May 19, 2011

Batsman against bowler groups: across ages

A statistical analysis of how batsmen have fared against various bowling attacks in Tests since 1900
381

Michael Vaughan: nearly 40% of his runs against top-level bowling attacks
Michael Vaughan: nearly 40% of his runs against top-level bowling attacks © Getty Images

(Revised on 22/05/11)

This article is a logical conclusion to the previous three articles. In these articles I looked at two teams which dominated the periods 1976-1995 and 1989-2008. I also looked at the batsmen who faced up these two outstanding sets of bowlers effectively. There was a discussion amongst the readers on the batsmen who faced up to strong bowling attacks, across years, effectively. It was also agreed that a composite single number indicating the weighted average bowling quality faced by a batsman across the career hides many truths.

Then Arjun Hemnani came out with a suggestion that I classify all bowling attacks into four groups and develop batting tables based on these groups. It seemed to be an excellent idea and I have created this article based on this idea. This is a quasi-rating work based on the most important of parameters, viz., the Bowling quality. I may do a similar exercise based on the Pitch conditions. Again some really tough work but at the end worthwhile.

I have summarized all relevant facts related to this analysis. First let me emphasize that this is not a Test innings Ratings analysis. There are many other relevant factors which would have to be considered in such an analysis. I have not done so in this analysis which is centred on Bowler quality. I would appreciate if the readers do not keep on repeating again and again that other relevant factors such as Pitch type, Innings status at entry, Result, Series status, Bowler recent form, Innings target et al, have not been included. That would be counter-productive.

1. As I have done in the Team strength calculations, I have considered only Tests played after 1900. It is impossible to fit in the Tests before 1900 because of uncovered pitches and many sub-20 averages. However we lose only 64 Tests.

2. The Bowling quality index (BQI) is based on Career-to-date values. This is the most dependable and accurate of the bowling measures. There is no situation where the Career-to-date figure is not the appropriate one. A bowler like Lee with a great start and tapering off towards the end or Muralitharan who had a poor start and wonderful finish will be taken care of equally well.

3. The BQI is based on the actual bowlers who bowled in the particular innings. This is very important. There is a Sri Lankan innings in which Wasim Akram and Waqar Younis bowled 14 overs each. That is all. This would have been a terrifying situation for the batsmen. Contrast this with an innings in which Akram and Younis bowled 50% of the overs.

4. The BQI is determined using the modified reciprocal method suggested by Arjun Hemnani which irons out the imbalance created by weak fifth bowlers. The career strike rate and career rpo are computed separately to arrive at the final BQI.

5. The BQI is determined for each innings. However in order to reduce wild variations, I will apply the BQI of the first innings to the second innings also in case the number of balls bowled in the second innings is less than 60. This is commonsense. This is explained through an example. Readers should know that this would not have much of an impact since no batsman is likely to score even 25 within the 10 overs.

Test # 1962: Win 231, Saf 346 (Win BQI 41.68), Win 161, Saf 49/3 in 8.4 overs (Win BQI 50.67).
In the above Test, the Saf second innings will be evaluated at 41.68 since fewer than 10 overs were bowled.

6. The BQI is reduced by 5% for Home games and increased by 5% for away games. Reader should remember that the lower the BQI, the more potent the attack is. There have been suggestions on increasing this quantum and on making this dependent on the specific country. I feel 5% either way is ample and the later requires some tricky work since I am not sure how to make it work. So that is for a later day. In general this concept is fine and works well in most cases.

It is possible that the visiting team has the right bowlers and can exploit the "away" bowling conditions. However there is no denying that, in most cases, the home bowlers would have the advantage of familiarity with and knowledge of local conditions.

The following italicised points are to be ignored in the current version.

7. The BQI is based on the Bowling average. However in order to recognize the importance of Strike rates in Test cricket is a special adjustment based on Strike rate. These concepts are explained in the examples.

8. The BQI is further modified by the Period related factor. The concerned table is given below. If the period average is lower than the all-time average of 31.76, it is a bowler-friendly age and the bowler/team averages are adjusted upwards. On the other hand, if the period average is higher than the all-time average of 31.76, it is a batsmen-friendly age and the bowler/team averages are adjusted downwards. I have adjusted the factor at a bowler level than spin/pace level since the later would have required a completely different way of working, at a player level. Even checking of results would have become very cumbersome and difficult. I also do not think that there is that much of a change.

Bowling average adjustment:  AMF - Average multiplying factor
Period       BowAvg     AMF
1877-1899     22.20    1.431
1900-1914     25.69    1.236
WW1-WW2       32.56    0.976
40s-50s       29.96    1.060
1960s         32.11    0.989
1970s         31.94    0.995
1980s         32.07    0.990
1990s         31.51    1.008
2000-2004     33.56    0.946
2005-2011     34.94    0.909
All Tests     31.76    1.000

Finally the bowling attacks are classified into 5 groups, as described below. The fifth group was necessary to separate the REALLY weak bowling attacks.

With the idea of short innings being tagged with the first innings, there have been 6837 qualifying innings until the West Indies - Pakistan Test which finished recently. The first cut-off has been fixed at 30 to have around 20% of the total innings into the top group. There may be a subjective element in this part of the exercise but that cannot be avoided. The basis on which we have decided that 30 will be the first cut-off point is not subjective. In fact Arjun's assertion that 20% means in a loose manner that at any time there are 2 really good bowling attacks makes eminent sense. The other cut-offs follow logically. The group cut-off details are given below.

Group  B Q I        # of Inns   % (out of 6837)

5 19.03-29.99: 1353 19.8% Very good bowling attack. (Prev: 19.2%) 4 30.00-34.99: 1703 24.9% Good bowling attack. (Prev: 23.9%) 3 35.00-39.99: 1753 25.6% Average bowling attack. (Prev: 21.3%) 2 40.00-44.99: 1095 16.0% Passable bowling attack. (Prev: 23.4%) 1 45.00+: 933 13.6% Very poor bowling attack. (Prev: 12.2%)

Note: In group 5 only three innings were below 20.00. Test 327: Australia (18.94). Lindwall (19.19), Johnston (18.24) and Miller (22.00) bowled 19 overs. Test 347: England (19.31). Lindwall (21.08), Johnston (19.25) and Miller (21.49) bowled 39 overs. Test 104: England (19.66). Barnes (20.14) and Blythe (19.20) bowled 47 of the 54 overs.

A note on the revised groupings. The S/R adjustments and the Period adjustments have been removed. The less-than-50-ctd-wickets situation has been ironed out. Established bowlers are assigned a better strike rate/rpos in line with their career averages. Only bowlers who did not even get 50 wickets in their careers are treated a bit roughly. All this has has resulted in some softening of the Bowling group allocations. The first two groups together have 2% of additional innings associated with them. However 4% more innings have come into the third group. This means that a total of over 70% of the innings now find place in the acceptable groups and below 30% in the two weaker groups. This contrasts with the 64-36% in the previous version. This has resulted in a better distribution of batsman performances.

In fact Arjun questioned the necessity to have the last group. He felt that the last two could be combined into one. However I strongly feel it is essential for the reason cited below. Let us look at the following two bowling attacks.

Test 1833, Bangladesh, Mortaza/Rafique/Shakib/Rasel/Sharif, BQI value is 41.45.
Test 226, Nzl (Hammond's 336), 4 "bowlers" had a ctd total of 26 wkts and a career total of 36 wkts, BQI value is 53.21.

There is no way I am going to put these into the same group. One is a good Test level attack and the other is barely at the level of a First Class side. So the last group is necessary.

The average BQI for this huge sample is 36.56 (37.13) and the median is at 35.93 (37.13). This indicates a fairly balanced distribution of values. The Standard Deviation is 7.30 (7.67). I have explained the whole concept of determining the BQI with the following examples.

First is MatchId 1267 between Sri Lanka and Pakistan, played at Kandy during 1994. In the first innings Wasim Akram (ctd 22.9) bowled 14.2 overs and Waqar Younis (ctd 19.3) bowled 14 overs and dismissed Sri Lanka for 71. The weighted BQI starts life at 20.81. This is multiplied by 1.05 (this being away Test for Pakistan). The final BQI value is 21.85 which puts this attack as a very potent one. Any runs scored in this particular innings, say A de Silva's 7 will get into the highest classification.

The second is MatchId 1844 between Pakistan and South Africa, played during 2007. These were Steyn's early years. As everyone knows he had a fairly average start to his career. Steyn bowled 12 overs, Ntini 8 overs, Kallis 7 overs, Harris 20 overs and Nel 16 overs. The base BQI is 30.77. The separation of strike rate and rpo in the reciprocal BQI calculation has benefited this attack because of Steyn's strike rate. This is multiplied by 1.05 (this being away Test for South Africa). The final BQI value is 32.31 which puts this attack as a fair one. Any runs scored in this particular innings, say Mohammed Yousuf's 25 will get into the second classification.

I have got into details here so as to give the readers a clear idea of the calculations.

There is so much data available that even the organization of the article is getting into trouble. I can only present in the article a certain amount of data. The serious reader should download the complete files and read the same. I have given below what I would be presenting within the article.

1. Top 20 batsmen for group 5, the top one. Ordered by batting average.
2. Top 20 batsmen for group 4, the second best one. Ordered by batting average.
3. Top 10 batsmen for groups 3-2-1. Ordered by batting average.
4. Top 10 batsmen for groups 5-1, the top one. Ordered by runs scored.
5. For selected batsman, their group-wise distribution of runs scored.
For all the above, complete files are available for downloading/viewing.

Let us look at the tables. First the Group tables based on Batting average. The batsman should have scored a minimum of 500 runs to be included. Otherwise we will have funny numbers.

Batsman              Cty Inns N Runs Grp  Avge

Hunte C.C Win 6 1 586 5 117.20 Bradman D.G Aus 15 0 1159 5 77.27 Sobers G.St.A Win 22 2 1328 5 66.40 Hutton L Eng 21 5 949 5 59.31 Wessels K.C Saf 14 1 763 5 58.69 Graveney T.W Eng 18 2 922 5 57.62 Hobbs J.B Eng 16 1 837 5 55.80 Lawry W.M Aus 14 3 613 5 55.73 Shoaib Mohammad Pak 11 1 539 5 53.90 Richards I.V.A Win 44 0 2338 5 53.14 Sangakkara K.C Slk 23 2 1045 5 49.76 Ponting R.T Aus 39 2 1794 5 48.49 Edrich J.H Eng 21 1 968 5 48.40 Martyn D.R Aus 24 2 1061 5 48.23 Lloyd C.H Win 39 2 1748 5 47.24 O'Neill N.C Aus 15 3 560 5 46.67 Cook A.N Eng 24 2 1017 5 46.23 Gomes H.A Win 21 3 830 5 46.11 Sehwag V Ind 33 2 1428 5 46.06 Amarnath M Ind 18 0 829 5 46.06

Hunte, by virtue of his huge unbeaten double century, has moved to the top with 117.20, but only 586 runs. Bradman's is the more significant performance since he has a 77+ average with over 1100 runs. Sobers, again buoyed by the 365*, has 1300+ runs at 66. Wessels is a surprise entrant as also Graveney. There is a complete churning of players with the result that Lara and Tendulkar have moved out of the top-20. The recent bowling attacks have lost their edge because of the removal of both s/r and period adjustments. Only Sangakkara, Ponting and Sehwag remain.
Bradman D.G          Aus  15  2 1275  4  98.08
Walcott C.L          Win  15  1 1067  4  76.21
Chappell G.S         Aus  49 10 2723  4  69.82
Sutcliffe H          Eng   9  0  617  4  68.56
McCabe S.J           Aus  16  1  986  4  65.73
Richards I.V.A       Win  54  9 2917  4  64.82
Wasim Raja           Pak  23  7 1019  4  63.69
Younis Khan          Pak  40  3 2352  4  63.57
Mohsin Khan          Pak  13  1  761  4  63.42
Hobbs J.B            Eng  15  2  758  4  58.31
Tendulkar S.R        Ind  85 10 4345  4  57.93
Crowe M.D            Nzl  34  5 1671  4  57.62
Smith G.C            Saf  53  4 2754  4  56.20
Walters K.D          Aus  46  6 2243  4  56.08
Border A.R           Aus  86 15 3935  4  55.42
Taylor R.L           Nzl  13  1  659  4  54.92
Richardson M.H       Nzl  18  0  977  4  54.28
Javed Miandad        Pak  44  4 2164  4  54.10
Kallicharran A.I     Win  26  4 1176  4  53.45
Gilchrist A.C        Aus  38  6 1701  4  53.16
Here we see 20 batsmen exceeding average of 50. Bradman has a near-career average of 98+ with a decent amount of runs. Walcott and Chappell come in next with 70+ and 70- averages. Sutcliffe and, nice to say McCabe, are next. Probably the most significant is Richards who has scored over 2900 runs at nearly 65. Quite a few modern batsmen have plenty of runs at 55+ averages. Mark Richardson is a surprise at the top, way above his career average of 44.77. He has done well against Pakistan and India.
Bradman D.G          Aus  17  3 1479  3 105.64
Sutcliffe H          Eng  22  5 1498  3  88.12
Zaheer Abbas         Pak  19  2 1495  3  87.94
EdeC Weekes          Win  15  1 1071  3  76.50
Amiss D.L            Eng  25  2 1744  3  75.83
Flower A             Zim  33  7 1940  3  74.62
Lara B.C             Win  52  2 3674  3  73.48
Compton D.C.S        Eng  14  3  808  3  73.45
Barrington K.F       Eng  34  4 2131  3  71.03
Walcott C.L          Win  14  1  912  3  70.15

Kallis J.H Saf 30 8 2333 2 106.05 Mohammad Yousuf Pak 16 1 1506 2 100.40 Headley G.A Win 12 3 890 2 98.89 Pietersen K.P Eng 11 2 851 2 94.56 Worrell F.M.M Win 14 2 1133 2 94.42 Woodfull W.M Aus 6 0 545 2 90.83 Prince A.G Saf 10 4 515 2 85.83 Shastri R.J Ind 15 4 932 2 84.73 Kirsten G Saf 16 3 1067 2 82.08 Vengsarkar D.B Ind 23 6 1389 2 81.71

Adams J.C Win 8 4 658 1 164.50 Bell I.R Eng 6 1 531 1 106.20 Mohammad Yousuf Pak 11 4 703 1 100.43 Dravid R Ind 14 4 987 1 98.70 Sehwag V Ind 9 2 672 1 96.00 Waugh S.R Aus 20 6 1268 1 90.57 Bradman D.G Aus 17 2 1342 1 89.47 de Villiers A.B Saf 12 4 713 1 89.12 Kallis J.H Saf 21 7 1228 1 87.71 Prince A.G Saf 12 5 614 1 87.71

Quite a few 100+ averages at the lower bowling quality levels. Kallis has a 100+ average against group 4, as so does Yousuf. Do not forget that group 3 is still a fair bowling attack.
Batsman              Cty Inns N Runs Grp  Avge

Stewart A.J Eng 97 8 3311 5 37.20 Lara B.C Win 76 1 3270 5 43.60 Atherton M.A Eng 98 1 2581 5 26.61 Gooch G.A Eng 73 1 2573 5 35.74 Tendulkar S.R Ind 63 2 2396 5 39.28 Richards I.V.A Win 44 0 2338 5 53.14 Dravid R Ind 65 3 2288 5 36.90 Border A.R Aus 70 10 2191 5 36.52 Chanderpaul S Win 67 9 2135 5 36.81 Hussain N Eng 66 6 2048 5 34.13

Tendulkar S.R Ind 85 10 4345 4 57.93 Border A.R Aus 86 15 3935 4 55.42 Dravid R Ind 75 10 3410 4 52.46 Kallis J.H Saf 85 10 3302 4 44.03 Waugh S.R Aus 76 10 3291 4 49.86 Lara B.C Win 70 1 3286 4 47.62 Richards I.V.A Win 54 9 2917 4 64.82 Atherton M.A Eng 66 2 2875 4 44.92 Inzamam-ul-Haq Pak 59 5 2791 4 51.69 Smith G.C Saf 53 4 2754 4 56.20

Ponting R.T Aus 97 13 5146 3 61.26 Kallis J.H Saf 79 9 3941 3 56.30 Lara B.C Win 52 2 3674 3 73.48 Jayawardene D.P.M.D Slk 55 1 3579 3 66.28 Tendulkar S.R Ind 69 7 3505 3 56.53 Hayden M.L Aus 67 5 3474 3 56.03 Dravid R Ind 64 6 3307 3 57.02 Gower D.I Eng 61 4 3063 3 53.74 Border A.R Aus 62 11 2993 3 58.69 Gavaskar S.M Ind 60 5 2940 3 53.45

Tendulkar S.R Ind 49 7 2982 2 71.00 Kallis J.H Saf 30 8 2333 2 106.05 Javed Miandad Pak 39 4 2242 2 64.06 Ponting R.T Aus 46 9 2181 2 58.95 Dravid R Ind 41 6 2071 2 59.17 Edrich J.H Eng 43 5 1894 2 49.84 Hammond W.R Eng 35 5 1890 2 63.00 Barrington K.F Eng 32 6 1860 2 71.54 Sangakkara K.C Slk 31 2 1820 2 62.76 Jayawardene D.P.M.D Slk 37 5 1813 2 56.66

Hammond W.R Eng 66 8 4133 1 71.26 Hutton L Eng 68 7 3599 1 59.00 Compton D.C.S Eng 47 5 2456 1 58.48 Harvey R.N Aus 35 2 2180 1 66.06 Sutcliffe H Eng 35 4 1897 1 61.19 Gavaskar S.M Ind 30 2 1882 1 67.21 Cowdrey M.C Eng 39 4 1761 1 50.31 Graveney T.W Eng 32 4 1704 1 60.86 Javed Miandad Pak 30 6 1660 1 69.17 Barrington K.F Eng 29 3 1624 1 62.46

The above tables are ordered on runs scored against the bowling groups. Stewart and Tendulkar lead the top two groups with 3311 and 4345 runs. Lara also has over 3000 runs against each of the top groups. Almost the same with Tendulkar indicating that, on balance, they faced better quality of bowling. The modern batsmen, led by Tendulkar top the middle group table. Tendulkar and quite a few current batsmen top the weaker group 2. Finally look at Hammond. Nearly 60% of his runs have come against the weakest of attacks, albeit, at no great an average. It must be accepted that when batsmen play 100+ tests, they are likely to figure in both groups 5 and 1 prominently. Hammond, Hutton and Compton are the surprises since they had career runs of 7000 and just above or below.

Now for the group-wise runs and % of career runs for selected 20 odd batsmen. The complete file is available for downloading.

Batsman              Cty Inns N  Runs    %  Grp  Avge

Hobbs J.B Eng 16 1 837 15.5 5 55.80 Hobbs J.B Eng 15 2 758 14.0 4 58.31 Hobbs J.B Eng 29 2 1642 30.4 3 60.81 Hobbs J.B Eng 20 1 854 15.8 2 44.95 Hobbs J.B Eng 22 1 1319 24.4 1 62.81

Sutcliffe H Eng 6 0 110 2.4 5 18.33 Sutcliffe H Eng 9 0 617 13.5 4 68.56 Sutcliffe H Eng 22 5 1498 32.9 3 88.12 Sutcliffe H Eng 12 0 433 9.5 2 36.08 Sutcliffe H Eng 35 4 1897 41.6 1 61.19

Hammond W.R Eng 4 0 124 1.7 5 31.00 Hammond W.R Eng 11 1 397 5.5 4 39.70 Hammond W.R Eng 24 2 705 9.7 3 32.05 Hammond W.R Eng 35 5 1890 26.1 2 63.00 Hammond W.R Eng 66 8 4133 57.0 1 71.26

Bradman D.G Aus 15 0 1159 16.6 5 77.27 Bradman D.G Aus 15 2 1275 18.2 4 98.08 Bradman D.G Aus 17 3 1479 21.1 3 105.64 Bradman D.G Aus 16 3 1741 24.9 2 133.92 Bradman D.G Aus 17 2 1342 19.2 1 89.47

Hutton L Eng 21 5 949 13.6 5 59.31 Hutton L Eng 21 0 884 12.7 4 42.10 Hutton L Eng 12 1 530 7.6 3 48.18 Hutton L Eng 16 2 1009 14.5 2 72.07 Hutton L Eng 68 7 3599 51.6 1 59.00

Sobers G.St.A Win 22 2 1328 16.5 5 66.40 Sobers G.St.A Win 47 4 2169 27.0 4 50.44 Sobers G.St.A Win 43 6 2060 25.6 3 55.68 Sobers G.St.A Win 26 4 1626 20.2 2 73.91 Sobers G.St.A Win 22 5 849 10.6 1 49.94

Lloyd C.H Win 39 2 1748 23.3 5 47.24 Lloyd C.H Win 40 3 1688 22.5 4 45.62 Lloyd C.H Win 54 5 2068 27.5 3 42.20 Lloyd C.H Win 35 4 1712 22.8 2 55.23 Lloyd C.H Win 7 0 299 4.0 1 42.71

Chappell G.S Aus 42 2 1490 21.0 5 37.25 Chappell G.S Aus 49 10 2723 38.3 4 69.82 Chappell G.S Aus 30 3 1465 20.6 3 54.26 Chappell G.S Aus 20 2 918 12.9 2 51.00 Chappell G.S Aus 10 2 514 7.2 1 64.25

Gavaskar S.M Ind 47 2 1604 15.8 5 35.64 Gavaskar S.M Ind 42 1 2122 21.0 4 51.76 Gavaskar S.M Ind 60 5 2940 29.0 3 53.45 Gavaskar S.M Ind 35 6 1574 15.6 2 54.28 Gavaskar S.M Ind 30 2 1882 18.6 1 67.21

Richards I.V.A Win 44 0 2338 27.4 5 53.14 Richards I.V.A Win 54 9 2917 34.2 4 64.82 Richards I.V.A Win 48 1 1795 21.0 3 38.19 Richards I.V.A Win 32 2 1439 16.9 2 47.97 Richards I.V.A Win 4 0 51 0.6 1 12.75

Gooch G.A Eng 73 1 2573 28.9 5 35.74 Gooch G.A Eng 53 2 2409 27.1 4 47.24 Gooch G.A Eng 52 1 2010 22.6 3 39.41 Gooch G.A Eng 26 2 1269 14.3 2 52.88 Gooch G.A Eng 11 0 639 7.2 1 58.09

Javed Miandad Pak 39 0 1434 16.2 5 36.77 Javed Miandad Pak 44 4 2164 24.5 4 54.10 Javed Miandad Pak 37 7 1332 15.1 3 44.40 Javed Miandad Pak 39 4 2242 25.4 2 64.06 Javed Miandad Pak 30 6 1660 18.8 1 69.17

Border A.R Aus 70 10 2191 19.6 5 36.52 Border A.R Aus 86 15 3935 35.2 4 55.42 Border A.R Aus 62 11 2993 26.8 3 58.69 Border A.R Aus 30 6 1038 9.3 2 43.25 Border A.R Aus 17 2 1017 9.1 1 67.80

Waugh S.R Aus 43 5 1643 15.0 5 43.24 Waugh S.R Aus 76 10 3291 30.1 4 49.86 Waugh S.R Aus 82 17 2920 26.7 3 44.92 Waugh S.R Aus 39 8 1805 16.5 2 58.23 Waugh S.R Aus 20 6 1268 11.6 1 90.57

Atherton M.A Eng 98 1 2581 33.4 5 26.61 Atherton M.A Eng 66 2 2875 37.2 4 44.92 Atherton M.A Eng 34 2 1534 19.8 3 47.94 Atherton M.A Eng 9 1 371 4.8 2 46.38 Atherton M.A Eng 5 1 367 4.7 1 91.75

Tendulkar S.R Ind 63 2 2396 16.3 5 39.28 Tendulkar S.R Ind 85 10 4345 29.6 4 57.93 Tendulkar S.R Ind 69 7 3505 23.9 3 56.53 Tendulkar S.R Ind 49 7 2982 20.3 2 71.00 Tendulkar S.R Ind 24 6 1464 10.0 1 81.33

Stewart A.J Eng 97 8 3311 39.1 5 37.20 Stewart A.J Eng 69 6 2240 26.5 4 35.56 Stewart A.J Eng 35 3 1628 19.2 3 50.88 Stewart A.J Eng 25 3 845 10.0 2 38.41 Stewart A.J Eng 9 1 441 5.2 1 55.12

Lara B.C Win 76 1 3270 27.4 5 43.60 Lara B.C Win 70 1 3286 27.5 4 47.62 Lara B.C Win 52 2 3674 30.7 3 73.48 Lara B.C Win 22 2 979 8.2 2 48.95 Lara B.C Win 12 0 744 6.2 1 62.00

Inzamam-ul-Haq Pak 40 1 1347 15.3 5 34.54 Inzamam-ul-Haq Pak 59 5 2791 31.6 4 51.69 Inzamam-ul-Haq Pak 52 7 2409 27.3 3 53.53 Inzamam-ul-Haq Pak 30 5 1474 16.7 2 58.96 Inzamam-ul-Haq Pak 19 4 809 9.2 1 53.93

Fleming S.P Nzl 50 2 1459 20.3 5 30.40 Fleming S.P Nzl 47 5 2093 29.2 4 49.83 Fleming S.P Nzl 55 2 1906 26.6 3 35.96 Fleming S.P Nzl 25 1 1147 16.0 2 47.79 Fleming S.P Nzl 12 0 567 7.9 1 47.25

Ponting R.T Aus 39 2 1794 14.5 5 48.49 Ponting R.T Aus 65 3 2751 22.3 4 44.37 Ponting R.T Aus 97 13 5146 41.6 3 61.26 Ponting R.T Aus 46 9 2181 17.6 2 58.95 Ponting R.T Aus 12 1 487 3.9 1 44.27

Kallis J.H Saf 31 4 1143 9.6 5 42.33 Kallis J.H Saf 85 10 3302 27.6 4 44.03 Kallis J.H Saf 79 9 3941 33.0 3 56.30 Kallis J.H Saf 30 8 2333 19.5 2 106.05 Kallis J.H Saf 21 7 1228 10.3 1 87.71

Dravid R Ind 65 3 2288 19.0 5 36.90 Dravid R Ind 75 10 3410 28.3 4 52.46 Dravid R Ind 64 6 3307 27.4 3 57.02 Dravid R Ind 41 6 2071 17.2 2 59.17 Dravid R Ind 14 4 987 8.2 1 98.70

Laxman V.V.S Ind 52 4 1876 23.7 5 39.08 Laxman V.V.S Ind 61 7 2480 31.4 4 45.93 Laxman V.V.S Ind 46 12 1709 21.6 3 50.26 Laxman V.V.S Ind 29 8 1591 20.1 2 75.76 Laxman V.V.S Ind 10 0 247 3.1 1 24.70

Jayawardene D.P.M.D Slk 29 1 1124 11.8 5 40.14 Jayawardene D.P.M.D Slk 44 4 1639 17.2 4 40.97 Jayawardene D.P.M.D Slk 55 1 3579 37.6 3 66.28 Jayawardene D.P.M.D Slk 37 5 1813 19.0 2 56.66 Jayawardene D.P.M.D Slk 25 2 1375 14.4 1 59.78

Sehwag V Ind 33 2 1428 18.6 5 46.06 Sehwag V Ind 46 0 2239 29.1 4 48.67 Sehwag V Ind 33 0 1760 22.9 3 53.33 Sehwag V Ind 29 2 1595 20.7 2 59.07 Sehwag V Ind 9 2 672 8.7 1 96.00

Sangakkara K.C Slk 23 2 1045 12.7 5 49.76 Sangakkara K.C Slk 36 4 1655 20.1 4 51.72 Sangakkara K.C Slk 43 0 2178 26.4 3 50.65 Sangakkara K.C Slk 31 2 1820 22.1 2 62.76 Sangakkara K.C Slk 23 4 1546 18.8 1 81.37

Hobbs, Sutcliffe and Hammond have made very few runs against top group, mainly indicating that there were very few top group bowling attacks during these 50 years. Lloyd, Chappell, Laxman and Fleming all have around 20-25% against the top bowling group. However the real numbers are for Stewart with 39%, Atherton with 33.4%, Richards with 27.4%, Gooch with 28.9% and Lara with 27.4%.

Looking at the other end of the tables, Hammond has scored 57% of career runs against the weakest of attacks. Hutton stands at 51% and Sutcliffe at 41%. Ponting, Laxman, Lloyd and Stewart have only 3-5% against these weak attacks. But the amazing number is for Richards, who in his cricketing life has almost faced no group 5 bowling attack (0.6%). That says something about the sign of the period 1970s-90s.

Group table - by Batting average: please click/right-click here.
Group table - by Runs scored: please click/right-click here.
Batsman table - by Group (for all 2000+ batsmen): please click/right-click here.
Batsman table - all Group 5 performances (for all 5000+ batsmen): please click/right-click here.
BQI table - ordered by BQI (for all 6827 innings): please click/right-click here.

Er-Sr calculations: please click/right-click here.
Weighted bowling quality table - ordered by WtBowQty value (gt 4000 runs): please click/right-click here.

I would like to inform the readers that we are away on holiday between June 2 and 25 and I will be taking the month of June off. I will be back in July. I am confident that most of the comments on this article, and I expect plenty, to be in before June 1.

Anantha Narayanan has written for ESPNcricinfo and CastrolCricket and worked with a number of companies on their cricket performance ratings-related systems

Comments have now been closed for this article

  • AD on July 2, 2011, 4:21 GMT

    Gerry_the_Merry: As regards perid adjustments we may agree to disagree. In my view the only correct method of judging a sportsman is relative to his peers. As far as batting goes this factors in not just one or two variables like pitch conditions or bowling quality ,but various others. For example, the trend of judging LBWs or umpiring standards in a particular period vs. another period. There are numerous variables which cannot be normalized across eras.

    As regards Ponting, he certainly has an impressive body of accomplishment and would classify as an All Time Great. However, the feeling that he is just that rung below some other Legends lingers.

  • Gerry_the_Merry on July 1, 2011, 11:21 GMT

    Ananth, Great that you are back. Life becomes normal again. The last item is CTD averages. Somewhere you have mentioned that you are adjusting for early career distortions. How are you doing this please? How about an Brett Lee / Ishant Sharma type of bowler - good start, then mediocre averages for some time, before improvement? Obviously you have to keep out unnecessary complications, but am just curious as to your method adopted. [[ Gerry Will do so in a couple of days after clearing up the 97 (!) things on the plate. Ananth: ]]

  • Alex on June 30, 2011, 3:45 GMT

    @Boll: Absolutely on Ponting. In fact, my all-time world XI reads:

    Tests: 9 fixed players ... Hobbs, Viv (as opener ... he was the best ever vs fast bowlers anyway), Don, SRT, Lara, Sobers, Gilly/Sanga, Imran(c), Marshall.

    - PLUS any two of Warne/Murali/Barnes/O'Reilly on spinning wicket. - PLUS McGrath and Ambrose/Lillee on fast tracks. - PLUS Ponting as the 12th man in all cases.

    ODI's: SRT, Lara, Viv, Ponting, Dhoni (c), Hussey, Bevan, Imran, Murali, McGrath, Garner.

    - PLUS Jonty as the 12th.

  • Gerry_the_Merry on June 17, 2011, 7:04 GMT

    In support of my previous point about developing separate team 1st and team 2nd inn bowling average BQI, here are some approx stats for % by which the 1st inn average is higher than the bowler's second inn average - McGrath 0% (21,21 in 1st and 2nd Inn); Ambrose 10% (22 in 2nd inn, 20 in 1st inn); Lillee ~10%; Holding ~11%; Murali 20%; Warne ~28%; Caddick 80% (Caddick's 2nd inn 20, 1st inn 37). So there...!

    Ananth, i seriously think you should do a piece on this prior to releasing your grand finale, so that readers are not shocked. it would give a better understanding of how bowlers are actually generating their bowling avg. [[ Just got back home and finally back to my own laptop and computer. On road was not able to chack mail regularly and made the mistake of not taking either the laptop (on holiday !!!, no way) and my Motorola (should have taken this, I was worried about nonsense calls following me right throughout Europe). Will take a few days to go through and plan what is needed to be done. Ananth: ]]

  • Boll on June 16, 2011, 14:43 GMT

    @AD re.Ponting cont`d. His prominent place in Ananth`s batting table vs top-class bowling attacks (well ahead of Tendulkar, Lara and contrary to your implication that he performed only against weaker attacks) confirms his place as one of the great batsmen.

    And of course we should remember that he remains the only player to have won 3 World Cups, 2 as captain (a record he shares with Lloyd). He also captained Aus to 2 Champions Trophies, and has won more tests as captain or player than anyone else in history.

    He was overwhelmingly voted the cricketer of the decade 2000-2009 by ESPN - Ponting scored more runs and centuries in both forms of the game than any other batsman during that time. He has also been the best all-round fielder of his generation.

    Just that rung below? Below whom exactly?

  • Gerry_the_Merry on June 16, 2011, 10:02 GMT

    Pallab, I dont disagree with Boll that the English attack is the best today. You seem to have a pretty disparaging view about Anderson and Swann, who are pretty dangerous bowlers, with success (at least in recent years) in English as well as Australian conditions. Tremlett, Finn and Broad are a handful because of their height, and Onions, Shahzad et all are handy backups.

    There are two immediate comparisons - 1) how the same Australian batting line up did against India - pretty OK, if you think, compared to their performance at home against the Englishmen and 2) how they will do against SAF in a few months - I bet they do better than lose thrice by an innings. Regarding Tostobe / Parnell / Morkel / Harriss, i dont remember any of them threating any attack on their own. Basically it is Steyn plus a few average bowlers (Morkel has a dismal average).

  • Gerry_the_Merry on June 16, 2011, 3:39 GMT

    ...Contd...This is also why I have been insisting that separate BQI for home/away and for 1/2 Innings and 3/4 innings are generated using home / away bowling averages and 1/2 and 3/4 innings averages. This is because only then will the CONDITIONS IN WHICH THE BQI IS GENERATED MATCH WITH THE CONDITIONS IN WHICH THE BATTING AVERAGE HAS BEEN GENERATED.

    You will find that many bowlers have a lower 3/4 innings average than 1/2. Similarly superior home average to away. Ananth has all this already worked out in different previous articles. My suggestion is that this is the grand opportunity to put it all together, because never before has anyone so beautifully placed batting average in the context of tightly defined bowling average groups.

  • Gerry_the_Merry on June 16, 2011, 3:30 GMT

    AD, on period adjustments, I completely, totally and utterly disagree. There must be no period adjustment. Here is the logic. Imagine that for a moment, we are not facing sample size limitations, and we are defining a tight band of BQI 20-to-25. In this band, REGARDLESS of whether the pitches were sunny, rainy, muddy, sticky; regardless of whether the bowling was fast or spin; regardless of helmets / no helmets; regardless of whether the period is 1950-75 or 1985 to 2010, a batting average of (say) 50 is equally merited. All the above factors would have got built in to the bowling average.

    The implicit underlying assumption is that the BQI has been generated in roughly SIMILAR conditions as the batting average, making them a fair face-off.

    If from two different periods, the bowling was terrifying / mediocre, then the difference will be seen in the proportion of innings / runs between BQI groups (more Group 5 innings in the high bowling quality period and vice versa).

  • Alex on June 15, 2011, 13:29 GMT

    @AD: Actually, Ponting's domination spanned a full decade ... which is the norm for just about every post-WW1 batsman who could be said to a stake a claim to an All-Time Top 20 status. Very few have had two dominant years separated by 15+ years.

    1. Ponting averaged 50 in year 1997 and averaged 48 in year 2008.

    2. In tests, his best years were 1999-2008 (a full decade). It's just that his Bradmanesque streak of 61 tests (ave=74, 24 centuries) spanned only 5 years (2002-06) unlike Don's which spanned 20 years (actually 13, after taking WW2 into account).

    3. Ponting has continued to play quite well in ODI's since '08.

    After the dust settles down, I think history will view Ponting as one of the 3 best batsmen of the 1990-2010 era, possibly just behind Lara-SRT. And only Greg Chappell rivals him for the title of the best Oz batsman since Don. In addition, Ponting might very well be the greatest fielder of all-time.

  • west indian follower on June 15, 2011, 11:33 GMT

    Ananth, I am not sure whether this is a fair comment, but, I have always wondered why India would play someone like Yujrav Singh over Badrinath, even though Badrinath's average is about 20 runs higher. However, after the fourth ODI against the Windies, it became clear. Seeing Badrinath visibly unsettled by Andre Russell bowling 82 mile an hour bouncers really drove the point home. Ironically, Ashwin, who averages 35 looked very settled for the runs that he scored. This must bring in a new dimension to the selection process in India; someone like Rohit Sharma, even though having the same first class average as Badrinath, looks a lot better player. Maybe this is a case when the selectors cannot go on stats alone. Also, why has Rahane not been selected, I mean averaging 67 and not getting the call up must be pretty frustrating... [[ Badrinath has always had a raw deal. this despite his coming from the same state as Srikkanth. There must be some personal likes/dislikes. Badrinath is a true world class Test player. However Yuvraj, and now Kohli get in. Ananth: ]]

  • AD on July 2, 2011, 4:21 GMT

    Gerry_the_Merry: As regards perid adjustments we may agree to disagree. In my view the only correct method of judging a sportsman is relative to his peers. As far as batting goes this factors in not just one or two variables like pitch conditions or bowling quality ,but various others. For example, the trend of judging LBWs or umpiring standards in a particular period vs. another period. There are numerous variables which cannot be normalized across eras.

    As regards Ponting, he certainly has an impressive body of accomplishment and would classify as an All Time Great. However, the feeling that he is just that rung below some other Legends lingers.

  • Gerry_the_Merry on July 1, 2011, 11:21 GMT

    Ananth, Great that you are back. Life becomes normal again. The last item is CTD averages. Somewhere you have mentioned that you are adjusting for early career distortions. How are you doing this please? How about an Brett Lee / Ishant Sharma type of bowler - good start, then mediocre averages for some time, before improvement? Obviously you have to keep out unnecessary complications, but am just curious as to your method adopted. [[ Gerry Will do so in a couple of days after clearing up the 97 (!) things on the plate. Ananth: ]]

  • Alex on June 30, 2011, 3:45 GMT

    @Boll: Absolutely on Ponting. In fact, my all-time world XI reads:

    Tests: 9 fixed players ... Hobbs, Viv (as opener ... he was the best ever vs fast bowlers anyway), Don, SRT, Lara, Sobers, Gilly/Sanga, Imran(c), Marshall.

    - PLUS any two of Warne/Murali/Barnes/O'Reilly on spinning wicket. - PLUS McGrath and Ambrose/Lillee on fast tracks. - PLUS Ponting as the 12th man in all cases.

    ODI's: SRT, Lara, Viv, Ponting, Dhoni (c), Hussey, Bevan, Imran, Murali, McGrath, Garner.

    - PLUS Jonty as the 12th.

  • Gerry_the_Merry on June 17, 2011, 7:04 GMT

    In support of my previous point about developing separate team 1st and team 2nd inn bowling average BQI, here are some approx stats for % by which the 1st inn average is higher than the bowler's second inn average - McGrath 0% (21,21 in 1st and 2nd Inn); Ambrose 10% (22 in 2nd inn, 20 in 1st inn); Lillee ~10%; Holding ~11%; Murali 20%; Warne ~28%; Caddick 80% (Caddick's 2nd inn 20, 1st inn 37). So there...!

    Ananth, i seriously think you should do a piece on this prior to releasing your grand finale, so that readers are not shocked. it would give a better understanding of how bowlers are actually generating their bowling avg. [[ Just got back home and finally back to my own laptop and computer. On road was not able to chack mail regularly and made the mistake of not taking either the laptop (on holiday !!!, no way) and my Motorola (should have taken this, I was worried about nonsense calls following me right throughout Europe). Will take a few days to go through and plan what is needed to be done. Ananth: ]]

  • Boll on June 16, 2011, 14:43 GMT

    @AD re.Ponting cont`d. His prominent place in Ananth`s batting table vs top-class bowling attacks (well ahead of Tendulkar, Lara and contrary to your implication that he performed only against weaker attacks) confirms his place as one of the great batsmen.

    And of course we should remember that he remains the only player to have won 3 World Cups, 2 as captain (a record he shares with Lloyd). He also captained Aus to 2 Champions Trophies, and has won more tests as captain or player than anyone else in history.

    He was overwhelmingly voted the cricketer of the decade 2000-2009 by ESPN - Ponting scored more runs and centuries in both forms of the game than any other batsman during that time. He has also been the best all-round fielder of his generation.

    Just that rung below? Below whom exactly?

  • Gerry_the_Merry on June 16, 2011, 10:02 GMT

    Pallab, I dont disagree with Boll that the English attack is the best today. You seem to have a pretty disparaging view about Anderson and Swann, who are pretty dangerous bowlers, with success (at least in recent years) in English as well as Australian conditions. Tremlett, Finn and Broad are a handful because of their height, and Onions, Shahzad et all are handy backups.

    There are two immediate comparisons - 1) how the same Australian batting line up did against India - pretty OK, if you think, compared to their performance at home against the Englishmen and 2) how they will do against SAF in a few months - I bet they do better than lose thrice by an innings. Regarding Tostobe / Parnell / Morkel / Harriss, i dont remember any of them threating any attack on their own. Basically it is Steyn plus a few average bowlers (Morkel has a dismal average).

  • Gerry_the_Merry on June 16, 2011, 3:39 GMT

    ...Contd...This is also why I have been insisting that separate BQI for home/away and for 1/2 Innings and 3/4 innings are generated using home / away bowling averages and 1/2 and 3/4 innings averages. This is because only then will the CONDITIONS IN WHICH THE BQI IS GENERATED MATCH WITH THE CONDITIONS IN WHICH THE BATTING AVERAGE HAS BEEN GENERATED.

    You will find that many bowlers have a lower 3/4 innings average than 1/2. Similarly superior home average to away. Ananth has all this already worked out in different previous articles. My suggestion is that this is the grand opportunity to put it all together, because never before has anyone so beautifully placed batting average in the context of tightly defined bowling average groups.

  • Gerry_the_Merry on June 16, 2011, 3:30 GMT

    AD, on period adjustments, I completely, totally and utterly disagree. There must be no period adjustment. Here is the logic. Imagine that for a moment, we are not facing sample size limitations, and we are defining a tight band of BQI 20-to-25. In this band, REGARDLESS of whether the pitches were sunny, rainy, muddy, sticky; regardless of whether the bowling was fast or spin; regardless of helmets / no helmets; regardless of whether the period is 1950-75 or 1985 to 2010, a batting average of (say) 50 is equally merited. All the above factors would have got built in to the bowling average.

    The implicit underlying assumption is that the BQI has been generated in roughly SIMILAR conditions as the batting average, making them a fair face-off.

    If from two different periods, the bowling was terrifying / mediocre, then the difference will be seen in the proportion of innings / runs between BQI groups (more Group 5 innings in the high bowling quality period and vice versa).

  • Alex on June 15, 2011, 13:29 GMT

    @AD: Actually, Ponting's domination spanned a full decade ... which is the norm for just about every post-WW1 batsman who could be said to a stake a claim to an All-Time Top 20 status. Very few have had two dominant years separated by 15+ years.

    1. Ponting averaged 50 in year 1997 and averaged 48 in year 2008.

    2. In tests, his best years were 1999-2008 (a full decade). It's just that his Bradmanesque streak of 61 tests (ave=74, 24 centuries) spanned only 5 years (2002-06) unlike Don's which spanned 20 years (actually 13, after taking WW2 into account).

    3. Ponting has continued to play quite well in ODI's since '08.

    After the dust settles down, I think history will view Ponting as one of the 3 best batsmen of the 1990-2010 era, possibly just behind Lara-SRT. And only Greg Chappell rivals him for the title of the best Oz batsman since Don. In addition, Ponting might very well be the greatest fielder of all-time.

  • west indian follower on June 15, 2011, 11:33 GMT

    Ananth, I am not sure whether this is a fair comment, but, I have always wondered why India would play someone like Yujrav Singh over Badrinath, even though Badrinath's average is about 20 runs higher. However, after the fourth ODI against the Windies, it became clear. Seeing Badrinath visibly unsettled by Andre Russell bowling 82 mile an hour bouncers really drove the point home. Ironically, Ashwin, who averages 35 looked very settled for the runs that he scored. This must bring in a new dimension to the selection process in India; someone like Rohit Sharma, even though having the same first class average as Badrinath, looks a lot better player. Maybe this is a case when the selectors cannot go on stats alone. Also, why has Rahane not been selected, I mean averaging 67 and not getting the call up must be pretty frustrating... [[ Badrinath has always had a raw deal. this despite his coming from the same state as Srikkanth. There must be some personal likes/dislikes. Badrinath is a true world class Test player. However Yuvraj, and now Kohli get in. Ananth: ]]

  • Boll on June 15, 2011, 7:00 GMT

    @Pallab I think you'll find that there a quite a few people who rate the current English test attack as No.1. Sure, some others would back South Africa but I happen to think Swann tips the scales England's way.This is not just based on their recent victory over Australia, they've beaten plenty of other teams over the past couple of years as well.

    I certainly didn't claim they were a great attack (yet), just that they deserved a little more credit than you give them - I'm not quite sure how this then became an argument about the best pace/seam attacks to have played in Oz over the past 30 years??

  • shrikanthk on June 15, 2011, 2:40 GMT

    Ponting and Viv have many similarities (attacking, superb under pressure, best fielders of generation, etc.). Viv was shown the door and let's hope Ponting goes out on a high

    Alex: There is one crucial difference. Viv post 1986, despite under-performing with the bat, was a part of a remarkably successful WI side which hardly lost a series during that period. That's the reason why Viv could carry on that long despite not being near his best.

    The problem for Ponting is that his team has not been winning. As a result, the criticism of his performances is a lot more severe due to higher expectations. Had this Australian side been winning, Ponting's average of 38 wouldn't have been a problem (just as Strauss' under-performance is not a serious issue in England right now).

  • AD on June 14, 2011, 6:04 GMT

    Re.Ponting , it is here that we differentiate the Greatest of players from those just that rung below. Ponting and a few others dominated batting for around 5 yrs in the mid 2000s. But have been well below par on either side of that period.

  • AD on June 14, 2011, 5:49 GMT

    Gerry_the_Merry - I would say Period adjustments are the only way to normalise batting conditions across eras. Even a Pitch factor will nor cover other numerous variables which affect batting figures . The Period adjustments may be looked upon as a sort of Peer adjustment/ratio. This is the only correct method to compare players.

    For eg. if we tacitly accept that 55 is the new 50 then we should not ignore it statistically.

  • Pallab on June 13, 2011, 7:09 GMT

    @BOLL:As an exercise,could you jog your memory and list your top 10 foreign pace/seam attacks to have played and faced generally reliable, strong OZ batting units (barring the mid-80s dip) in OZ since 1982? Let’s keep spin out as barring Kumble in 2003 (24 wickets), none of the rated spinners have really done well on OZ tours. Give some perspective about the 80s attacks as telecast of non-Indian team tours of OZ began only in the 90s. Even now I will stake my analyst reputation and say that Flintoff, Jones, Hoggard(if fit, able and in form) and Harrmison (stronger mindset)would have been the first choice attack (and not the current trio/quartet) for Eng in 2009, if not for the caprices of selection and their early culling. Incidentally, my strong OZ batting line-up would have been Katich,Watson,Ponting, Hussey,Hodge (yes, bloody-minded enuf and desperate for a Test legacy) and Symonds(along with Katich, even I am fed up of the ruthless,untimely culling of Roy,Clark, Dizzy, Bracken)

  • Pallab on June 13, 2011, 7:02 GMT

    “I think Eng have the best test attack.” @BOLL:Are you seriously serious?Just cos they beat OZ?Till as recently as Aug 2010, Pak with Gul, Amir, Asif (forget the issues for a while) had the best fast/seam Test attack. Also SAffers would have grievances that you have overlooked Steyn, Morkel, Kallis and Tsotosbe/Parnell's claims as a better Test attack. Boll,I remember you mentioning elders waking you up to applaud Botham’s heroics in 1981. So if from 1982 you seriously started following/watching Test cricket in OZ summers till 2010, would the 2010 Eng pace attack even feature in your top 10? Hell, even Hadlee, Lance Cairns,Chatfield/Snedden in 1985 had better potency/class.So that makes 4 WI pace attack-led teams (till 1997), Imran, Javed, Akram’s teams, 2 SA team of 90s, Gatting’s team in 1987 making the top 10 pace-led attacks to take on Aus in OZ. That’s why I mentioned the lack of depth/quality of current OZ batting unit facing this Eng attack barring the high-quality spin of Swann

  • Alex on June 13, 2011, 1:23 GMT

    @shrikanthk: Ponting at MCG vs SA '09 was one of his best. On enjoying Ponting, how many tests is he left with now? He can still score a 50 quite regularly but, just like Viv after '86, his problem is in converting those into big scores ... that makes his average look poor.

    Ponting since '08: 25 tests, 46 inn, 2 100's, 13 50's, ave=38. Viv after '86 (final 4 yrs): 36 tests, 55 inn, 4 100's, 17 50's, ave=42.

    Ponting and Viv have many similarities (attacking, superb under pressure, best fielders of generation, etc.). Viv was shown the door and let's hope Ponting goes out on a high.

  • Gerry_the_Merry on June 12, 2011, 17:28 GMT

    Boll, thanks for the correction. I mistakenly remembered that 5th test as having been much closer. Ponting is among the players I admire the most, and has been a spectacularly successful batsman, and indeed, some of his feats are incredible, like the Sydney twin centuries, especially the second one, and the MCG test Shrikanthk talks of above.

    Most of his runs have come in important contexts. His WC Quarter final century was thrilling also. So also his 156 in Manchester. My heart was in my mouth for the few minutes after he got out...

  • shrikanthk on June 12, 2011, 16:09 GMT

    With Anderson, Broad, Tremlett, Bresnan, Finn they have great depth, pace, bounce and swing. Swann is probably the best spinner in the world and I think they have the best test attack.

    Boll: Agree. More importantly, a lot of the guys you mentioned are still young and can serve England well for several years. Swann should be around for another 3-4 years. Anderson is still under 30. Broad, Finn and Bresnan are quite young.

    Among batsmen, you've Cook, Bell and Trott who should be around for another 6-7 years. Even Pietersen can stick around till 2015.

    That's what makes the prospects of English cricket seem healthy.

    India has just as good a side. But it is an ageing side and we are not quite sure how good the replacements are (in the longer form of the game). Gambhir will be a key figure around whom the batting will revolve. Rohit Sharma is the best young batsman in the country. As far as attack goes, a lot will depend on how well our leggies shape up (Mishra and Chawla).

  • shrikanthk on June 12, 2011, 8:22 GMT

    Ponting generally played those high-pressure innings when he was part of very strong batting units and settled,DOMINATING sides

    Again, I don't agree with your hypercritical, captious approach while judging non-subcontinent/non WI players. You can't blame Ponting for being born an Australian in an era of Australian dominance!

    Even post 2007, he has played a good number of fine Test innings (200 at Hobart against Pak comes to mind). Unfortunately, his bowlers have often failed to close out Test matches whenever he has scored big to put his team in good positions. Eg : Cardiff Test of '09, B'lore Test of '08 and most famously the Melbourne test of 2008 against SA.

    I remember that Melbourne test vividly. Ponting scored a 100 in the 1st innings and a 99 in the 2nd innings and yet lost the test match rather easily. It wasn't Ponting's fault that his bowlers allowed SA to score 459 after reducing them to 180 odd for 7!!

    Let's make the most of Ponting while we can!

  • Boll on June 12, 2011, 7:29 GMT

    @Pallab, I also think you`re being a little unfair on Ponting. Admittedly, his performances over the past 3/4 years have below his best, but if we look at the series Aus have lost in that time, apart from the recent Ashes series, his performances have been pretty good. Aus in India 2008, ave 38; SAf in Aus 2008/9, ave 48; Aus in Eng 2009, ave 48; Aus in Ind 2010, ave 56.

    I think he`s obviously been a great player under pressure or otherwise. A 4th innings average of 54 suggests as much.

  • Boll on June 12, 2011, 7:01 GMT

    @Pallab, I think you`re being a little unfair on the England bowlers to be classing them as `such an average unit`. Over the past couple of years they`ve consistently shown they have the ability to bowl teams out twice. With Anderson, Broad, Tremlett, Bresnan, Finn they have great depth, pace, bounce and swing. Swann is probably the best spinner in the world and I think they have the best test attack.

    Time wll if they`re good enough to take England to No 1.

  • Pallab on June 12, 2011, 0:16 GMT

    “Ponting has been one of the best players under pressure over the past 10 years.” SHRI:10 Years!Overstating it. Pointing generally played those high-pressure innings when he was part of very strong batting units and settled,DOMINATING sides. His weakened team needed him to deliver MORE post-2008 which his peers SRT and Lara almost always did under pressure in weaker overall units. SRT from 1992-2001 and Lara from 1995 till almost retirement.Ponting has done almost nothing to shore up a weak OZ batting unit for 4 years which actually shows how much his skills have declined (remarkable similarity with Dravid since 2007 in their collective freefall and their AWESOME purple patches b/w 2001-06).I still can’t believe how such an average Eng bowling unit beat OZ 3-1 recently –shows the lack of depth of reserve high-quality OZ batsmen since Ricky's fadeway and retirements of Martyn, Langer, Hayden.Retrospectively,one realizes the value of SRT’s 116 at Melbourne and his 278 runs in 1999 series

  • Pallab on June 12, 2011, 0:09 GMT

    @Alex/Gerry:I agree about SRT not being able to deliver that runaway/breakout legend innings in Tests since the 90s-SIMPLY because he puts TOO much pressure on himself to deliver(Zen-like SRT might not show it but criticism does rile him making him feel the pressure EVEN more).He still believes (rightly or wrongly cos of the presence of MIGHTY Sehwag, improved, highly-motivated –for-greatness Gambhir and 4th innings titan VVS now) that his presence matters all the time for India to seek decisive positions in Tests with the result he fetters himself and his unbounded talent.The best thing about his LONG career is that he is ticking off the boxes which he always wanted: improving his record in SA, quality- chase innings in OZ in ODIs in 2008, Test 100 chasing 387 against Eng. The 300 and that Laraesque 153 and Gavaskar –like 221 last innings scores in Tests should not be too far away. @Alex: Almost outlier comment to compare eccentric genius Fischer to that Incandescent Dazzler Lara!

  • Alex on June 11, 2011, 6:04 GMT

    @Ananth and everybody: In the face of Katich's epoch making press conference, I have to ask this non-thread related question which I asked last month as well: how come Greg Chappell is so terrible at selection policies and PR skills? He wrecked a team India that was on the rise and looks set to guide Australia to the bottom of the ranking chart.

  • Boll on June 10, 2011, 17:29 GMT

    @shrikathk, re.Ponting`s perfomance in the 2005 Ashes. I would agree, it`s hardly fair to suggest Ponting (360 runs at 40) crumbled in that series.

    @Gerry. re.Ponting(2005)/Lloyd(1975/1976). Yes, Lloyd had an excellent series with the bat (470 runs at about 47), but comparisons between the pressure situations in the series are hardly justified. The 2005 Ashes series remains probably the closest 5-match series of all time. WI/Aus 1975/76 was not only a comprehensive 4-1 victory for Australia, but none of the matches were vaguely close; Aus by 8 wks, WI by an innings, Aus by 8 wks, Aus by 7 wks, Aus by 160 runs. You mention the 5th test (dead rubber) that `may have been won` by WI if Lloyd had had more support. Come on! Aus declared at 3-300, to set 500 to win, WI were 8-230 and Lloyd`s 90 off 90 balls, got them to 330 odd.

    Australia won in a canter, and were rarely stretched.

    My first bat was a Clive Lloyd County Colt

    I`ll leave my comments on the mighty Ponting f` later.

  • Alex on June 10, 2011, 10:04 GMT

    @shrikanthk: Amen! I would mention Ponting's WC QF century as well.

    @Gerry: SRT has a very carefully hidden persona of a brat marathi soldier at his core. Nothing illustrates this better than what happened after Waqar broke his nose in the 4th test in Pak ('89-'90). Along with a few Pak players, Waqar advised the 16-year old SRT to get treated by a doctor in the dressing room. As Sidhu later recalled, SRT responded with a completely unexpected "Tujhya aaiicha gho!" I became his life-long fan with that. A puzzled Sidhu had to get the meaning of this phrase deciphered from Ravi Shastri. SRT hit the next ball from Waqar for four!

  • Alex on June 9, 2011, 15:45 GMT

    @Abhi and @Ramesh: On a batsman's utility, I had suggested Ananth a method on evaluating how often a batsman was (i) dominant and (ii) supporting. Roughly, if he faces 90 deliveries and/or scores 35 runs, he has done his job as a "supporter" ... if he faces 150 deliveries and/or scores 70+ runs, it's a fairly dominant innings. More shades can be added. Last year, Ananth did an article on something close to this but not exactly this.

  • Gerry_the_Merry on June 9, 2011, 13:13 GMT

    AD, the period wise difference pointed out by you could also have resulted from the wide range of the group 5 - 19.99 to 29.99. If we were to see the Group 5 BQI faced by each batsman, you would see a comparably wide dispersion, just as you see in the batting averages (I suspect). Hence I had proposed that the group 5 cutoff be fixed at 26, so that we can be absolutely sure that 2-3 all time great bowlers are there in each group 5 attack (right now, some Zimbabwe attacks are also there in Group 5).

  • Boll on June 9, 2011, 8:43 GMT

    @Abhi, haha, yeah we do seem to have our disagreements! Sorry, don`t mean to jump at your every comment, and always enjoy our `chats`.

    @Ramesh Kumar. Agree with almost everything you`ve said there - very considered points.

  • Gerry_the_Merry on June 9, 2011, 8:16 GMT

    Yesterday Ravi Shastri mentioned in a meeting that "In 1992, in Sydney, Mike Whitney (12th man) was going overboard sledging Tendulkar, who was practically boiling over...the other Aussies were stoking him too...Tendulkar swore he would punish them...I had to calm him down and tell him to let his bat do the talking...he said he would make a 100, and made 148*". Must have been a different sort of bloke then. I admired him a lot until 1997, and he reacted very positively to pressure. Somewhere along the line, expectations got to him I feel, and he changed.

    However, this is not a "courage theory" as Ramesh puts it. It is as i said earlier, a matter of personal preference. Ricky Ponting drew many admirers with his WC QF century. So did Dhoni. So I guess different yardsticks for different players...!

    AD - What i am trying to say is that period adjustment is happening through weightage redistribution between groups when comparing eras. Hence ANY explicit period adjustment is incorrect.

  • west indian follower on June 9, 2011, 7:30 GMT

    I believe, of the past 30-5 years that Viv Richards is the greatest batsman. This is because we have seen him as a true match winner (World cup 79), at true match saver and a very dominating, aggressive batsman, (career from debut-captaincy). However, despite his declining powers in the late 80s, when his eyes started to go, he managed to squeeze out the important knocks,60s, 70s to get the windies out of trouble and ensure their unbeaten streak. Whilst West Indian batting under Lloyd was very strong, under Richards it was not as potent with the likes of Gus Logie and Carlisle Best, solid, but not in the class of Lloyd, Kallicharan etc. Even Greenidge was not at his best during this time, although he did come up with some big hundreds, and he managed to mantain West Indies dominance during a transition phase, with Holding and Garner leaving the team. IMO, have we ever seen such a necessary, multifaceted, driven batsman and winner?

  • shrikanthk on June 9, 2011, 3:16 GMT

    Ponting - came out looking sorry in 2005. One more major contribution from him would have won that Ashes. The pressure was just incredible. I felt he crumbled

    Ponting almost single-handedly saved Australia the Old Trafford Test match in 2005 with a 156 in the 2nd innings, when nobody else crossed 40. He very nearly won the game off his own bat. Surely, you don't expect him to do that in every game especially against an attack as good as the 2005 English attack in helpful conditions.

    There's no doubt in my mind that Ponting has been one of the best players under pressure over the past 10 years (Laxman being just as good). I still vividly remember his "coming-of-age" innings against Pakistan in '99 at Perth. He batted at No 6. He had scored 3 consecutive ducks leading up to the test match. Aus was teetering at 54-4. And this guy scored a counter-attacking 197 in quick time against Akram/Younis/Akhtar on a Perth wicket.

    That innings pretty much summed up Ponting for me.

  • Abhi on June 9, 2011, 1:40 GMT

    Alex, Again flies in the face of all logic. You state "there exists a set of 40 innings that will reveal 80% about a batsman provided the batsman has played enough # innings across a variety of conditions"

    What you perhaps mean (and what you previously implied)is that 20% of inn. make up for 80% of a batsmans STATs. That is perhaps true for most batsmen.

    But ,the whole point is that in cricket how in the world can we hold up 20% (any 20%) of a batsman's career and say it tells us more than the remaining 80%?? The vast bulk and majority of a player's career!

    It's something like looking at the highlight reels of an error strewn Fed at the French Open..If the winner/unforced error ratio is 20/80 Fed will lose everytime. The 20% of trick shots tell us nothing.

    Again, not a very apt comparison- but simply looking at any arbitrary 20% of a players career without any concrete reason (was he injured? Was the team boycotted? Was there a war break? etc)is a gross distortion of a career.

  • Ramesh Kumar on June 8, 2011, 11:45 GMT

    Alex/Gerry/Boll,

    I need to jump into defense of SRT(not that ne needs one). There is no problem in analysing various parameters/criteria of batsmenship and say somebody is better in one's opinion on that aspect of the game. The problem comes when you conclude that one is best due to that overriding factor(s). Unfortunately we form our opinions on such overriding factor, like Gerry's theory on courage. Even with SRT fans, some aspects of his personality have caught the imagination over actual results. When comparing SRT with other greats, we can always say VIV is better than him on some areas, Lara on some, Kallis on some etc. But if we put all the objective parameters and analyse all these players against all the parameters across formats, I think we will probably see SRT coming out on top. That does not mean we should not have other favoutites or like some others play. The problem is that SRT fans go over the top in praising him inviting SRT bashing-both are unwarranted

  • AD on June 8, 2011, 10:54 GMT

    I have already mentioned the considerable difference between the first and this table. To ascertain why I looked at the batsmen in two rough groups, Period wise and Team wise. Period wise: 1) Chappell -37.3 Richards- 53.1 Gavaskar- 35,6 Miandad- 36.8 2)Border- 36.5 Waugh-43.2 Tendulkar- 39.3 Lara- 43.6 Atherton- 26.6 3)Ponting- 48.5 Kallis-42.3 Sangakkara- 49.8 Mahela- 40.1 Dravid- 36.9 Sehwag- 46.1

    Team wise: 1)Chappell- 37.3 Border- 36.5 Waugh- 43.2 Ponting-48.5 2)Richards- 53.1 Lara-43.6 3)SRT- 39.3 Dravid – 36.9 VVS -39.1 Sehwag-46.1 4) Sangakkara- 49.8 Mahela- 40.1

    The reasons for the variance from the initial tables seem to be - 1) Period factor – This seems to indeed be a very important factor. Given the same quality bowling attacks the 2000s batsmen outperform prior modern generations. This doesn’t seem quite right. 2) Pitch factor – Perhaps the reason for the outperformance of the predominantly 2000s batsmen could be the generally batting friendly pitches as compared to earlier generations. 3) Cut outs- Some batsmen may have had either good or poor innings clustered in areas just before or after a particular cut out. Minor differences between top class batsmen are understandable due to the many factors involved. However, the best 2000s batsmen outperform some of the best batsmen of earlier generations by more than 20% in some cases. Doubtless a Period/Pitch factor and a broader or more varied grouping would throw up different results.

  • Alex on June 8, 2011, 10:34 GMT

    @Abhi: I can't make a fool-proof argument for an idea in 300 words ... it's always a suggestion for Ananth. What I mean is this: there exists a set of 40 innings that will reveal 80% about a batsman provided the batsman has played enough # innings across a variety of conditions. Of course, scorecard alone cannot reveal the full worth (certainly not the aesthetics) but Ananth's work can be used to extract that out as much as possible.

    SRT has played great successful innings in dire situations: e.g., his first hundred & 103* vs Eng. Having played 290 innings, he had the opportunity to play more of those but is a bit lacking there. E.g., take his 84 vs SL. It was a dire situation. As far as the art of laying the bat on ball goes, it was a top drawer clinic. So, he was in full form but failed to reach the heights of a VVS/Lara special. That often happened with him. I do think his 204 in the next test was a great innings in a tough situation (albeit on a batting paradise). I am publishing these comments from a London hotel hobby, with no access to my Laptop, Androd phone and Cricket Database.. So will not be able to do my usual extensive responses. I can only act as the umpire. Ananth

  • Abhi on June 8, 2011, 9:18 GMT

    Alex, I'm afraid your logic is flawed. The 80/20 rule as you are attempting to use it in cricket amounts to the equivalent of say the Top 20% inn making up 80% of a batsman's stats.

    This, infact could well be a method of checking out how often a batsman flops to how often he plays well. The higher this ration-It may be argued such a batsman is truly "useful" "closer" to just 20 % of the time. (A better way to judge how "useful"would be perhaps to take "matches" instead of "innings")

    But,this too would indicate a "flop ratio". I would say that most Top batsmen would actually be pretty high up on such a ratio.

    But ,inadvertently you may have hit upon a new way of generating a "usefulness" quotient of a batsman.

    After all ,If for eg your Premier batsman performs only 20% of the time - just how useful can he be said to be for the team?

  • Gerry_the_Merry on June 8, 2011, 9:06 GMT

    Alex, of the lot you mention, Lloyd would probably make the cut, as per my criteria. Regarding Lara, he has played several blinders, but what makes me hesitate is that I remember many (generally forgotten) occasions where the Aussies / McGrath overcame him. 1) World Cup S/F 1996 - clean bowled by Waugh 2) WC 99, bowled McGrath 3) 1994-95 series, no real contribution while at his peak 4) 1996-97 series, same 5) 1996-97 OZ ODI series final, blown away by Akram. All high-pressure matches.

    Ponting - came out looking sorry in 2005. One more major contribution from him would have won that Ashes. The pressure was just incredible. I felt he crumbled. Compare with Lloyd in '75-76. Team went through hell, but Lloyd kept his batting together, and had he got better support, 5th test may have been won to make it 3-2, and then who knows?

    But no disgrace to any of the batsmen I have mentioned to be knocked out by Lara, Ponting, Lloyd. Kallis...I would pick Miandad ahead of him. And Gooch.

  • Gerry_the_Merry on June 8, 2011, 8:53 GMT

    Boll: I would hate the be the selector who had to choose a captain from that team - the best captains are the best at handling pressure...

    Ananth: So far I have made 3 suggestions : 1) on weightages (i think it should be fixed based on CTD balls / innings and not actuals in that innings) 2) period adjustments (no need for explicit adjustment as it is happening via the groupings) 3) creating home/away bowling averages based BQI rather than blended. Let me make a 4th suggestion (this one should not be limited by sample size problems): Since most bowling attacks have superior Inn 3/4 avgs, compared to inn 1/2, and since this exercise is placing batting averages relative to bowling averages cleanly, why dont you develop Inn 1&2 bowling averages and Inn 3&4 bowling averages separately? That will give a far more accurate distribution of innings BQI between groups. In fact, now that i think of it, items 3 and 4 can be combined, and I am sure the results will be in sharp focus.

  • Abhi on June 8, 2011, 8:41 GMT

    Alex, The type of inn.s you refer to , esp. the Lax specials - often just "happen". If you look at Ananth's best 100 inn. lists there are many batsmen who will not figure in any World XI.

    Tendulkar has saved and played pivotal roles in umpteen matches. In fact ,he first started saving matches since he was 16 /17 .Actually Lax is unique in that he has so MANY of such cliffhanger type of inn. I would say he is completely unique in that regard. These last ball /wicket type inn. occur perhaps just once or even not at all in the careers of most batsmen.

    I know Ananth doesn't like parallels with tennis- but for eg. till the recent Fed-Rafa match the greatest Wimby final was the Mac-Borg '80 one. Rest assured neither of them "planned" any of it. They just went out and gave it their all. Sampras ,by contrast has never had a great Wimb. final in the same league- he dismantled all comers. At the end of the day - Borg won. Was Mac a "loser" then? Yes- he lost the tennis match. But- without him you wouldn't have had a match of that quality in the first place. Similarly, in practically all the epic inn in Ananths lists the batsman offered chances. Had the batsman been caught just before the line it would have been a let down for the batting side , but how much do you take away from the inn. value?

  • Alex on June 8, 2011, 8:33 GMT

    @Boll: You are right on the volume observation ... Hobbs played only 60-odd tests in 25+ years, Sobers played 93 in 20 yrs, and SRT looks set for 200 in 24 yrs. If we cannot accept that Sobers would score 14000+ in 180 tests in modern era, we should be prepared to accept that Kallis is better than him.

    Stats have several limitations in deciding the true worth but a metric that I like a lot is the Top 40 innings simply because the 80/20 rule works so well in all aspects of life ... one can assign 80% weight to the top 40 and 20% to the rest. Top 20 percentile innings for most modern batmen will make a sizable set of 40 innings ... it is a bit too small at 16 for the Don. Now, how you rate these 40 innings can be quite subjective. In his 2009-10 analysis, Ananth had accounted for the peak performances ... I am not sure if he used the 80/20 rule.

  • Abhi on June 8, 2011, 8:20 GMT

    Boll Phew I always seem to get into some hot soup with you. My comments were basically towards some ppl from whose comments you glean a certain "tone". This "tone" often has little to do with actual cricket/batting/bowling per se. I have absolutely no issues with solid cricketing arguments- Howevever to MY mind Any such "solid cricketing arguments" Must comprise: All known factors including injuries. We have commentators passing snide remarks about "so-called" injuries- Apparently these things are all figments of our imagination and made up.

    Using plain vanilla stats are fine but the true and complete picture requires a finer understanding of the limitations of the stats and a certain "judgement" too(copied this from the latest Roebuck article) - which includes the inclusion of all known factors in addition to the stats.

    I've never had any issue with your comments- they have always been fair and square- and your "tone" has always been great. You have your faves and so do others. It’s all got to do with the “tone”.

  • Alex on June 8, 2011, 0:43 GMT

    @Gerry: I too agree with your rationale on why SRT has not risen that well under intense pressure ... 136, 98, 85, etc. are blazing exceptions. As Imran says, he probably needs to trust others more. I thought he had turned a corner after the 117* in CBS '08 & 103* vs Eng but, sadly, that did not happen ... his 84 in SL (2010) illustrates this the best: brilliant stroke-play but nothing to suggest that he would turn the match around.

    On your list, the glaring omissions are Lloyd, Barrington, Lara, Kallis, & Ponting. Ponting's stock has plummeted in the last 2 years but he used to thrive under pressure.

  • Ravi on June 7, 2011, 22:28 GMT

    Gerry, Alex, Keeping BQI aside for a moment, is there a general view that playing (scoring against) good pacemen is harder or easier than playing good spinners? Speaking of 'back to wall' innings, do you recall efforts such as Sunny's vs Pak in '87? And I totally agree with the view that a batsman is wrongly penalised when a captains decides to bring on a lesser bowler. This sometimes breaks long partnerships or dismiss a well set batsman- eg Ponting v Vaughan at Old Trafford '05. Also I notice Clive LLoyd, Kallicharan and Kanhai not mentioned as frequently in discussions. Ravi, if we go past the simple "whoever bowled" method, too many doors are open. How did Kumble bowl at the end of his 77 over spell against Slk. Was he better than, same as or worse than Kulkarni. Where do you place Hammond or Watson. What do you do if the fifth bowler bowled better than, say, the third and fourth bowlers. There is no clear solution, other than going with the accepted basis, which, if nothing else, does not fall a victim of subjective considerations.

  • Boll on June 7, 2011, 12:33 GMT

    @Abhi, I`m sure we`re all guilty of `arranging` statistics, or at the very least choosing which statistics to focus on. Still, some statistics require little `arranging, filtering or sorting` - batting average and highest score. While they are not perfect, they are certainly fairer than cumulative figures, which overwhelmingly favour those who have played in the last 20 years. re.averages (30 tests min) Bradman, Sutcliffe, Barrington, Weekes, Hammond, Sobers, Kallis, Hobbs, Tendulkar, Walcott (Top 10) From 2-10, the difference is 3.05 runs per innings. The test careers of these men spans a century and more, and yet you classify people who dare to disagree with your `calcified` opinion as `pointless` to argue with. For the record, in my test and ODI teams of all-time, I would have Tendulkar at No.4, and opener, respectively. However, I`m perfectly willing to consider that Sobers or Hutton (No.11 as per test averages) were just as good, or indeed that IVA was the best ODI batsman. Surely if this `venue` has any meaning or value, it should challenge our opinions, our biases, our recollections, and the oral/written history of the game. It`s great when these analyses confirm our pre-conceptions, but when they bring them into question, at the very least we should take time to reconsider.

  • Boll on June 7, 2011, 11:37 GMT

    @Gerry, love the team, hate to be the umpire! @Alex, re.Mohali 2010. Yes, it was obviously a test Australia should have won, even taking into account a few poor umpiring decisions. India got a few rough ones as well, but 1-1 or 1-0 India (including a tie?) would have been a fairer result. In the end Australia weren`t quite good enough, and Laxman was superb, as you noted. I also feel that a 1-1 was the most appropriate result. However this is not to take anything away from India who fought as very few have done before. Australia have also not been able to finish off matches recently. A failure which has cost and will cost them dearly.

  • Alex on June 6, 2011, 15:16 GMT

    @Abhi:

    1. First, I daresay Viv will not consider me an amateur on qualitative and/or math analysis. Second, since we can never know how A would have performed in B's shoes, your line of reasoning has limited validity. Also, Viv need not give me a look ... an autograph is more than enough!!

    2. SRT's durability is almost unique. Only Hobbs (& maybe WG) rivals him on it. Lara was denied it & Sobers-Bradman could have lasted 2-3 more years.

    3. SRT piling on runs in his *career* is a bit similar to Lara in one of his marathon *innings* ... after some point, everybody throws up their hands and says "OK, we know & accept that you are a class apart at this. Pl stop now unless there is something else to be seen."

    4. For me, the only thing missing with SRT is a back against wall innings of the calibre of VVS' 96/74* or Lara's 196/213. Will he ever play it? This being my favorite topic, let me clarify that I didn't refer to those innings as "match-winning" but as "back against wall". E.g., we all know that VVS' 73* would have been in a losing cause had the umpire given Ojha out lbw to Johnson ... Johnson & Aussies got robbed big time in the 2010 Mohali test but did not complain about this. Regardless of the end result though, VVS's was a very special knock in a dire situation.

    5. Incidentally, SRT has 12 hundreds in the last 24 tests ... 17 more in the next 28 and he will tie Don's 29 in 52!!

  • Gerry_the_Merry on June 6, 2011, 12:32 GMT

    Abhi, no serious follower will fail to recognize Tendulkar's greatness. But I completely agree with Imran Khan (who is the commentator I hold in the highest esteem) when he says that Tendulkar becomes tense under extreme pressure and this is visible in his facial expression. At times, like in Supersport Park, where also he made a tentative start and reached 111*, he overcomes it (in this instance, Dhoni was the catalyst). Generally, others, like Laxman, Gambhir, Dhoni etc. who are cut out for these situations, outperform in exactly these situations. Beyond this, it is personal preference - i admire courageous players - my World XI would be Gavaskar, IM Chappell, Richards, Laxman, S Waugh, Imran, Dhoni, Marshall, Warne, Lillee, Ambrose (12th man McGrath). I havent seen players with greater courage and steely will.

    I am no Brearley, but I think Tendulkar peerless consistency prevents him from raising his game under pressure, whereas others who blow hot and cold do.

  • Mahendran on June 6, 2011, 12:26 GMT

    "if Lara had played in different surroundings" Ananth..I would disagree on that. We all know what happened to Lara when he was in his pomp with almost all the support you had quoted. Success got to his head and all of a sudden he lost form and his average dropped below 50 before he revived himself due to the pressure created by his mediocre team mates. Great batsman like Lara will perform best when pushed against the wall. The same applies to SRT. In spite of all the support SRT had got he did not let his mind go astray like Lara and kept performing. If he had not got the proper support he would have performed even better. Look at his 90's record.

  • Abhi on June 6, 2011, 10:18 GMT

    Ananth, Oops ,MY line has come under Italics in your comment...viz: " It is an opinion which had begun to form several years back but now after 20 yrs it is pretty calcified- and the opinions of some random commentators in here is hardly going to change an opinion formed in this manner. " Precisely,we are not going to change our views at all. Certainly not because someone who for whatever personal, geopolitical or other reasons has a grudge against "X" batsman.

    So,we will all eagerly look at whatever stats or arguments suit us- and ignore whatever doesn't. Abhi, I gave my counter-points as an example of what can be given. Not for you to take it point by point and rebut it statistically or otherwise. Your views on Tendulkar are not going to change. Similarly others' on other players. I am on road and do not want another chain of arguments. However I have published one comment each.

  • Abhi on June 6, 2011, 1:57 GMT

    Alex, Gerry, I've realised for a while that it is pointless arguing with a certain type. You can always, always arrange, filter, and sort out the stats or any other argument to suit a preconceived notion. (I'm probably guilty of that myself)

    So, not matter what Gerry (or you) say, or what sort of particular filter you use – it is not going to change my opinion. Certainly some childish baiting is not. It is an opinion garnered not merely on subsets such as “generation on the average" or even smaller ones such as "Top 40 innings", “Worst 60 inn” ...or whatever filter you and others may chose.

    Let's say my view is thus- If Tendulkar had been placed in any other of his contemporaries shoes, he would have performed much better than them than vice versa (This means if any of his contemporaries had to go through the gamut that Tendulkar has- 16 yr old debut, ODIs, injuries etc the works – they would have fared much worse than if Tendulkar were in their shoes. The metrics I refer to when I say “worse” are the standard ones of “average” etc. I will only say that the alternate view, that, "if Lara had played in different surroundings, an assured support from world-class batsmen through most of the career, complete support from the board, public and fellow players, a complete financial security, adulation all-round, option to chose one's own retirement with confidence, maybe he would have averaged 56+ or in your words, done better". So let us all accept that we are not going to change the others' views. That is all. It is an opinion which had begun to form several years back but now after 20 yrs it is pretty calcified- and the opinions of some random commentators in here is hardly going to change an opinion formed in this manner.

    Incidentally, Richards too seems to have formed the same opinion. Try using some of your amateur arguments with him. Wonder what sort of look he would give you?

  • west indian follower on June 4, 2011, 20:32 GMT

    Ananth, I was readin GideonHaigh's ashes 2009 book, and same something very strange. Aussies top 6 barsmen average = range 42-64. England top 6= 32-80 (52 if you discount trott's one match. Also,england top 3 bowlers = 18@30, 14@40.5 and 12@45 Aussies = 22@27, 20@30 and 20@32. Aussies had 8 hundreds to england's 2. Aussies lost. Could you un an analysis on the value of individual one off performances to a test series, because clearly here England having players in the right place e.g Broad, Trott trumped Aussies superficialstatistical dominance Good idea. Will do some time in the future since I have got the Series work completed.

  • Gerry_the_Merry on June 4, 2011, 12:46 GMT

    Ananth, I have since read some of the earlier comments on the first point - fixed weightages for top 4 bowlers. I have seen your counter arguments also. Let me clarify. I disagree with the principle of giving varying weightages to bowlers in different innings depending on how much they bowled. What I am proposing is that the weightage be fixed for an innings based on stats BEFORE that innings.

    TO illustrate, you are taking CTD averages. This gives you an expectation of the bowling quality, against which batting must be measured. By the same logic, you must take the EXPECTED weightage of each bowler. There is a simple method for this. Compute each bowler's CTD balls per completed innings, and use those to weight each bowlers CTD average for that innings BQI.

    E.g. if Holding bowled 90 balls per innings and Garner 100, Garner will get slightly higher weightage. On the other hand, if Hadlee has bowled 180 balls per innings and Snedden, Chatfield etc 100, those are the weights. Gerry, I am on road and can send you only a minimal response of having seen the comment. Will look at it after I return

  • Alex on June 4, 2011, 11:36 GMT

    @Gerry: Amazing that you kept track of my comments. I was a expecting a stinging slap from Abhi and got a ringing endorsement!! SRT is not an outlier in his generation on the average and if we reduce the sample set of each batsman to the Top 40 innings, a few will arguably better him on a few valid metrics. That is the only reason why I cannot call him all-time #2 without having second thoughts.

    What separates SRT from most legends is not just the # runs but the art of batting that made those runs ... that, unfortunately, is never reflected by the score-cards. So, no stat-based argument can clinch the case there. IMO, Lara is certainly SRT's equal. Should anyone put a gun to my head to pick only one of them, I will figure out a way to take the gun from his hand and will shoot him instead ... so much for him never facing a McGrath special!!

  • Gerry_the_Merry on June 3, 2011, 7:18 GMT

    ...would mean an attack like McDermott/Hughes/Kapil Dev/Kumble, which in no way would make it to a "nightmare" combination for batsmen. But certainly a good competent attack. Deserving of group 4. I would propose cutoffs at 26, 31, 36 and 40. Anything above 40 is group 1, if you must have another group.

    5) It would be great, no AWESOME, to have such an analysis for bowlers also (Holding and Ambrose are sure to come out on top against their Group 5s, I would bet). Ananth, if you did that, would you not pick the top 7 batsmen, and compute an average, rather than take exact weightages according to runs scored in each innings (for instance, when Pakistan folded up for 62 in Perth, Sarfaraz scored 26, doesnt mean that Lillee's performance should get downgraded, right?).

  • Gerry_the_Merry on June 3, 2011, 7:05 GMT

    (b) Another argument in favour of considering fixed 25% wightages only for top 4 bowlers: A quick comparison of the top 7 spinners (I am arbitrarily picking Underwood, O'Reilly, Warne, Murali, Bedi, Qadir, Kumble) will yield a much higher average than top 7 pacemen (Holding, Marshall, Ambrose, McGrath, Roberts, Lillee, Imran, Steyn). Hence I bet that group 5 has the least 4th innings bowling performances. I think if your method is to be completely true, group 1,2,3,4,5 ALL MUST HAVE ROUGHLY EQUAL REPRESENTATION IN 1st, 2nd, 3rd, 4th innings.

    4) Last main point: I cannot believe that Steyn+Tostobe+Harris+Morkel almost made it to group 5. Good as that attack was, Ananth, you will remember we started on the trail of the best performers against the best attacks. The highest average i can think of for 4 all-time great bowlers is 25 (Roberts/Walsh/Gillespie/Warne). Give the batsmen an additional 1 run for unevenness, and you have a cut-off of 26 for a super group. 28, which someone else...

  • Gerry_the_Merry on June 3, 2011, 6:58 GMT

    ...Continuing on period adjustment: In your methodology, the adjustment is automamtically happening by shifting a 90s attack from group 4 to 5, or 3 to 4 etc. Hence to weight by period factor would negate this adjustment. Put differently, a batsman in the 90s will have had a higher % of runs against a group 5 attack, compared to a 2000-10 batsman, but in each case, the batting average against a group 5 attack will be equally merited.

    3)BQI bowler weightings: My previous post on this is several days old, but others (Alex, May 23, 2011 8:57 AM)have raised this point, and I dont agree with your reply. I think only the top 4 bowlers, with a fixed weightage of 25% each should be considered. There are three arguments in favour of this:

    (a) If a batsman forces the bowling side captain to bowl part time bowlers like Hooper (S Waugh 200) or Bradman (Hutton 364), by downgrading the attack into a lower quality group, you are penalizing rather than rewarding a deserving batsman...

  • Gerry_the_Merry on June 3, 2011, 6:44 GMT

    B) A bigger digression has already happened, with the core Tendulkar fans discussing whether he was the greatest or not, again. However I wish to butt in with the following observations:

    1. The classification methodology, admirable though it is, does throw up some counter intuitive results - for instance, Mohinder Amarnath is at 46 in group 5, along with Sehwag. We know why, and i suspect that this is due to the home away delta. Ananth, would you consider the following suggestion - for each bowling attack, if played in the bowler's home, use the SR/Avg/ER etc. of home only. If played away, use each bowler's away record only. This will AUTOMATICALLY re-calibrate all attacks without your having to consider home/away factor.

    2. Period Adjustment - this is a strict no-no. It is simply because the original rational in your own articles, which i completely agree with, was for adjusting the batsmen's averages: if a batsman averaged 50 in the 90s, and 50 in 2000-10, the former is superior...

  • Gerry_the_Merry on June 3, 2011, 6:38 GMT

    A) Alex, I am struck dumb at your comments. A 180 deg turn if there ever was one. First you say that you are having second thoughts about Tendulkar being a top 5 batsman, given his group 5 performances, and now calling him a terrible captain, critisizing his strike rates. You are encroaching upon my territory, though i dont mind rubbing a bit of salt into Tendulkar's fan's wounds by saying that considering him as top 5 would need to give excessive weightage to the quantum of runs he has scored. Barring that, given the massive gulf between teh top 20 group 5 batsmen and him, given his penchant for buckling under pressure (World Cup final, semi final), and his punishing record against Bangladesh when he had an injury, I anyway had a much lower opinion than his core fan group (which includes all other regular readers here but me).

    Anyway, this is not the place to rant. I have some comments for Ananth, after having been away tiger spotting for 2 weeks (actually a tigress, but i digress...

  • Alex on June 3, 2011, 4:47 GMT

    @Boll: What Trott has managed is not that extraordinary. He has scored 2,000 runs in 31 innings at ave=67 ... great stuff but his age was between 28 & 30 years during this period. Age 25---32 is the peak period and since the days of Hobbs, several have managed 2,000 in sub-40 innings during that patch.

    What's peculiar about Trott is that, just like Hussey, he started pretty late so that his start looks really amazing whereas by age 28, all players you mentioned (except Hussey) were already legends at the highest level. I do feel Trott might be able to pull a Gooch on all of us though.

  • shrikanthk on June 3, 2011, 4:28 GMT

    A very unattrractive batsman

    Ananth : I don't agree. He's just a very orthodox batsman who plays in the V to the extent possible. That doesn't make him unattractive. He's not "ugly" in the same sense as Shivnaraine Chanderpaul or Gary Kirsten.

    I haven't yet seen enough of his square-of-the-wicket strokes yet. If he develops that part of his game, he could well be a complete batsman. You would never be found wanting when it comes to defending Test players. You are correct. I could have used A "typical Test batsman". Ananth

  • Alex on June 2, 2011, 14:38 GMT

    @Ananth: I think top players missing even the test tour is OK since WI really are a terrible team now.

    SRT missing out on 6-8 league games in IPL4 would have worked better for MI since other guys would have had a chance to play long innings , different batting order could have been tried out, etc. etc. But, then again, we all know SRT is a terrible terrible captain.

    Somebody needs to say out loud that SRT was mediocre as a batsman too in IPL4. Forget the # runs and look at his SR instead. Him in T20 is only marginally better than SMG in ODI's. He routinely consumed 40+ balls near SR=110 when the par SR for 20 overs was 135. Must have been humiliating for Symonds-Pollard to watch this sorry spectacle and then come out to swing through the final 2 overs, if at all.

  • Boll on June 1, 2011, 11:54 GMT

    @Ananth and Shrikanthk, re. Trott. I must admit he`s far from my favourite batsman, but he`s had one of the most spectacular starts to a career ever. Without wanting to put a jinx on him (he remains about 200runs short of 2000 after 31 test innings), only 5 people in the last 50 years have reached that mark in less than 40; G.Smith(39), IVA Richards(36), Lara(35), and M.Hussey(33) - fine company indeed. G.Headley(32) will take some equalling, and Bradman(22) has well and truly flown the coup. Still, over the past few years Trott has looked as unlikely to get out as anyone I`ve seen. (SR of under 50 during such a dominant run probably doesn`t endear him to many) Boll, Trott is something else. A very unattrractive batsman, but impossible to shift. 2010 records along the same lines include Sanga(152) becoming fastest to 8000 (just pipping Sachin - 154), Ponting (247) joining Tendulkar as fastest to 12000, and Amla(40) smashing KP and Zaheer Abbas` joint record(45) as fastest to 2000 ODI runs.

    cheers

  • Alex on June 1, 2011, 3:21 GMT

    @Abhi & @Ananth: Ananth didn't publish my comment on this but I have strong views on this. So, sending it again.

    1. Too much cricket is being played. MSD has been complaining of it for 2+ yrs now. Just look at the Ind calendar since Aug '10. Till '76 or so, most top players would play 8-10 tests/year and that was it for the official international cricket ... now, they play at least twice as much.

    2. The WI have been minnows for a decade now. So, the WI tour is a perfect opportunity for younger & stand-by players for the international stage without BCCI/Ind running the risk of incurring a poor win-loss record.

    3. Aging stars have to be phased out anyway. It is not their fault that IPL is being like a merry bonanza on the way out.

    4. Why should SRT/MSD skip 8-10 IPL games? They had committed to the job and professional ethics demands that they give 100%. They are not skipping a tough tour: WI are a bottom of the pyramid & this is international cricket just in the name. No, Alex, that was not a deliberate non-publishing decision by me. I deleted your comment, along with a few other junk comments, by mistake and did not have the time to revert to you. Now that I am in Glasgow, I am told the weather is good and we are using the heater !!!, I will take corrective action. I have no problems with anyone skipping the ODI tours. If India loses that will be a lesson for the future. If they win, very good, great for the future. I object to top players skipping Test tours, that is all.

  • Suraj K Shetty on May 31, 2011, 16:05 GMT

    Ananth, I think we should look it from a different prespective. everybody is human..If someone pays you 2 million $ (mind you even for a rich it is not petty change) and also other undisclosed perks that we dont know about.. then you owe up some responsibilitiess.. [[ If you mean "a player" by "you", where does one draw the line. If a combination of Packer and Getty offered players 5 million dollars would you accept them if they gave up international cricket. Players play for their country. Actors act for themselves. So no comparisons. Ananth: ]] Sad but true, Present cricketers are not just players but have now become entertainers too, If we or any one of the @above guys, were to place ourself in their shoes.. only one in a few thousands might refuse such a offer.(even if they are rich take our bollywood brigades..they have so much money and earn a lot too.. still they dont mind getting paid to entertain at private weddings and parties.. easy earnings!!! IT IS HUMAN NATURE!!

  • Knowwho on May 31, 2011, 11:47 GMT

    @abhi,

    I do thk bcci is forcing players to play IPL. SRT would have probably gone to WI to play test series.He has nothing to prove by playing IPL. this holds good for almost all the senior indian players.

    During one of the interviews in IPL 4 SRT himself said as a 19 year old i would expect no rest but as a 37 yr old your body needs rest to recuperate.Probably tought he would have skipped later part of IPL and to my astonishment he carried on playing.May be corporate pressure from IPL big bosses.

    Same goes with lot of Srilankan players too Mahela, Sanga.

    RD did not fit here becoz he doesnot play odi series.IPL is it a boon to senior indian players (i dont thk so). [[ I agree with all what you say. One thing which intrigues me is the releuctance of the top players to take breaks within IPL. What prevented SRT or Dhoni from playing in no more than 8-10 preliminary matches. This would have enabled them to have rest in between, spend time with their families, visit Ranchi etc. They have the stature to insist that this be part of their contract. Ananth: ]]

  • Abhi on May 31, 2011, 8:24 GMT

    Going through the tables again, here’s some more interesting finds:

    1)At around Bqty 27.6 – SRT and lara are both around 40 (40.5 vs. 39.6 ) . After a 226 Lara then jumps to 44.6 in the next line. Richards at 27.6 is around 44. 2)Till Bqty 28.14 Richards is 42.7 . He then scores at an avg. of 78.1 from Bqty 28.14 to 29.89 ending up with an avg. of 53.1 just before the cut off of 30. 3)Richards avg. shows a straightish line up from Bqty 20-30. Lara is up ,down,up. Tendulkar is up till 26 then down. Then up from 30 on…etc.

  • Abhi on May 31, 2011, 7:15 GMT

    Ananth: Yes, Tendulkar and co.skipping a Test series is absolutely ridiculous. The SA tour, WC players should at least have skipped the first half or so of the IPL. They all looked absolutely jaded through the entire circus.

    So, now the play the whole circus and skip International Test series?! Am at a total loss for words.

  • Abhi on May 31, 2011, 7:13 GMT

    @Harsh Thakor, Knowwho and other typical Tendulkar lovers,

    As several ppl have mentioned the “problem” (depending on how you look at it) with the tables is the arbitrary and steep cut off ( 30,35 etc.) – As Alex says it is probably better to look at the top 2 groups in conjunction for a broader based perspective. Ofcourse,we prefer whatever format suits us (Me included ). As an eg. of my point I did a cumulative average exercise for Richards, Lara and Tendulkar- in Grp5 from best bowling to worst i.e from 20 towards 30. Tendulkars avg. peaks at 52.4 (at a Bqty of 25.55 and Below. At that level Richards avg. is 35.8 and Lara’s is 42.8 Thereafter Tendulkars avg. goes DOWN as we go towards Bqty 30…(i.e He does much WORSE against the 26-30 than the Below 26 group) Richards conversely improves towards the end of the scale (30) and infact Peaks right at his last inn. So, if we used groups such as Till – 26, Then 26-30 , 30-34 etc….Tendulkar would have been ahead in the First “Group 5” group. As I have mentioned pretty often- use with a pinch of salt and stats are like bikinis….. As per my reckoning then Tendulkar would be the best against the very best? So, it very much depends of what choice of grouping you use.

  • Santhana Krishnan on May 31, 2011, 6:02 GMT

    " What is this sudden pick-and-choose attitude of the Indian top players. "

    Ananth, there is nothing sudden about this. Tendulkar has been doing this for the past two years, though I agree this is the first Test series he has skipped. Dhoni skipped India's Test series against Sri Lanka. I think there is no reason to get emotional here.What else do you expect the players to do ? Skip IPL ? Can anyone ( except perhaps Sachin and Dhoni )even dare to do that ? After all, it's the BCCI's tournament. If I was in Sachin's position may be I would have done the same. What Gavaskar says is funny. He actually means that the BCCI should punish its own players who participate in its own tournament . We are expecting our players to do the "right" thing all the time.If the IPL was shorter and there had been about 10 days gap before and after the IPL, I am sure Tendulkar would have played the Tests.The players are no Gods or saints, they are just bloody good cricketers and average human beings.

  • shrikanthk on May 31, 2011, 3:48 GMT

    Their top two players were rusty

    Just goes to show no amount of Limited overs cricket can attune you for Tests. Proper 4-day cricket remains the best prep school for batsmen, young or old.

    Maybe we are reading too much into Sanga and Mahela's failure. [[ Let me put it this way. Their double failures have allowed people like me to hang this on the IPL and lack of acclamatization. They may prove me wrong in the next two matches. I hope for Sri Lankan cricket's sake that they do so. And Jayawardene's first innings was a nightmare. Ananth: ]]

    Nevertheless, I don't agree at all with all this talk of IPL being a better learning ground for young cricketers than English county cricket, as Eoine Morgan apparently claimed.

    Ofcourse one can learn a fair bit in IPL. But those learnings derived from 4-hr workdays in conditions and circumstances that are identical match after match can hardly be a match to the learnings on offer in highly variable and dynamic conditions that prevail in the English domestic circuit.

    English FC circuit, what with all its flaws, remains the best prep school for a cricketer. [[ You could add the Australian and South African tough FC cricket-spaces. The only thing that gets sharpened in IPL/T20 is the fielding. Ananth: ]]

  • shrikanthk on May 31, 2011, 3:37 GMT

    The Cardiff Test was a perfect example of the importance of, inherent excitement in and endless possibilities vested in Test Cricket

    Agree with the "endless possibilities" angle. Vindicates what we were discussing in one of the earlier threads. Test cricket cannot possibly be stereotyped. Each Test match follows its own pattern and is unique. Whereas, there's always a sense of deja vu with Limited overs cricket (be it ODIs or T20s).

    Four days of pedestrian play

    In my book, it wasn't pedestrian at all. It was fascinating to watch Trott and Cook adhering to first principles and beating the temptation of throwing their bat against a seemingly "weak" attack. And their approach was, in a sense, vindicated. Had England thrown their bat around on the 4th day and got bowled out for 450, we might not have witnessed this result. [[ You are right. I should probably have used the term "seemingly draw-oriented" play. The point was that England scored slower than Sri Lanka. However they had a clear plan of taking a lead of around 100 and then throwing down the gauntlet. So from that point of view we should term England as more purposeful. The Trott-Bell partnership was the key one in implementing this plan. It is my feeling that they would be no.1 by end of the year. They also have the finishing power. Compare Australia at Mohali and England at Cardiff. That too with 3 bowlers. While on England, you know that many of my tables have 2000 runs as the cut-off. Imagine Trott scoring the next 200 runs and crossing 2000 at 65+ average. He is now 19/1803/66.78 Hussey at this stage was 19/1898/82.52. Even today Hussey is comfortably 50+. So Trott should be at least a comfortable 50+ batsman, not often seen now. Ananth: ]]

  • Ananth on May 31, 2011, 1:33 GMT

    The Cardiff Test was a perfect example of the importance of, inherent excitement in and endless possibilities vested in Test Cricket. Four days of pedestrian play, rain and no great will to win, on either side was compensated by 3 hours of intense effort on one side and capitulation on the other side. To conjure up a win with a sub-100 lead on a rain-shortened last day was the stuff no.1 teams in the world are capable of. I have no doubt that England are on their way to no.1. This also raises the question of unpreparedness of the Sri Lankans. Their top two players were rusty. I love the Sri Lankans and am amongst a few Indians who wanted them to win the World Cup. However I cannot but think that their Board has not been tough enough and their top stars have let them down. I may be a doomsday predictor, but mark this, IPL will slowly kill Test cricket. Already we have an Indian 'B' team touring West Indies and the cat would be amongs the pigeons if they lose. And we have Tendulkar skipping a Test tour. Then he should gracefully retire from ODI Cricket. That would give him more time with his family. What is this sudden pick-and-choose attitude of the Indian top players. Normally Gavaskar is too parochial for my liking. But he has now hit the nail on the head. The selectors must not pick players who opt out. But catch Srikanth and Srinivasan doing that to MSD or SRT. Srinivasan has made Dalmia acceptable. Ananth

  • Knowwho on May 29, 2011, 18:05 GMT

    Hi,

    I was just going through the runs scored chart of recent indian batsmen against grp5 bowlers.found that dravid has played more grp5 innings than srt anf vvs only 4 less than sachin. if my memory serves rt vvs started to find regular place only after 2001 aus series. I would be interested to know how many grp5 innings did srt play during his so called injury period 2004/06. Did it actually have an impact on his overall grp5 average. [[ You have the Group 5 table right there with you. Ananth: ]]

  • Ananth on May 29, 2011, 10:40 GMT

    I have attached Andrew Simeos's work related file along with his comments. I am not able to add any comments since we are leaving for our holiday on Tuesday morning and I have to complete 99.94 things to do before we leave. Readers can draw their inferences. For a week after we leave I will have limited access and can publish comments, provided these are short and not in huge numbers. ""Thus spake Andrew" The Excel Workbook is attached. The format is as follows : Sheet 1 : Your data copied as is with unnecessary columns deleted Sheet 2 : Analysis page ordered by ascending BQI with box for entering Batting average tolerance and Go Button for analysis Sheet 3 : Analysis page ordered by descending Batting average with box for entering BQI tolerance and Go Button for analysis To exclude the problematic Bradman from Sheet 2 analysis just change his batting average to 0. When you want to include him , correct it back to 99.94. My workbook contains VBA code which you can look at if you want to understand the logic :But the working is quite self-explanatory . You will have to enable macros if they are not . Andrew P S : I have since analysed and found that it's 1:4 is not always cool. There is a lot of variation . Probably sine-ish curve. """ To download Andrew's Excel sheet: please click/right-click here

  • Alex on May 28, 2011, 13:33 GMT

    @Ananth: I speak 2 non-Indian languages excluding English and will cite URLs for SRT's interviews in those 2 but SRT can speak only Marathi, Hindi, & English ... apart from a mysterious one spoken with a bat in hand!

    @AD: Chess is an individual's game (bar the preparation). Still, if the comparison is to be made then, IMO:

    1. Don = Kasparov. 2. Hobbs = Lasker. 3. Sobers = Alekhine. 4. Viv = Capablanca. 5. Pollock/Kanhai = Tal. 6. Lara = Fischer. 7. SRT = Anand but I would rather equate him with Anand's own favorite Karpov ... a phenomenally balanced player. 8. Chappell = Spassky. 9. WG Grace = Botvinnik (they were alike in mind-games as well). 10. Ponting = Kramnik. 11. Hutton = Max Euwe. 12. SMG = Smyslov. 13. AB/Waugh/Miandad = Petrosjan. 14. Ranji/Trumper = Morphy.

    The above 18 batsmen might make the All-Time Top 25 list of most in terms of talent, record, & impact. A few of these 18 don't make my own Top 25.

  • arijit dasgupta on May 28, 2011, 11:53 GMT

    CONTINUED FROM PREVIOUS POST When my late great uncle, a very cultured and patriotic man and a Gandhian freedom fighter, told me the story (when I was a kid), his eyes shone with patriotic pride --- over Hunte’s act, though he was a foreigner and an opposition player. My elderly relative must have felt that his own patriotism --- which had led him to daring and selfless acts that I wouldn’t want to discuss here --- and Hunte’s patriotism were the same. He would go on and on about Sobers’ and Kanhai’s magnificent batting without a thought that their feats had helped WI give India a thrashing. A true patriotic man and a true cricket lover. Hope I have been able to express my point clearly. [[ I would have been proud and privileged to have met your uncle. Ananth: ]]

  • arijit dasgupta on May 28, 2011, 11:51 GMT

    Ananth, your comment about Indian fans being more sporting 30 years ago is spot on. Let me tell you two anecdotes. In the 1996 WC semi-final when Aravinda began lifting Lanka out of a mess with one of the best ODI innings I have seen, I was stunned at the crowd’s deathly silence. When I began clapping in front of the TV, my fellow paying guests told me to stop the noise and blasted me as “anti-national”. Later, the crowd set fire to the stands when India lost. [[ Again, you are spot on. I received dirty looks when I applauded Gilchrist's power-play. When Gilchrist's wicket fell, one guy who had a small kid in his arms started jumping up and down. I told him to take care since he had a kid with him. He rudely asked me to shut up. Recently I went to the India-Ireland game rather reluctantly. It was worse. This time the noise levels were deafening and so were the silent moments when Porterfield hit a few boundaries. Ananth: ]]

    The previous time stands were burnt in Calcutta was in 1966-67 --- in a protest against the police beating up an elderly spectator, not because India was losing. Conrad Hunte had then climbed to the top of the stadium amid the mayhem to prevent the West Indies flag from catching fire. When my late great uncle, a very cultured and patriotic man and a Gandhian freedom fighter, told me the story (when I was a kid), his eyes shone with patriotic pride --- over Hunte’s act, though he was a foreigner and an opposition player. CONTD

  • Harsh Thakor on May 28, 2011, 9:52 GMT

    The analysis proves the true greatness of Sir Vivian Richards,the greatest batsman of his day ,and arguably the best since Don Bradman.Viv simply devoured the greatest pace bowling like no batsman,including Bradman and adding his Packer record would have been on another pedestal.Viv was more prolific agaisnt great pace bowling than Sachin Tendulkar,who as the stats showed was very prolific against the weaker bowling attacks.The stats also show what Sobers and Hobbs may have achieved had they played in the modern era and that Brian lara at his best was ahead of Tendulkar.Ofcourse in overall analyis Sachin may well be second to the Don being the most complete batsman but Lara and Richards were much closer to him than the stats show.

    Justice is also done to Graham Gooch,the best batsman in my book against the graet Caribaen pace attack as he showed in 1981 ,1988 and 1991.However I feel Alan Border is beeter than the stats show as he was prolific against the greatest bowlers.

  • Mahendran on May 28, 2011, 6:03 GMT

    "You all talk about SRT. Watch him walk out to bat at the SCG, and you`ll realise why Australia remains the soul of cricket." You are right Boll as SRT himself acknowledged this fact during the last tour to AUS. When he played in Sydney he said that he could not make out whether he was in Mumbai or Sydney. The same goes to Lara. Both these players were welcomed with thunderous applause all over the world. They held that kind of sway.

    "My last experience in Test cricket was the complete silence which greeted the exhilarating strokes of Gilchrist/Clarke on the second morning of the Bangalore Test." Ananth..you have to give credit to the Chennai crowd. They were always sportive. Others were overwhelmed by Indianism I suppose. Nothing wrong with that though. [[ Mahendran, I have attended Test matches in various places in India including Chennai and Bangalore, two cities which were the fairest in supporting all players. My last experience at Bangalore changed that. I have not recently attended a match at Chennai. Unfortunately it is difficult to glean anything from telelvision. I hope they have not changed much. Ananth: ]]

  • AD on May 28, 2011, 4:56 GMT

    Re. recent crowd behavior, it has definitely gone south. Nastier and more immature. This is actually a worldwide phenomenon; Witness the incessant booing of Ponting in England, though it may be more detectable in India. There are a very few players who transcend national and other barriers.

  • AD on May 28, 2011, 4:38 GMT

    As an aside - On Chess, Fisher has been my hero since childhood. Kasparov was like SRT – the all encompassing genius, with skills to eventually overcome any opponent and conditions. Fisher was more like Lara- the romantic, eccentric genius. A well thumbed copy of “My 60 memorable games”, though completely dated now, is still prominent on the bookshelf. [[ Those are good comparisons. Although I feel Fisher under-achieved due to his reclusive nature and phobias. Lara was/is the opposite self. I think Anand has the potential to be the greatest if he wins the next World Championshio where he will be lucky to face Gelfand against whom he has a good record. Grischuk would have been tougher. Carlsson cried off. His loss. If Carlsson, still only 21, thinks he needs special treatment and withdraws from qualifiers, he does not deserve to be in the frame. Ananth: ]]

  • Alex on May 28, 2011, 2:22 GMT

    @Mahendran: AFAIK, the closest SRT came to commenting on the ball tampering incident was on the "Talk Asia" interview series of CNN back in '08 (see 7:00-8:30, esp. 7:56-8:30 on http://www.youtube.com/watch?v=Rdb0H5EOT_M&feature=related). It's a good interview.

    I personally feel his best ever interview was for IBN-Lokmat, Nov 2009, celebrating 20 yrs of international cricket. It ran for almost an hour. It is in Marathi though and is available on YouTube. [[ Imagine "Alex" referring to a Marathi video !!! Ananth: ]]

  • shrikanthk on May 27, 2011, 18:39 GMT

    Can the reason be that each team plays many (or all) others, so the BQI averages out?

    arjit: Spot on! Said the same thing in my previous comment.

    Yet so few of you can find a decent thing to say about the Australian cricketers, or teams, who have dominated test and ODI cricket for so long.

    Boll: I'm Indian. So there are exceptions :) Don't jump to conclusions based on the absence of cheers. It isn't disappointment but awe that accounts for the silence in many cases. The culture is different here. That is all. The rambunctious cheer for the home team is more of a ritual performed with gusto. The silence, in response to the opposition performances, is a mark of appreciation. [[ I differ slightly. It has now gone past the culture. The ODI/IPL crowds are no less revelling-noise making-parochial-aggressive-jersey wearing crowds than the Briish Football crowds. The only difference is that they are almost totally non-violent. I would have a problem in going to a local derby in London or Manchester. But not in India. However the level of appreciation is markedly different. When a multitude of fours/sixes, some of these part of a near-100-before-lunch from Gilchrist, are greeted by a stony silence and singles of Sehwag and Ganguly fetch rapturous applause, I am not going to analyze this into "silent appereciation". And I also realize that I am the wrong man in that crowd, not the others. Ananth: ]]

  • Boll on May 27, 2011, 17:23 GMT

    I suppose my point is that Australian fans, parochial as they may be, learnt to support `cricket`, a long time ago. Whether it was great bowling, or exciting batsmanship, or stunning fielding; it didn`t matter where you were from, if you produced it, you were applauded- loud and long. (Of course if you were soft, you got an absolute pasting.)As a young boy I well remember being woken up to watch Botham`s innings in 1981, or IVA a year later- and you stood to clap the century. Sachin, at any ground in Australia? VVS? Absolute respect, and often the appalling silence from Indian crowds in similar situations(am I speaking to 95% of people here?) is a disgrace. Yet so few of you can find a decent thing to say about the Australian cricketers, or teams, who have dominated test and ODI cricket for so long. [[ That is true. A four by a visiting batsman, in most centres in India, is greeted with stunning silence nowadays. My last experience in Test cricket was the complete silence which greeted the exhilarating strokes of Gilchrist/Clarke on the second morning of the Bangalore Test. I then decided never to go to a match in India again. If we get time next month hope to see a day of English cricket next month. However let me say that 30+ years back the Indian crowds were as quick to applaud the visiting team as they did the home team. Ananth: ]] Just read Ananth`s response @Alex 27/05, and previously. Wonderful to see a bit of chess/cricket discussion. I was always a Capablanca fan myself, but it wasn`t until Karpov/Kasparov that you could get (almost) live games replayed in the papers and play them out on your board at home. Exhilirating stuff.

  • Boll on May 27, 2011, 16:54 GMT

    @Abhi, apology accepted - you`re a generous man. We all get a bit carried away on occasions, not all of us are men enough to admit it though. I`ve certainly been guilty on occasions. [Of course the mighty `Tiger` (statistically, and by reputation, the greatest legspinner before, and perhaps after, Warne) remains one of the greats of the game.] I suppose what annoys me most about writing on here though, are the constant accusations of bias. Whether it be O`Reilly, or Bradman, or Ponting; or a particular Australian team; or Greg Chappell performing best against the Windies, there is rarely any comment which does not attempt to discredit their prowess. Sure, all of these players/teams have their weaknesses, but the team with easily the best test/ODI record of all time deserves some recognition for its contribution to the game. You all talk about SRT. Watch him walk out to bat at the SCG, and you`ll realise why Australia remains the soul of cricket.

  • arijit dasgupta on May 27, 2011, 16:40 GMT

    Re: any specific reason for 'bowling faced BQI averages' being closely grouped (35.52-37.41)? Bowling bowled BQI averages' are in the range 33.81(aus) to 48.15(bng)

    Can the reason be that each team plays many (or all) others, so the BQI averages out?

  • Boll on May 27, 2011, 16:13 GMT

    `Only a few days away, and so many comments which deserve a response.` I think that`s been a recurrent thought of mine over the past year or two, and I`m sure for the many of you who have been more regular posters than I. I do sense a strange disturbance in the force, however, and I hope it`s not a terminal thing. This has been a wonderful place to discuss cricket, primarily because of the stunning work Ananth has done, but also due to the reasonable discussion which has taken place. It`s a privilege to be able to talk about the game we love (and argue about it) in a place like this, and long may it continue. OK-back to the cricket!

  • shrikanthk on May 27, 2011, 16:00 GMT

    Ananth, any specific reason for 'bowling faced BQI averages' being closely grouped (35.52-37.41) ? Bowling bowled BQI averages' are in the range 33.81 to 48.15

    There is a straightforward cricketing interpretation for this.

    "Bowling faced BQIs" will naturally be clustered closer together. WHY?

    Let's consider the "BQI faced" index of WI. Now, WI batsmen of any generation would've faced a wide mix of bowling attacks. For instance, Richards faced bowlers of variable quality ranging from Roger Binny to Dennis Lillee. So, the mean "BQI" faced by him will regress to the mean. You can extend Richards to include all WI batsmen from 1928. Each generation of WI batsmen would've faced bowlers of variable quality resulting in a regression to the mean of 35.52

    When it comes to "BQI bowled", no such regression to the mean happens. Some countries have produced more than their share of great bowlers like Aus(33.81). Others have struggled to produce quality bowling talent(like NZ's 40.02).

  • Mahendran on May 27, 2011, 15:59 GMT

    "His behaviour is outrageously impeccable and there has been no black spot over 21 years of Cricket" Ball tampering??? I wish to think that this particular incident was a stupid mistake than anything else. He did not do it surreptitiously as ball tamperers do. A case of "Much ado about nothing" - thanks to the media. But somehow this remains a black spot in his otherwise spotless career a la "distraction" (how do you call "Dhrishti"?) to ward off the evil eye. [[ Even that, a minor incident was blown up. However as you rightly have said that is the "dhrishti". Ananth: ]]

  • Vinish Garg on May 27, 2011, 13:48 GMT

    Ananth

    For my comment on *not being a statistician* and your corresponding response, I posted another comment last night (about 20 hours back). Guess that you missed it by mistake though I know you do not miss a single comment. Can you please check that?

    Best Regards [[ No, Vineet, I saw the comment but did not publish it. I had only said that one other reader had gone off the blog citing the reason I had mentioned. I have no problems with your comment. In fact the blog is for the non-technical follower. Ananth: ]]

  • shrikanthk on May 27, 2011, 13:31 GMT

    Even gods have a limited shelf life

    Agree on all your points! Every era has its own conceit that dwarfs giants of an earlier era. I'm not denying that.

    However, this thread isn't just a reflection of the times we live in. It is a forum where more rational fans can attempt to transcend the transient hype and assess personalities with lenses that are unbiased to the extent possible.

    As a result I see no harm discussing WG or even Fuller Pilch as long as the context is valid.

    you INADVERTENTLY miss a Gavaskar

    You honour me by taking my impromptu list so seriously! No. I didn't miss Gavaskar inadvertently. Nor was that a list of the "most influential" batsmen. Influence was just one of the criteria I mentioned. I also said that one can feel free to add or subtract a name or two.

    Someone else might think that Muralitharan ranks among the greatest batsmen of the 20th century given the sheer enjoyment his swipes provided. That would be a perfectly valid viewpoint! Chill.

  • Alex on May 27, 2011, 12:29 GMT

    @Ananth: You are right on why SRT's peers might want to put him on the same pedestal as Bradman - at least in media if not in private.

    SRT *must* always be taken at his own word: to the best of my knowledge (& I have followed his career very closely since '88), he has never said anything on public platform that he did not believe in ... he will _not_ utter a word rather than say something he believes to be false. For all his greatness as a player, I put him on the highest pedestal for this rare level of integrity. [[ I have the same view as you. To me he and Federer are on the same plane. The word "role model" was invented for them. His behaviour is outrageously impeccable and there has been no black spot over 21 years of Cricket. All the more reason why everyone should take what he says as what he believes in. Why suddenly say he says something and means something else.That is not being fair to him. Another valid point. If one choses not to accept what Tendulkar says, maybe the same feeling of disbelief should extend to Warne, Lara et al. If Anand says he does not believe he is the best ever Chess player, we should give him credit for saying what he means. I still think Kasparov is a better Chess player overall. If anyone wants a different answer, why ask at all. Ananth: ]]

  • Pallab on May 27, 2011, 10:34 GMT

    “To Abhi: Would Tendulkar (or any modern great) have averaged 99.94 over 52 Tests if they had played in the 1930s. Again I would never ask this question myself. But you have made me do it.” Ananth: Could not resist since you PROFERED the question almost as a Debate issue. I can turn around and say that if SRT was made to play in just so-called flat pitches/dust bowls in South Asia in the ‘30s (and 1 other country) against primarily 2 countries or a Lara or a Sobers playing just in WI and Eng might have ended up with a HIGH average.. SRT with his talent might have ended with an 80 plus average (maybe not that MIGHTY Average of 99) in 50 plus Tests. [[ Pallabh Not just a pinch of salt is needed but a lot more. You have picked up 80 from thin air just as you plucked 70 for Bradman from thin air. I would appreciate your telling me how to run this blogspace. At the drop of the hat you come out with 10+ comments, despite my repeated requests to you to summarize the comments to 2 or 3. And if I edit the same you get upset. Only thing I can say is, THIS IS NOT THE WAY TO DO IT. I am afraid I have no time to read and publish all your comments. I am going to publish only what I feel are relevant comments, which I have been forced to edit.. And if you do not like that, nothing can be done about it. Ananth: ]]

  • Pallab on May 27, 2011, 9:59 GMT

    “The press going ga-ga over SRT's 20th year international year and completely ignoring RD's 12,000-th run in the same A'bad test is a total morale crusher Unfortunately, Dravid has been dealing with that for almost all his career.” ALEX: Nothing to do with the media ignoring his 12,000 runs landmark as most knew he had crawled (as against his high standards) to that landmark as against the speed of reaching the other multiples of 1000 landmarks. The same media has been generous to him and hardly dissecting his poor Test record in recent years. In earlier eras, Vengsarkar had to deal with Gavaskar’s shadow, Greenidge with Richards’ and Kanhai with Sobers’ as far as recognition went (am not talking about mere statistical comparisons). Dravid though has had his fair share of plaudits; [[ What about the unique landmark of 200 catches. Just to be swept under the carpet as one which comes in with the rations. Do you seriously think the media sits there and analyzes that Dravid crawled thru his last 1000 so we should not applaud his 12000 that much. Ananth: ]]

  • Pallab on May 27, 2011, 9:41 GMT

    “SRT himself has said that the Don was beyond everybody and Abhi should accept his hero's verdict as the unadulterated truth.” ALEX: Surprised (cos of your quite fantastic, nuanced assessment of Indian cricket and storied batsmen’s legacies in Ananth’s last blog) that you are accepting SRT’s public statement verbatim as a ready admittance given that SRT’s personality and values systems (along with the other gents VVS, Dravid, Kumble) are too well known. You would not expect a SRT to publicly boast that he is greater than so and so or that he has practically butchered Warne or that Warne is ordinary. But you can expect a SRT to say the same perhaps privately to Atul Ranade and Atul Bedade (two of his closest cricketing buddies). As I have mentioned previously, Indian cricketers are generally too gentlemanly or submissive to be publicly bombastic about their achievements or even undermine/underrate achievements of opposition players. A Ranatunga, a Imran, a Ganguly were made of CONTD.

  • Pallab on May 27, 2011, 9:20 GMT

    "Any serious cricket follower who puts Don the batsman second to anyone needs to schedule an urgent appointment with his family doctor with a vague complaint that "something is wrong with me". Lara’s quote: “He is our period's Don Bradman. Forget the difference in averages with Bradman but whoever I have spoken to who have seen very old players in action, they believe that he [Bradman] would not have averaged 99 in today's cricket.” “SRT himself has said that the Don was beyond everybody and Abhi should accept his hero's verdict as the unadulterated truth.” ALEX: Are you serious? Are cricketers such as Lara (MSN Cricket archives) and Gavaskar (rediff cricket archives) who have just recently mentioned the same not cricket followers? ALEX: Surprised (cos of your quite fantastic, nuanced assessment of Indian cricket and storied batsmen’s legacies in Ananth’s last blog) that you are accepting SRT’s public statement verbatim [[ I am sorry to say, it is now the fashionable thing for peers to put Tendulkar at the same level as Bradman. I may hazard a few reasons why. As far as Tendulkar's statement he knows that what he says is the reality. Give him credit for accepting the gulf. And if you do not want to accept what he says you should not ask him the question. Ananth: ]]

  • Pallab on May 27, 2011, 9:14 GMT

    “It may also be essential to consider the batsman's influence on the game (technique among..) and on cricket in his country. ..to make a shortlist of the greatest batsmen of the 20th century, I'd come up with the following names Ranji, Hobbs, Bradman, Headley, Sobers, G.Chappell, Richards, Tendulkar, Lara. “ SHRI:You traverse oceans and across generations to vociferously read about cricketers of the “ancient” past (and even comment on some obscure bowlers with 10 Test records with ease!) but you INADVERTENTLY (or otherwise) miss a Gavaskar who along with Hobbs was among the 2 foremost technicians in all cricket history. I would surely hope that as a serious historian, you didn’t have any bias in drawing up that list. Gavaskar ticks all the boxes for influence on technique (the only one after a generation gap after the 3 Vijays retired including S. Manjrekar’s father ) and influence as the first neo-colonial Asian batsman to exert influence on world cricket. [[ Yes I think no list of influential batsmen, if it is of size 10 and above, can exclude Gavaskar. Ananth: ]]

  • Alex on May 27, 2011, 9:01 GMT

    @shrikanthk: I take Ananth's permission to state this comment since I have derailed the comments into a discussion of the Don whereas the original objective was to simply defend his placement in these tables.

    Even gods have a limited shelf life: 50 years back, WG Grace had to be named among the all-time top 2-3 cricketers. Today, people disregard him and, at least in India, quite a few argue that the Don (whose career started 50 yrs after WG's started) ranks below SRT ... recently, even Ian Chappell said that the SRT-Don comparisons do make sense to him.

    WG is so far away in the past that I have little connection with his achievements ... even my grandfather couldn't have seen him bat. One fears that as the years roll by and cricket changes with T20's & all, the same will be the fate of the Don with younger generations.

  • Andrew Simoes on May 27, 2011, 8:09 GMT

    @Alex : Thanks for the interest. As I mentioned , I have subsequently found that a round 4 is a poor indicator. My workbook , however , provides a good means for analysis. If you want to normalize Lara or anyone else, I found that their averages must all be graded to the mean of the BQI sample space. (Hammond and Sutcliffe could be excluded but here , I did not). This worked out to 35.5 . Tendulkar's BQI was 35.09. The difference from the mean : 0.41 Players generated by tolerance of 0.41 : 15 . Resultant batting average correction : + 0.4 . Tendulkar's corrected average = 56.95 + 0.4 = 57.35 Lara's BQI : 33.44 . Difference from mean : 2.06. Players generate by tolerance of 2.06 = 56 . Resultant batting average correction : +10.25. Lara's corrected average = 52.89 + 10.25 = 63.14 . Sachin fans note : I am NOT saying Lara is superior to Tendulkar, we are just having a little fun with averages here. Similarly Bradman's correction : -5.5 . Corrected average = 99.94 - 5.5 = 94.44.

  • Arjun on May 27, 2011, 6:36 GMT

    Ananth,

    The average BQI of WI and AUS during their peak period(according to your previos article) is....

    For WI - 30.78 (from Test# 764 to 1371)

    For Aus - 30.06 (from Test# 1121 to 1879)

    from warne's debut it comes down to 29.36 from mcgrath's debut it further comes down to 28.69.

    In warne-mcgrath ERA it is '28.55' although it contains few tests in which atleast one or both of them didn't played; like 2003-4 AUS-IND series.

  • Abhi on May 27, 2011, 4:50 GMT

    Arjun,Ananth Re. the BQI paradox ,If we could check the figures pre and post 1990 this may give us some idea about the impact of the "minnows" [[ Good idea. Only that I am running behind three trains and barely have the time to post responses. So I would appreciate if someone else can do it. Import into an Excel sheet, take couple of copies, sort and do a subtotal; that is all. Ananth: ]]

  • shrikanthk on May 27, 2011, 3:07 GMT

    I don't rely too much on Cardus: a professional romantic journalist like him cannot be that different from Harsha Bhogle

    Alex: Maybe. But I think Cardus is often unfairly labeled as a "romantic". If you read his match reports in the Manchester Guardian (especially the 1938 Ashes tests), one senses a very perceptive hard-headed analyst who obviously understands the game very well and has a genuine appreciation for the game's techniques and nuances.

    Don was not averse to lunging well forward

    You cite his last ball dismissal. I'd like to cite clips of his defensive play at Trent Bridge in the 2nd innings of the '38 Test, when he scored one of his slowest FC hundreds.

    http://www.youtube.com/watch?v=n5Dl-uCnEH8

    Check out the section from 1:48 to 2:05. Just goes to show how eager he is to play back in defence to the extent possible unless he is absolutely confident of reaching the pitch of the ball. Most modern batsmen would've lunged forward to many of those deliveries

  • Alex on May 27, 2011, 1:43 GMT

    @Ananth: I meant that the huge &/ frequent losses of BD/Zim/NZ helped their opponents (England of yore, Aussies, and Pak of 90's ... Waqar used to wreck them) take plenty of wkts at low averages with good SR. That adds to the BQI mismatch.

  • ecottle on May 26, 2011, 20:27 GMT

    A drawn series wasn´t bad for West Indies. However, Pakistan is a team going through a trsnsition, much like the West Indies.

    The old West Indian problems are still there. The top order batsmen are not leaving enough space for the incoming batsmen to build on. Instead, the new batsmen are always facing bowlers who are on a high.

    The team has a lot of players who star for an innings. There is no constancy in play. Here is where West Indies need to get things right. Ramdin should be back, and room should be made for Gale, Bravo and Pollard. They are still the best we have got, and cannot be ignored for the Indian tour. We must put out the best team against the World´s No.1. Its useless bringing in untried players against the Indians. We must play our best.

  • Alex on May 26, 2011, 20:12 GMT

    @Ananth & @Arjun: A few reasons why the BQI mismatches exist:

    1. Better bowling attacks (Aus, WI, Pak) have played more # tests than poor bowling attacks (Zim, BD, & NZ). [[ Yes, that seems to be the major reason. Ananth: ]] 2. Zim, BD, & NZ have possibly played a lot of matches vs the good bowling attacks and suffered losses. This increases their own bowling BQI while lowering the bowling BQI of their successful opponents. [[ I do not know wjy losing matches should increase their BQI and lwoer opponents BQI. After all BQI is a function of the Ctd average and Balls bowled, irresoective of the result. Ananth: ]]

    @shrikanthk: Don was not averse to lunging well forward. E.g., the YouTube clip shows him in full forward lunge to _defend_ the last ball he ever faced in test cricket. What is true is that he strongly preferred staying within the crease while playing the ball. How sacrosanct he held this can be seen from how he dragged his front leg back after getting bowled on that zero ... eerily similar to how SRT does it, BTW. I don't rely too much on Cardus: a professional romantic journalist like him cannot be that different from Harsha Bhogle.

  • Aditya Nath Jha on May 26, 2011, 16:33 GMT

    dear ananth - for me, george headley has been the only serious challenger to bradman (though, admittedly, he does fall short on all objective parameters). his numbers on this list do nothing to make me change my opinion :)

  • shrikanthk on May 26, 2011, 15:09 GMT

    Don did not play primarily off the backfoot. Even the youtube clips show him bat well off the front foot.

    Alex: By "primarily", I meant a strong preference to play off the back foot atleast in defence. Ofcourse, no one can play off the back foot or the front foot all the time! Footwork is determined by the length of the bowling, after all.

    Nevertheless, I quote Cardus:

    "Bradman is seldom, if ever, detected in the act of playing forward - full stretch".

    I've also read Harvey, Morris and Loxton in different places talking about Don's strong preference for the back foot.

    This suggests that he was a practitioner of Ranji's famous maxim - "Play back or drive". The forward defensive lunge wasn't quite his favourite.

    Moreover, Bradman played in an era when the "forward defensive" wasn't quite a craze as it became later in the 50s. CLR James writes on this subject perceptively in his essay The Welfare State of Mind which can be found in the book Beyond a Boundary

  • Arjun on May 26, 2011, 14:10 GMT

    Average BQI faced by batsmen. Win(35.52) Aus(36.12) Ban(36.41) Pak(36.44) Zim(36.66) Nzl(36.67) Eng(36.81) Ind(36.85) SAF(37.35) SLK(37.41)

    Average BQI bowled by bowlers Australia(33.81) Pakistan(35.69) England(35.97) WI(36.19) t Saf(36.27) Slk(37.54) i Ind(37.97) Nz(40.02) Zimbabwe(42.14) Bangladesh(48.15).

    Ananth, any specific reason for 'bowling faced BQI averages' being closely grouped (35.52-37.41) ?

    Bowling bowled BQI averages' are in the range 33.81(aus) to 48.15(bng) [[ Arjun This is absolutely amazing and requires thinking and some contributions from other readers, especially those who can give a cricketing interpretation. In fact after I sent you the message on asking for Bowling BQI based, I did the Excel work myself and can confirm that I got the same figures. And the 6837 sample has a range of 18.94 to 60.00, distributed fairly uniformly, typical distribution. Each country has the same wide range. However when we do the averaging of the BQI faced by Batsmen, this has a data range of 2.5% either side of Pakistan which is almost at the dead centre with 36.44. The BQI of the Bowlers is lopsided. The mid point is 40.98 and only the two weak teams exceed this, as expected. But the range is 40.98 and a spread of 17.9% on either side. Can it be the fewer number of matches played by the poorer bowling teas such as Bangladesh and Zimbabwe. Only 68 matches and 83 matches respectively. Anyhow you guys, Aditya, Andrew, Alex, Abhi, ShrikanthK, AD et al comment on this seemingly peculiar distribution. Ananth: ]]

  • Alex on May 26, 2011, 12:48 GMT

    @Ananth: Andrew's comments are really interesting. Can you pl email me his work? [[ Yes, once I go through it myself, I will upload and provide a link. Ananth: ]]

    1. I am happy to see Viv & SRT on the same level. As per Andrew's method, Lara's adjusted average becomes 52.88 + 4*0.65= 55.48. Likewise, Chappell's moves up to 57.4. Makes sense.

    2. SRT fans need not despair: if the unfortunate Mar '04 - July '06 period is removed, his average in Andrew's setting moves up to almost 59 ... even including that phase, his performances in your Groups 4 through 1 are the most rounded (bar the Don, of course) with the lowest ave=57 and highest ave=81: we saw something similar in his innings-wise breakdowns last year.

    I have followed your blog for almost 3 years. This article is probably the best on eliciting brilliant ideas from the readers. I suggest you follow the same route in obtaining definitive All-Time Top 100 lists.

  • Arjun on May 26, 2011, 12:34 GMT

    West Indies have faced, on an average, the toughest bowling attacks over the years. They have batted in 824 innings, 35.52 being average BQI. Srilanka(37.41) and Saf(37.35) have faced 'easiest' bowling.

    RE the bowling attackof teams, the list is given below. These numbers are across 130 years.

    Australia(33.81) Pakistan(35.69) confirming their bowling supremacy England(35.97) aided by yesteryear attacks WI(36.19) recent years bowlers damaging index a bit Saf(36.27) Slk(37.54) aided a lot by murli Ind(37.97) poor SRs of spinners not helping index Nz(40.02) Zimbabwe(42.14) Bangladesh(48.15) the worst ever. [[ Arjun, I presume you must have used the testgrp6.txt file. If so the only thing I must warn you is that I have capped the BQI at 60.00 for the few innings exceeding 60. This really should not matter, but you need to know. There is no lower limit since only three innings went below 20 (these are explained in the article) and I decided to leave these as they are. What about the complete list of 10 teams for the first average. Ananth: ]]

  • Alex on May 26, 2011, 8:06 GMT

    @shrikanthk:

    1. Don did not play primarily off the backfoot. Even the youtube clips show him bat well off the front foot.

    2. Just like McCabe, he looks very modern (unlike Hammond & Hutton) on shot-making and on running hard between the wickets ... wouldn't need many adjustments there for ODI's & T20's.

    3. On talent, Bradman himself thought that a few could bat better than him and said "but they kept getting themselves out". Of course, the players since then may well have surpassed him on technique, range of shots, etc., etc.

    Doesn't make much sense to compare "A" to "B" anyway ... much better to appreciate the uniqueness of such players and thank them for the memories.

  • Andrew Simoes on May 26, 2011, 7:35 GMT

    Now , as per the analysis workbook I gave you I decided to compare Tendulkar and Richards . The difference between their BQI is 1.79 . So I entered 1.79 as the tolerance in the BQI analysis page ( sheet 3 ) and it threw up 41 players . Trying to achieve 41 players from the Batting Average analysis page ( Sheet 2, excluding Bradman ) I found that 7 does the trick! But if we Include Bradman , we need 40 others . So the true value is not 7 but 6.7 . Hence corrected batting average of Richards = 50.24 + 6.7 = 56.94 Tendulkar's is 56.95 . Quite uncanny I thought! [[ Andrew, I will go into it and assimilate what you have done and then see what further can be done. The problem is that I am also hard pressed for time since we are going on a holiday in UK and Europe between 1 and 25 June, hoping that the black cloud from Iceland moves further eastwards. But will spare some time for your work. Many thanks. Ananth: ]]

  • Andrew Simoes on May 26, 2011, 7:19 GMT

    I have mailed you. Check it. I have since revised my conclusions ... you can help me with that , perhaps .

  • Vinish Garg on May 26, 2011, 4:38 GMT

    Ananth

    To say that this analysis is your best effort will be a disrespect to your past works. However, hats off to your passion, diligence, and love for the game, to your patience with the readers.

    Since I do not understand statistics (BQI or whatever) like some folks do here regularly, I cannot really participate in technical details of analysis. All I can see is that I enjoy every comment, every response, every update, passively... le "They also server who only stand and wait". God Bless, always. [[ I lost a reader whose main complaint was that I had not played cricket and I was not a statistician. Mind you he did not say my articles were not good. In other words my not being a statistician was the problem. I have always tried to make the articles readable and understood by 90% of the readers. The day I don't do is the day I will lose readers like you which I hope never happens. I would rather lose 10 like the above-refernced gentleman than one like you. Ananth: ]]

  • craigmnz on May 25, 2011, 19:40 GMT

    (contd) I'm afraid I've not downloaded the entire table so haven't been able to do this comparison myself but would expect to similar disparaties in percentages between Hammond and Ponsford and between Hutton and Harvey.

    so the question really should be not did these batsmen inflate their averages by feasting on substandard attacks but did they when the opportunity presented take advantage and cash in as they should?

    Hammond certainly did - he warmed up for his 336* with a knock of 227 against an equally inexperienced attack in the previous test in 1933. Mind you the inexperienced attack did remove the English openers (Sutcliffe, Paynter) in the first 2 overs. It was all downhill after that.

    I'd like to think I'm arguing for an understnading of these excellent statistics in a larger context than simply debating the merits of our favourite players. After all for my money Glenn Turner was NZ's best batsman and my father always favoured Bert Sutcliffe. [[ Craig, again my mistake. I will go back to your second comment and see what has to be done. Now I am publishing this as it is to clear the comment. Ananth: ]]

  • craigmnz on May 25, 2011, 19:29 GMT

    Anath Thank you for your gracious responses to my slightly bitter comments - possibly more than they deserved. I want to explore my second comment at little further because I don't think the real point I wanted to make came through in my venting at the patronising attitudes of Australian cricket authorities down the years. Another contributer has pointed out that 70% of Bradman's test cricket was played against England. Far less of Hammond's test cricket was played against Australia. The differences in percentage of innings played agsinst the different quality attacks attest to this. Even Hutton's post-war career reflects this. People has in the past made much of the fact that Bradman never played on the sub-continent, but let's be fair, he was never given the opportunity. He never played on NZ's slow seaming tracks either. My personal belief is that had he done either he'd've flogged the attacks on offer - in much the same way as his English contemporaries did but moreso.

  • shrikanthk on May 25, 2011, 19:00 GMT

    Ananth, Arjun: I saw the illustration of the "reciprocal" method just now! Wonderfully elegant. [[ 80% credit to Arjun for showing me the method and 20% to me for realizing the impact of this and implementing the same !!! Ananth: ]]

    Ananth: I only had one question. You must have used limiting values for both CTD SR and Eco rate right? What are those values like? [[ Overall limiting values are s/r: 30 to 100 & rpb: 0.25 to 1.5. However the within-50-wickets values are differently set within these numbers, that too depending on a player's career achievements. Ananth: ]]

  • shrikanthk on May 25, 2011, 18:20 GMT

    the name of this blog is itfigures. as the name suggests, this is a statistics based blog

    Aditya: Agree on that. Nevertheless, I wonder whether it is possible to divorce cricket from cricket stats, try hard as we might.

    Cricket is about personalities. Numbers can illuminate personalities yes, but not entirely. I may not agree with Pallab's mode of discussion. Nevertheless, discussion on history/personalities/subjective notions of talent and cricket-worth are bound to creep into the discussion.

    We have to accept that. If we confine ourselves to numbers only, we may have to reach the conclusion that Johnny Wardle is the greatest spinner of all time :) (Just kidding...but I guess you got the drift) [[ Shri, I think Aditya, or for that matter, are not saying that we should confine out=rselves to stats and numbers. I myself have been guilty of pushing non-numbers-based issues, even recently I brought in Gayle/Pollard issue. But that was to express my displeasure at someone forsaking his country for club. What we do not want is to bring in some other factor with a clear view to put down a player. Ananth: ]]

  • shrikanthk on May 25, 2011, 18:08 GMT

    his art of batting was low on aesthetics

    I am not sure about this. From what I've seen (the Bradman Era documentary which made a huge first impression when I was 8), he seems a very dynamic player with excellent footwork.

    By all accounts, he played primarily off the backfoot nearly all the time. Also, used a very light bat which meant greater convenience at playing horizontal bat shots. Neville Cardus once wrote that he always got on top of the ball and seldom played away from his body.

  • shrikanthk on May 25, 2011, 18:01 GMT

    Was he a good captain?

    I think he is in the top 3-4 captains of all time in terms of winning percentage. Ananth can confirm this.

    Led by example. Good tactical captain as evidenced in '36-37 when he responded dynamically by shuffling the batting order to maximise team's chances on a wet wicket.

    It is vital to recognise the opposition- good Eng, avg SA, poor Ind,WI.

    I'd say : Good Eng, Avg WI, Below par SA and India. West Indies had a pretty decent side, with some good fast bowlers back then. In fact, WI didn't lose a single series to England in West Indies back in the '30s.

  • shrikanthk on May 25, 2011, 17:49 GMT

    Was he ever the highest scorer in FC/Tests?

    Yes to both. He held the highest FC score record (452* against Queensland) for nearly 30 years, I think, before Hanif Mohammad broke it in 1958-59. Held the highest Test score record briefly (334) between '30 and '32 before Hammond broke it against NZ.

    Was he seen as a match winner against strong teams?

    The only two series Aus lost during his playing career were 1928-29 and 1932-33 Ashes. The first one was his debut series and the second one was the Bodyline series. Besides these two blips, he was winning matches for Aus all the time.

    In 1936-37, he did what most "match-winners" aspire to do. Lead his team to a 3-2 win against a strong English side after being 2 down. He did this by scoring two double centuries and one century.

    Another memorable match-winning performance under pressure : A fourth innings 173* on a wearing pitch in '48 Leeds test.

  • Alex on May 25, 2011, 17:08 GMT

    @Ravi: Answers to your questions in that order.

    1. Yes. E.g., British tabloids ran the headline "He Is Out" as early as 1930. Also check the Wisden reports starting 1930. 2. Other milestones yet to be surpassed include most runs in a series, 300 in a day, etc., etc. 3. Yes. His 334 was a WR in tests. 4. England was the only strong team back then and thought him a match-winner. In fact, bodyline happened because they believed it was the only to overcome the Don and that once he was gone, they will beat Oz easy. 5. As a captain, he won often but many of his team-mates did not like him as a captain. Undoubtedly a strong personality.

    The Don had only Eng as the strong opponent during his entire 20 years. He faced them for more than 70% of his career and averaged 90+. By all accounts, his art of batting was low on aesthetics. However, it was unique & original. People knew that a genius of rare order had arrived even before his run orgies started in tests.

  • Suraj on May 25, 2011, 16:47 GMT

    Thanks and many thanks.. This link has been bookmarked for ever!!

  • Abhi on May 25, 2011, 15:14 GMT

    Srikanthk You are right of course. And it isn't all Ananth's fault. He reckons some of us should show some restraint and stick to the main line of the Topic. But,as you know- in cricket (all sport?) the simmering passions quiver just below the surface and require constant vigilance to keep 'em capped! Phew...No more Tendulkar and Bradman! [[ This is the other Srikanth, Abhi. Ananth: ]]

  • Abhi on May 25, 2011, 15:00 GMT

    Andrew Simoes, I did a vaguely similar study - but with just a few select batsmen. You get roughly the diff. (in % terms) working to vaguely the diff. in batting avg. Obviously nowhere near the range and depth of your analysis- since i simply dont have the stat and computer knowledge to go about it.

    So, does this simple correlation also work? i.e Diff. in Bqty in % terms = Diff. in batting avg in %?

    At least for the top batsmen?

  • Ravi on May 25, 2011, 14:17 GMT

    Ananth, sorry for digressing. Reading holy cow mentioned got me thinking. Do you/historians have records of how the Don was seen as his career progressed? He was self-coached and was once asked "Who do you think you are, Vic Trumper?" Fact: His avg. touched 99+ in his 7th test & stayed there till he retired. Outstanding. Period. However a few questions come to my mind: - Was the media then obsessed with his avg. like it is today? - Do we harp on his avg and frequency of big hundreds because his other milestones have been approached or bettered (Lara, Viru for big scores, by many for 100s, by 50-odd players for aggregate, by half a dozen for the highest score etc). - Was he ever the highest scorer in FC/Tests? - Was he seen as a match winner against strong teams? - Was he a good captain? It is vital to recognise the opposition- good Eng, avg SA, poor Ind,WI. Don't have answers to these & the subjective what-ifs. So welcome comments on facts and media reports. [[ I think our resident historian ShrikanthK is probably well suited to answer some of your queries. The quality of opposition was mixed. That is reflected by the rather high Career Bowling Quality index of 36.97. Ananth: ]]

  • arch on May 25, 2011, 14:10 GMT

    This is extensive and brilliant work, and excellent to see how the comments have refined this analyses. Looking forward to a similar analysis of bowlers. If anyone can do it, you can.

  • Aditya Nath Jha on May 25, 2011, 13:14 GMT

    we all know that pitches were uncovered earlier, that modern batsmen have helmets and other protection, that fielding standards have improved crazily and that the LBW laws have changed. We can only fantasize about how people would have played in different eras. We also know that style can't be quantified and neither can be charisma. We also know that Trumper will never figure highly in any list on this blog and that McCabe will be remembered for only 2 innings on this blog. But this blog is not the last word on cricket. It is "itfigures". There are other places to discuss talent - unless, you have a statistical model on talent as a function of x variables. [[ Aditya Many thanks. I could not have put it better. Your two comments and Andreq Simeos's comments have restored my faith in the blog. I was beginning to get the depressing deja vu feeling I had earlier when the pro-SRT comments, some bordering on the nasty and beyond, were pulling the whole blogspace down. This blog is not even the last word in cricket numbers. There are equal or better webspaces there. Once again many thanks for putting in words the right perspective. Ananth: ]]

  • Aditya Nath Jha on May 25, 2011, 13:09 GMT

    someone has to do this, so here goes. dear pallab - appreciate your passion and commitment towards all aspect of cricket. the name of this blog is itfigures. as the name suggests, this is a statistics based blog. ananth - and the readers - have invested a lot of time and effort in statistical analysis and discussion. the whole notion of talent is, to put it exaggeratedly, alien to this blog. talent cant be measured by itself and therefore it doesn't figure here. i am sure there are other equally eminent places where talent is discussed. what this bog does do, is to discuss the statistical manifestation of talent (or, lack of it). harvey, walters, chanderpaul, barrington, headley (the grandfather), botham, morris, tate, crowe..and what the hell, even ravi shashtri - hugely talented or untalented all - they have all been discussed here by people who are equally passionate as you. and equally well read and equally keen followers of the game. but statistically. CONTD

  • West Indian Follower on May 25, 2011, 13:03 GMT

    @ObelixThe RomanBasher Ultimately, if you look at that Aussie side, everyone who was playing had not started their careers in the 1960s, but early 70s, apart from Ian Cahppell. They were in no way a young side, but a seasoned professional one, unless you think that successful cricket careers come to an end after 10 years of play. The West Indies team of 1975 was young and unused to a tough standard of cricket that the Aussies presented. It took Packer to make them into the hardened successful unit. As for their attack, the fact that it performed and took wickets ALL OVER the world suggests that the apparent lack of balance was not an issue. You call my myopic, but you said that South Africa would thrash and strongest Windies side, for what reason? West Indies could match uo with South Africa in the batting department, maybe just missing the extra extraordinary batsman (Wi had Viv, SA had Pollock and Richards), but their bowling was definitely stronger.

  • James on May 25, 2011, 12:58 GMT

    @Ananth

    You can be a great player in a useless team.

    Post 2000, Brian Lara was the only great player for the West Indies; Chanderpaul, Gayle, and Sarwan were decent, and the rest were 'nobodies'. [[ Tendulkar/Dravid/Laxman/Atherton/Inzamam/Saqlain/Yousuf ??? Ananth: ]]

  • mohanlal on May 25, 2011, 12:55 GMT

    Sir, In my earlier comments i had put forward 3 instances of match situations where a captain over a series is prompted to give a more effective bowler against a particular batsman(Lara Vs Bickel for eg: ) more overs taking in mind the chances of getting the batsman out is more.This surely decreases the BQI of the concerned bowling group which the batsman is not at fault at all because it was his effective resisting of more potent bowlers in general (warne & mcgrath in lara's case) that the captain was forced to use the generally less potent but more effective option against this particular batsman more.Can you pls inform as to how this method effectively counter this situation? Eager to know that. [[ The BQI is determined by taking in the actual overs bowled in a particular innings. There is no change to this whatsoever. All situations you are referring to will come within this framework. Ananth: ]]

  • Andrew Simoes on May 25, 2011, 12:42 GMT

    Similarly , if the BQI was less than or greater by only upto 0.2 than the current benchmark , I would let the batsman into the list. Experimenting with different delta values simultaneously , I came up with two values ( -4 for batting average and +1 for BQI that yielded lists of exactly 28 ! - Different players of course in each list ) . Trying with -2 for batting average and +0.5 for BQI I was happy to see that one yielded 19 and the other 20. This firmly establshed the correlation that excepting Bradman from the space . 1 BQI = 4 Batting average approximately . Provides an accurate means for comparing batsmen . Who said stats don't tell the whole picture! [[ Andrew, this is something. I cannot say that I have understood all what you say. However I could try and make sense if you can send the Excel sheets to me. I will reply to you from my mailid so that you send me the same. This is so refreshing that someone does something positive after extracting the files after the huge number of pointless comparisons and putting down players. May your tribe flourish. Ananth: ]]

  • Andrew Simoes on May 25, 2011, 12:38 GMT

    I did this by choosing players who bettered the current one. For example if I'm moving down your list ordered ascending BQI , if I start of with the first guy ; K J Hughes , any player with a lower average than he will not be meritorious because he has also faced inferior bowling to Hughes! If I get someone with a better average , he replaces Hughes and becomes the new benchmark. The proceeding was Hughes, Atherton , Stewart , Thorpe , Richards etc etc ... Sobers , Bradman. If I used my other list and applied the same procedure only this time ordering by average and using lower BQI as benchmark , I obviously obtained the same list .

    Here's where the fun begins. This list was too exclusive, so I decided to allow delta errors to be included. For example in the first sheet , if the average is greater than or less by only upto 1 than the current benchmark I would allow the player into the list.

  • Andrew Simoes on May 25, 2011, 12:30 GMT

    Anantha, What a superb analysis. I can't really find faults with it ; as you are focusing on batting , a bowling average consideration with your two corrections suffices because discrepancies will get normailized over such a large sample space (People argue that Warne is better than Murali , and therefore you cannot consider bowling averages as Murali's is superior to Warne's and therefore a BQI involving Warne should be higher than one involving Murali etc etc) So intrigued was I by your discoveries that I saw the need to establish a good relation between BQI and Batting average. Excluding Bradman from the sample space (as he was skewing results) I made two Excel sheets from your Weighted Bowling Quality Table - the most decisive overall of your tables. One sheet was exactly like you have put it , the other , I ordered descending by batting average with Bradman at the top and Healy at the bottom. I then proceeded to descend down the lists selecting players who were meritorious.

  • James on May 25, 2011, 11:55 GMT

    @Ananth

    I never said any of those guys were nobodies, and I don't know how you came to that conclusion.

    The point being made was that there have been a lot of average, fundamentally flawed, and very incomplete cricket teams walking around the past 15 years, which is true.

    Some of these guys seem to think that cricket in the 90s and beyond has been incredibly strong and competitive, and that cricket in the 50s, 60s, 70s, and 80s, was useless, which is just absolute rubbish.

    The modern Indian team troubled Australia the most, but they did it with great batsmen, and an average bowling attack.

    If a guy like Agarkar or Prasad is opening the bowling for an International team, then that's a pretty good sign that the fast bowling stocks are fairly weak. And then you look at Kumble, who averages 37 in Aus, 41 in Eng, 40 in NZ, 42 in Pak, 44 in SL, 32 in SA, and 31 in WI, and you start to get the point.

    The standard of cricket in this era was not as strong as some seem to think. [[ You mentioned that SW's Aussies defeated teams of nobodys. These are the players in these teams. So how could these teams be what you have described. Pl use proper terminology. Ananth: ]]

  • shankar on May 25, 2011, 11:49 GMT

    Hi Ananth,

    sorry for asking again. The uploaded file for all batsmen (batsmen table by group) has only 1200+ rows and data for 261 batsman. Not sure if I got this wrong. Thanks. [[ This is not for all batsmen but for those who have scored 2000 runs and above. Ananth: ]]

  • Pallab on May 25, 2011, 11:17 GMT

    And I had an occasional blog on CNN-IBN website (cricketnext.com) in 2008 without ridiculing any cricketer. Being “real cricket” followers are not the preserve of a mere/chosen few or the many millions who revere Bradman. Let’s not take a judgment call on that. This is not directed at you as I have been mentioning privately to you about your measured and thoughtful ripostes/remarks on issues (Gayle/Pollard mercenary conundrum) and players.

  • Pallab on May 25, 2011, 11:11 GMT

    "My advice is "do not respond at all". Nothing will be gained ... I think the real cricket follower will know the true facts." Ananth: What exactly are you wary about even if someone responds to ONLY the Talent “parameter” which I have raised? And I have pointed out facts too about innings, conditions and the like. And why would you preclude others from debating as this BLOG article is a dissection of batsmen. Do the issues raised sound hare-brained or outlandish? I have long learnt to critique even world history (and the distortions) as we have been told and taught. Just for the record, I gave up a career between 1995-2007 (school and college obviously did not allow fulltime watching from 1979 onwards) to watch Test cricket FULL-TIME (till my last day in India) without interruptions and managed my independent assignments AROUND Test schedules.

  • Pallab on May 25, 2011, 10:49 GMT

    "No media person, no commentator, no analyst, no cricket follower can utter one word against Tendulkar" ANANTH: Disagree. I have done so: maybe not in this blog per se, but on other cricinfo forums. I can bring up those published comments. But I have dispassionately also pointed out SRT’s almost matchless talent without necessarily talking about the 100 100s which I too feel is being hyped by the media. In your blogs, I was the ONLY person to point out that India won 12 Tests (beginning with Aus’s real domination period in 1999 versus Aus 10 -@BOLL subtly ignores that). I was the ONLY person to point out that Imran’s Pak teams never lost to WI in Tests and dented their absolute dominance. And then I went ahead and REASONED why Don might not be the “holy cow” of statistics as he is being made out to be. I know you might not have the time to read other forums, but SRT is/has been criticized vociferously and quite VICIOUSLY (on many parameters) on many online forums for the past 5 years.

  • Pallab on May 25, 2011, 10:39 GMT

    @BOLL: Unlike “Sheeple” (people –even those wise enough- who will be blind enuf to accept anything at face value) -not directed at you as I have already classified you, I am perhaps the first commenter who has raised the “talent” PARAMETER and not records while critiquing Bradman and simply asked whether his talent at playing particular strokes or technical proficiency at tackling particular bowling types -pace/spin- was X or 1000 times better than peers/other acknowledged greats.And I had also mentioned in my disclaimer posting to Ananth (not published then) that SRT is NOT my favorite Test batsman.(Though his supreme and peerless young age talent is embedded in my consciousness as it is in the millions of cricket-loving Indians –MIND YOU, not those mythical billion fans!).

  • Pallab on May 25, 2011, 10:32 GMT

    “By your extremely harsh standards Pallab, I believe The Don may be the only man in history to have `passed ` more times than he has `failed`” I had forgotten to mention my shock at Pallab`s contention that any score less than 50 was a `failure`, and that Don had therefore failed on 38 occasions. BOLL: Why are you shocked as I am not imputing harsh standards but pointing out the fallacy of your blanket yet disingenuous statement about “there being daylight between Bradman and other batsmen” . If there is daylight, how has he failed in 38 out of 80 innings? And yes, popular PERCEPTION is that Don never failed. Even otherwise he has just “passed” in 42 innings (100s and 50s) hardly that BIG a difference between pass and FAIL. (and NOT as you exaggeratedly SAY “may be the only man in history to have `passed `more times than he has `failed`. I am talking abouut the innings and not the humungous scores per se).

  • AD on May 25, 2011, 10:19 GMT

    Ananth: Yes,I have received your file. Please use the above email address for future private correspondence. I will also be sending a mail to you from the above ID.

  • Pallab on May 25, 2011, 10:17 GMT

    "Is it not the easier solution to accept that Bradman was quite simply, and by a fair stretch, the best player to pick up a bat?"

    BOLL: Now that Ananth has published my earlier reasoned out points about Don (PLEASE READ THEM), I have precisely raised the point made by you:” the best player to pick up a bat”. This is a qualitative statement made by you which is subjective. About TALENT. Bradman’s records on Average and 100 to 200 conversions criteria are INDISPUTABLE but not Talent as my post has pointed out.

  • Pallab on May 25, 2011, 10:11 GMT

    “28,000 first class runs, 117 centuries in 338 innings, at an average of 95. At the age of 22, both the first class (452*) an test (334) high scores.” @BOLL: You have this “quirky” habit of buffeting records and stats with non-Test figures of players whom you favor or when you want to state points and then pass them off as nuanced, subtly sophisticated points (as you have also been doing while overstating Waugh/Ponting’s Aussies over Lloyd’s WI). I have been telling you repeatedly to please add the prefix “Unofficial” to the SuperTests records of G. Chappell and Lillee and now you are showcasing Don’s FC records (which I know about). Let me point out the record of an Indian colossus Vijay Merchant who averaged 98 in India’s first class championship. But I never raise these figures while pointing out Merchant’s small Test playing record. Please STOP pointing out FC figures.

  • Pallab on May 25, 2011, 10:07 GMT

    “The lessons are there in front of us and Rahul. Kapil and Gavaskar.” Ananth: Kapil was actually forced to retire in 1994 unlike Gavaskar who called on his own retirement (though Chandu Borde tried to ease him out of ODIs in 1986). I have it on good stead that Azhar ousted Kapil and put him on notice after passing Hadlee’s record (having been an obsessive hound about power plays in Indian and Pak cricket for decades, I kept track. Have a lot of inside dope of the match-fixing era before Hansie’s confession in 2000 and genesis of the Azhar-SRT feud too) almost to pay back Kapil in kind who was planning/"plotting" a putsch on Azhar after the failed SA tour in 1992 - but which coach Wadekar nipped in the bud.

  • Pallab on May 25, 2011, 10:03 GMT

    “That, Abhi, is DRAVID’S PROBLEM. If he ends his career with a sub-50 average..But he should retire gracefully.. The lessons are there in front of us and Rahul.” Ananth: Glad that you have been belling the Dravid “cat” for a while. Dravid has failed against top-rated /ranked opposition for the past 4 years (SA home and away -3 series, Eng away, OZ home and away -3 series, SL away twice). He will again score runs against the average WI attack before coming a cropper against Eng. This has been his pattern: score against NZ, fail against SA….so on. He is hurting his legacy. Ironically, India’s surfeit of cricket engagements(or luckily IPL) has made Dravid’s extremely poor Test returns in recent years (in direct comparison to his career record and that prolific run between 2001-2006) go under the analytic radar in the Indian media. I mentioned last year that he is CLEARLY blocking Badrinath’s re-entry into Tests (after Badri’s promising 50 against Steyn, Morkel in debut Test). [[ I strongly feel Dravid should announce his retirement tomorrow. He has earned a phenomenal amount of goodwill and acclaim. He has a 1.2 million dollars retirement fund signed up. He could command a premium in the mediaspace with his ariculate skills. What does he want to prove and to whom. Certainly one of the all-time greats. Ultimately if he has the bad run Abhi has talked about, that would stay on, like Kapil's huffing and puffing to the 433rd wicket. Ananth: ]]

  • Sreekanth on May 25, 2011, 8:32 GMT

    Ananth,

    First off, thanks a lot for this blog which has been a constant source of enjoyment and illumination for amateur cricket historians such as myself.

    But lately, there has been a tendency for rather tedious and occasionally inane arguments to pop up. Couldn't you use your discretion to vet out the comments which you know for certain, arent really adding to the discussion but are merely meant to provoke responses? [[ You are right. I did this fairly strictly for over two years. Somehow these things slip in and once the floodgates open the comments and counter-comments increase in GP. Ultimately it is my respobnsibility. Ananth: ]]

  • Alex on May 25, 2011, 7:17 GMT

    @Abhi: Incidentally, I think RD is in his final year. SMG has been advising him the past 2 years to enjoy himself on the field, fully knowing that he had served India with great distinction. That is a sound advice ... SMG walked that talk in his own final 2-3 years with really nice results. I guess Dravid is such an intense player that he probably is not able to implement this advice. It is a pity. He was such a great batsman as well as a great player and deserves to go out better.

    It would be if the media showers some affection on him now. The press going ga-ga over SRT's 20th year international year and completely ignoring RD's 12,000-th run in the same A'bad test is a total morale crusher. Unfortunately, Dravid has been dealing with that for almost all his career.

  • Alex on May 25, 2011, 7:07 GMT

    @shrikanthk: Sutcliffe was not rated a true great probably because he was an effective but ugly batsman (like a Chanderpaul perhaps). Contemporary reports say that Sutcliffe himself knew that he was not a great batsman and tried to make up for it by scoring well.

    @Abhi and @Ananth: Abhi sort of expects every analysis to rank the batsmen on how good they "really" were. He has produced a WW2-marred data-set in which the Don played only 23 innings in a decade ... now a days, most batsmen play 23 innings in a single year. SRT himself has said that the Don was beyond everybody and Abhi should accept his hero's verdict as the unadulterated truth.

    I swear by SRT but cannot rate him as an all-time Top 5 batsman without having second thoughts. Much less separate him from Sobers/Viv/Hobbs who played in other eras, I cannot separate him from even Lara. To me, these are undoubtedly the two best modern batsmen and are best ranked together as the single entity "SRT-Lara".

  • bks123 on May 25, 2011, 6:59 GMT

    I am tired with the Bradman/Tendulkar discussion everywhere in cricinfo. Spare this nice blog and leave that for other articles in cricinfo. Bradman was way ahead of his generation. And that's a fact. Every sport evolve with time. Today's cricket is a thoroughly professional outfit. The most improved part of cricket in today's world is fielding (except for pak). Today fielding influences the game a lot. So many times we have seen a tendulkar or a lara getting out in 20 or 30 by an excellent catch by someone like jhonty rhodes. Personally, I believe, sachin is not far ahead of his contemporaries especially in tests. I would rate both lara and dravid ahead of sachin in tests till 2008. I also rate viv richards higher than sachin in ODIs. In fact many others like ponting, heyden, ganguly will be very close to sachin in ODI performances. But if you consider all formats combined, sachin is the best and there is no doubt about it. So, Bradman vs Sachin is like Maradona vs Messi comparison.

  • shankar on May 25, 2011, 6:22 GMT

    Amazing analysis Ananth. Thanks for your hard work to provide this insight.

    I had downloaded the files per your links. However, in the Batsman table - by Group (for all 2000+ batsmen): (testgrp4.txt)I could find only 261 batsman. Just thought will check with you. Thanks.

  • AD on May 25, 2011, 6:00 GMT

    Regarding the DGB/SRT debate,everyone familiar with cricket is well aware of the figures and 99.94. So the truly amazing part is that some ex cricketers have even dared to venture that SRT may be better than DGB.To date anyone who was silly enough to do so was immediately branded a heretic with scant knowledge or regard of the game.

  • Pallab on May 25, 2011, 5:57 GMT

    conversion rate to 200s criterion. But let’s talk talent wise now and DISPASSIONATELY. I mean how much MORE talented was he than a Tendulkar, a Lara, a Sobers, a Sehwag, a Richards? Was he the foremost technician of his era ?(like Hobbs/Hutton were of their eras and Gavaskar of a later era). Did he play the straight drive 1000 TIMES better than a Gavaskar? Did he hook 1000 TIMES better than Richards? Did he score faster than Sehwag? All in the negative judging from first person accounts and written histories. Just the 29 100 record in 52 Tests is not all that great as it is made out to be as Gavaskar scored 22 100s in 56 Tests Tests (hardly all that far behind). I want to point out to @ BOLL and others that SRT was ALREADY a FINISHED Test product at 16! -which Bradman (and other greats) could only dream of till he turned 20-21. [[ A request to Boll (and others who might want to respond). My advice is "do not respond at all". Nothing will be gained and we will only be taking the route which occured couple of years back. I think the real cricket follower will know the true facts. I am also going to publish these without any response. Ananth: ]]

  • Pallab on May 25, 2011, 5:55 GMT

    another BENCHMARKING PARAMETER). And 100 international 100s (almost a formality now) should COUNT for a lot. I remember the hype about Zaheer Abbas getting to 100 first class 100s and SRT now goes ahead and reels off 100 international 100s!. And let @SHRI not turn around and talk about importance of FC 100s vis-à-vis ODI 100s as SRT is/was well capable of hitting 100 100s in FC alone if he had played that many FC matches instead of just ODIs. @BOLL The only 2 UNIQUE superhuman qualities which distinguished Bradman from all batsmen in all cricket history are his extreme powers of concentration when making those those humungous BIG SCORES and his insatiable appetite for massive 100s (doubles, triples and the like).I FAIL to understand how Bradman is called the greatest batsman (I meant talentwise) of all time SIMPLY because of his awesomely superior average. He is INARGUABLY the statistically greatest superior batsman of all time on the average criterion and possibly the century CONTD.

  • shrikanthk on May 25, 2011, 5:55 GMT

    What if Tendulkar plays for another decade and ends up avg.45? nevermind the thousands of runs.

    Good point. This is precisely the reason why I ignore people who claim that WG isn't a big deal since he averaged only 39 in First class cricket (and 32 in Tests).

    What they forget is that this guy WG played his first FC game in 1864 and the last game in 1908. That's 44 years of cricket.

    No wonder he averaged that low. Which is why, we need to judge WG by his record in his youth (1860s and 70s) when he was MILES ahead of everyone else.

  • AD on May 25, 2011, 5:54 GMT

    Ananth: Managed to correctly transfer the data.Thank you. Would it be possible to put up a list of variables and their weightages that you intend to use to calculate Pitch Quality? [[ I hope you have received the document I have sent to you for comments on the topic. Ananth: ]]

  • Pallab on May 25, 2011, 5:54 GMT

    “daylight” between Bradman and other players which is why I brought up the issue of talent and not just Bradman’s awesome records. @BOLL, Others Bradman has actually “failed” (yes a FACT) in 38 innings (out of 80) where he has scored less than 50 –with actually 30 innings a lot less than 50 (WHICH is a LOT as the general PERCEPTION again is that Bradman NEVER failed in Tests). BOLL should accept this as he has mentioned that there is daylight between Bradman and other great Test batsmen and that he wants a 30% differential between one great (future or current) batsman and his peers in his generation or across eras to be considered an equal of Bradman (BOLL used Don’s average as the BENCHMARKING PARAMETER for the 30% differential. Let’s not bring in the 12 200s as Sehwag has a superb 14 150 plus scores as a Test opener! and BIG 200 scores). Well Tendulkar (SRT) has that now with both his Test and ODIs 100 hauls almost a clear 30% better than his peers (using a century as CONTD

  • Pallab on May 25, 2011, 5:50 GMT

    I did not take issue with your quite “cavalier” summarization of my SRT-Bradman comments.This has given some of the posters rebutting what you have summarized that I have not reasoned out my argument rationally.(I am possibly one of the most unbiased poster on cricinfo blogs and have critiqued “holy cows” SRT, Dravid, Sehwag and the like.)I have pointed about SRT’s zeal to accumulate certain records on cricinfo. Also, at no point, did I mention that SRT was far ahead of Don;I just pointed out that matched with SRT’s supreme acknowledged talent and his awesome records (please note I am also not swayed by the 100s),the title of “greatest” should now pass onto him. But now after you have again replied to BOLL” I am glad that I posted only a summary of Pallabh's comments.”, I request you to publish my next 3 comments which I have summarized from the 7 earlier mailed to you. I had just taken ISSUE with BOLL using a blanket statement about there being CONTD.

  • shrikanthk on May 25, 2011, 4:53 GMT

    So,this question of "average" is actually very loaded... As per this metric there are numerous batsmen “better” than say Lara of Viv. Now who is going to accept that?

    Agree. That's precisely why it is essential to go beyond stats once in a while and read a bit of history to form a subjective judgment.

    To form an informed subjective judgment especially on pre WWII players, it may even be essential to think beyond Tests and consider FC stats. It may also be essential to consider the batsman's influence on the game (technique among other things) and on cricket in his country.

    Keeping all this in mind, if I were to make a shortlist of the greatest batsmen of the 20th century, I'd come up with the following names -

    Ranji, Hobbs, Bradman, Headley, Sobers, G.Chappell, Richards, Tendulkar, Lara.

    You may add or subtract a name or two.

    Within this list, Bradman (and to some extent Sobers and Sachin) stand out. I don't really care how you rank the rest.

  • Abhi on May 25, 2011, 4:05 GMT

    Srikanthk In another recent blog on the Tendulkar/Bradman thing. Bradman's avg. is 99 but he has actually never got out in the 90s in his Test career- not once.

    Alex also raised a very tantalising point- the delicate nature of a well timed retirement. If Dravid goes on for another couple of years it is possible he may end up with a Sub 50 avg. But is he "really" a sub 50 avg player? [[ That, Abhi, is HIS PROBLEM. If he ends his career with a sub-50 average, then he has played probably 10+ more tests at an average of 30. Why should he ever do that. To further his IPL career. Don't get me wrong. I am not saying he does that. But he should retire gracefully if he knows in his mind that he would fail in 3 out of 4 tests. The lessons are there in front of us and Rahul. Kapil and Gavaskar. Ananth: ]]

    Relatively few players sign off in a blaze of glory so enhancing their reputations and leaving us pining for more. The vast majority slink into the slow vortex.

    What if Tendulkar plays for another decade and ends up avg.45? nevermind the thousands of runs. So, his "overall" average wouldn't be too pretty. Isn't it a fact that injuries can hammer stats? [[ That, Abhi, is HIS (a different 'his') PROBLEM. If he ends his career with a 45 average, then he has played probably 50 more tests at an average of 25. Why should he ever do that. To further his IPL career. Don't get me wrong. I am not saying he does that. But he should retire gracefully if he knows in his mind that he would fail in 5 out of 6 tests (to reduce his average by 20%). Ananth: ]]

    Would Bradman have avg.99 over 100 more inn, plus ODIs. Rem. Hussey avg.80+ too for 35 inn. [[ Why should you pose this question, knowing that it cannot be answered and knowing that ODIs came in 1971 and knowling that that players travelled 3 months on ship and knowing that there was no way anyone could have played 100 Tests in those times and knowing that Bradman, at his peak, missed 5 years and knowing that his post-WW2 figures, after the age of 38 were 1900 runs at 105+ . I will ask you this question. Would Tendulkar (or any modern great) have averaged 99.94 over 52 Tests if they had played in the 1930s. Again I would never ask this question myself. But you have made me do it. If you say "Yes", you can send all comments what you want and I will publish those with no contribution from me.. Ananth: ]]

    So,this question of "average" is actually very loaded,and one has to answer a lot of questions before we can just trump it around as an end all. As per this metric there are numerous batsmen “better” than say Lara of Viv. Now who is going to accept that?

  • shrikanthk on May 25, 2011, 4:04 GMT

    I've always wondered WHY Sutcliffe and Barrington (despite their humongous averages) are not discussed often enough when we talk about "all-time greats".

    Ananth's analysis now tells me why! Fascinating insight into those two.

    Sutcliffe was a product of the 20s, who made utmost use of the old LBW law.

    Barrington was something similar. Very effective batsman against slow bowlers on the subcontinent. But perhaps not as dominant against quick bowlers.

  • shrikanthk on May 25, 2011, 3:44 GMT

    Finally, I must say this- The question - "What would player X average in the current era" is a fundamentally flawed question. It's the sort of debate which doesn't befit this blog. [[ I agree with this. This cannot ever be answered. The final number will be what the individual wants. He will use required stats to get to either 70 or 85 or 105 for Bradman's average today. A better method would be to use the collective brains of a set of free-thinkers, as I think has been done successfully in this webspace, to do an incremental adjustment to all such figures based on available stats, such as Period index. In that case all figures have to be adjusted. However the most effective method of eveluating an individual's performance is by comparing Peer performances. Ananth: ]]

    It's the equivalent of asking - "Will Adam Smith be able to compete in an Economics PhD program today, given all the emphasis on math in modern economics?"

    OR

    "Can Kautilya, the ancient Indian political strategist, be an effective and influential king-maker in modern Indian coalition politics?"

    These questions are so very pointless and inane, and also disrespectful.

  • Abhi on May 25, 2011, 3:32 GMT

    Srikanthk Actually 99% of stats,arguments etc. aren't mine..They are all over the net in numerous blogs and forums. I merely copy/paste.

  • shrikanthk on May 25, 2011, 3:18 GMT

    Just had to comment on this.

    Abhi: You work very hard indeed to prop up Tendulkar's case! I appreciate the commitment.

    Now, you cite all kinds of possible reasons why the average of 99 might come down very significantly in the current era. I may agree with some of your points. But my question is - If you were to do that, you'll have to discount EVERYBODY's average. Not just Bradman. Go on and say that Hammond's avg will be 30.xx, Hobbs' avg 35.xx in the current era. After all those guys faced weaker attacks than Bradman (as per Ananth). But then nobody ever does that. It's always poor Bradman who bears scrutiny. Let's just accept that Bradman was WAY ahead of his time, unlike Sachin.

    Vs England Headley avg. :71.2 Don avg : 89.8

    Firstly, Headley faced slightly weaker English attacks, unlike Bradman who was mostly up against full-strength English attacks.

    FYI: Headley never faced Tate, Larwood and Bedser in Tests. 3 of the best seam bowlers during his career.

  • Abhi on May 25, 2011, 2:30 GMT

    Ananth, Probably right about the O'Reilly comment. My old (bad) habit is to type first, send.... then Think!

    And then even if there was something a bit off colour it is human tendency to try to justify yourself!

    So,Boll,Srikanthk...and anyone else...Sorry Guys!!! [[ That is the Abhi I like. Owning up to a mistake is hard and I salute you for that. O'Reilly turned out to be the Australian cow. Ananth: ]]

  • Abhi on May 25, 2011, 2:18 GMT

    And Why, Oh why, did they drop Chanderpaul? It's like dropping Dravid. All that flashy Carib strokeplay will lose you more matches than you will win without some Steel in there.

  • Abhi on May 25, 2011, 2:06 GMT

    Boll, And you are “still smarting” at the O’Reilly comment?! Partly because he was Chief guest at a prize distribution?. You should then at least have a feel for how irritated some fans get when the hear the anti Tendulkar garbage. Here is a guy who has given innumerable ppl pure joy and kept their hearts thumping for over two decades because of purely one thing and one thing alone- the pursuit of excellence. Just his bat- no bull talk, no swagger, no nonsense, no nothing. Just him and his bat. [[ The big difference, Abhi, is the way you put down an all-time great bowler as no good and the others, including me and Boll, have never spoken one word against Tendulkar. All of us have only expressed our admiration at Tendulkar. We were only trying to justifiably put him in second spot. And that does not wash in India now. According to you that is garbage. If you think that is garbage, your own credibility is at stake now. Every possible statistical element is gleaned to bring down Bradman and push up Tendulkar. This defies description. As an Indian, I would say, the holiest of holy cows in India is Tendulkar. No media person, no commentator, no analyst, no cricket follower can utter one word against him. And in this blogspace, where no one ever utters a negative word about Tendulkar, I cannot put him in second spot. You yourself have changed. Once you commented that you accept that Bradman is the best but cannot see anyone else in second place other than SRT. At that time you made a lot of cricketing sense and did not allow your adulation to cloud your thinking. Not now. And I suggest you re-read my response to Boll's comment. Where I myself have placed SRT. You will then realize what I am trying to say. Ananth: ]]

    As rgds.5% avg. differences etc in the past couple of years there have been numerous stats and analyses on Bradman vs Tendulkar. An eg. Several batsmen avg. 70+ for periods of upto a decade in Bradman’s time. Headley avg. around 70 too, before his ill lfated and ill advised comeback battered his stats. And of course Headley had to face the Aus bowlers too…..against whom his avg. too got further battered (37) Vs England Headley avg. :71.2 Don avg : 89.8 (Again before his comeback Headley was avg. 78 vs. England) The minnows then were arguably India and SA who Bradman milked for an avg. of 179 and 201 resp. So, essentially ,Bradman’s claim to fame is a sllightly better avg. than Headley vs. England …for longer. The case against Tendulkar is that he apparently only has durability. Much the same could be argued about Bradman. Hussey avg. 80 over his first 35 inn. Bradman did it for “longer” For the ’90s (again EXcluding Ban and Zim) for players who have played THROUGHOUT the decade: SRT : avg. 59.4 Steve: 51.8 Lara: avg. 51.6 During Bradman’s best 10 yr stretch 1938 to retirement he avg. 105.7 over 23 inn. A.Morris avg. 74.1 over a similar 22 inn. If you ignore the “minnows” then (India) whom Bradman ravaged at an avg of 178.8, you get an avg. of 84.9 vs. England in 17 inn. Arthur Morris avg.79.9 vs England in 17 inn. too (his overall avg. drops because of a 52 avg. vs India) SO- Bradman’s Best 10 yr stretch was actually 84.9 vs. 79.9 (if you use the above context) over the “Next” best A difference of 6.3% Tendulkar’s 15 % over the next best for his best decade. The Million Dollar question- Throw in a completely different format, hundreds of matches….Does Bradman maintain those standards through inevitable injuries ,matches etc. There are also MANY other fascinating comments re. LBW laws etc( I think on one of those Brit blogs)…On the surface Bradman’s stats blow away everyone- dig a little deeper and you glean some new info.

  • craigmnz on May 25, 2011, 1:54 GMT

    The figures for Hammond and Sutcliffe point out the impact of cricketing politics on performances. After WWI England and the MCC had a clear policy of fostering cricket abroad (at least in their terms) as evidenced by tours to West Indies, New Zealand and India including test matches and reciprocal tours by those sides to England. New Zealand first toured Australia with test matches in 1973 (after we'd toured the West Indies and India (the latter 3 times). Australia played a single test in NZ between 1930 and 1973. Now I know the Tasman Sea is not exactly small but clearly the Australians method of fostering cricket was quite different to that of the English - we saw numerous Australia B sides over the years. In discussing Hammond I'd suggest that a lesser performance against the newcomers such as NZ would be even more noteworthy. [[ Craig, all these discussions are never with the intention to put down NZL cricket, which had a peculiar history before 1950 when they settled down as a Cricketing nation. Ananth: ]]

  • Abhi on May 25, 2011, 1:44 GMT

    Ananth, Looks like there are more holy cows than initially thought. Re.Tendulkar- the 100s and Thousands of runs are undoubtedly also due to longevity. But the true beauty of it is that inspite of all the runs,hundreds and injuries he still averages as good as any modern day player. Except for the few years in the mid 2000s ,where he almost crashed and burned, he has been among the top few batsmen for most of his career.

    It is somewhat like saying that Federer has 16 slams partly because of longevity. [[ I knew you would bring this in. A totally wrong comparison. There is nothing else in Tennis. Everything is measured in the number of Grand slams. There is nothing like a Slams won % to participated in. A GS won is a GS won, it is the final thing. Any player,who is good enough to play 100 Tests, as a batsman, is going to score a minimum of 5000 runs. A better batsman, 6000 runs. A good batsman, 7000 runs. A very good batsman, 8000 runs, An outstanding batsman, 9000 runs. One in a million batsman, 10000 runs. On the other hand, a Tennis player, good enough to be in the top-20, say, Ferrer or even Murray, could play 100 Grand Slams and WIN NO GRAND SLAM. The truly greatest would probably win 15-20. So, in Cricket, you play at your level and accumulate runs at your competence level. How long and how effectively is what matters. Please stop this matter of comparing two totally different sports, just to prove your point. It does not wash. Ananth: ]] I have mentioned several times that the problem with cricket relative to other sports like Tennis is the "assymetry" of schedules. Ponting and others were very lucky. They scored 40% or more of their career runs in a 4 yr mid 2000s period when in top form. In a mere 4 yrs they “caught up” .This is somewhat like suddenly scheduling many more slams and Djokovic winning 4 slams out of say 8 for the next 4 yrs-hence "catching up" with Fed...Luckily in Tennis ,longevity( in terms of number of slams)directly translates into Excellence- Because of a Fixed schedule. Langer in his recent book said getting a hundred is like a holy grail for a batsman.Ponting mentioned how hard it was to get one international hundred- never mind 100. It’s not like you just turn up in international cricket and the bowlers give you right of way. [[ It is my personal feeling on 100s. You may feel differently. We have our individual rights to think in our own ways. For me a 100 is the 100th run, no more or no less. What matters is what it did to the team. Tendulkar's awful 85 in the WC SF was, in my opinion, more important than all his centuries, especially the 200 and the most significant innings he has ever played. But for that innings, one of the other Asian teams would have won the WC. This is not to put down Himalayan achievements. The 200 and 400 were milestones, never achieved before and would remain like May 1953, until they are crossed, if ever. They may never be crossed. Give importance to such momentous achievements. A wonderful century, on the other hand, might end up in a losing team. For that matter, many a century. Again a personal request. Please do not bring in Tennis or Golf into this discussion. Ananth: ]]

  • James on May 25, 2011, 0:03 GMT

    @Boll

    I looked in the dictionary for the meaning of 'nobody', and I saw a picture of Mike Atherton's English team, Steven Flemming's New Zealand team, the Indian team that had Ajit Agarkar and Venkatash Prasad as their opening bowlers, the Pakistan team that was involved in a new controversy every five minutes, the West Indian team that has become a joke, and the South African team that would choke on its own cereal.

    Happy?

    Ironically, these are the same teams that Australia's overrated side would occasionally lose a series to..... [[ I think you are crossing the line, if you have not already done so. In your anxiety to pull down the Australian teams you are putting down other teams. That is not on. It is one thing to talk of bowling quality, strength or batting quality. Another thing to put down Lara, Walsh, Ambrose, Tendulkar, Dravid, Laxman, Kumble, Atherton, Stewart, Fleming, Inzamam, Yousuf, Anwar, Saqlain as a bunch of nobodies. Ananth: ]]

  • ObelixThe RomanBasher on May 24, 2011, 18:55 GMT

    Well we will have to differ in thinking. I see the 1975 win is attributed to WI being a young side, but the 1979 win by the WI is not due to an aging Australian outfit !? What sort of double standard is that? I also see Barry Richards "only" averaged 50 in England, so what did Viv Richards do ? I guess some peoples runs carry more credibility. Very myopic views.

  • Abhi on May 24, 2011, 18:19 GMT

    Hey Alex, Did what you said and visited the family doc- He said "Wha... But you are absolutely right!" I said that someone may have suggested that something may be wrong with me. So,he said "Hmm, may be we should visit the local shrink" And so we went... And even the shrink agreed! We're thinking of hitting the ortho,dental and cardio chaps next week.

  • Boll on May 24, 2011, 17:31 GMT

    Still smarting over the derogatory comments about the great Tiger O`Reilly (products of the same Goulburn school, and the man who presented me with the best U/14 cricketer award in 1986 - been wanting to say that for a long time!!!). I will quote Wisden - He was `emphatically, one of the greatest bowlers of all time`. Emphatically. Amen. Ananth

  • Boll on May 24, 2011, 17:16 GMT

    @Alex, but surely an oppositon captain of the stature of Hammond can be taken at his word. re. Gower vs Richards, I remember the sounds of the same trumpets. Gower was a beautiful player, but you`re right, it simply wasn`t true.

    Your comments re.The Don, very brave! Hammond's high average and the bowling attacks off which the bulk of runs were scored might be dicey. But that should not take away from his captaincy related achievements and the ability to judge and appreciate a great opposition player. Ananth

  • Boll on May 24, 2011, 16:48 GMT

    Ananth, thanks as always for your thoughtful and measured comments to posts. I had forgotten to mention my shock at Pallab`s contention that any score less than 50 was a `failure`, and that Bradman had therefore failed on 38 occasions.

    All those hard earned 30s and 40s I scored at school,( not to mention the 10s and 20s) were for nothing! I languish in the very low single digits on that measure.

    By your extremely harsh standards Pallab, I believe The Don may be the only man in history to have `passed ` more times than he has `failed`. Maybe that`s cricket.

  • Alex on May 24, 2011, 16:36 GMT

    @Boll: Any serious cricket follower who puts Don the batsman second to anyone needs to schedule an urgent appointment with his family doctor with a vague complaint that "something is wrong with me".

    Anyway, best not to cite Hammond's testimonial to qualify the Don's credentials ... this article has shot Hammond down to earth. I never thought Hammond was in the Hobbs-Hutton league but this article suggests that he probably was no better than even, say, David Gower. I clearly recall the pom press trumpeting Gower to be better than Viv Richards in early-mid 80's. That is the time-honored British propaganda and them calling Hammond the Don's near equal back in 30's was almost certainly no different.

  • Boll on May 24, 2011, 16:23 GMT

    @James. Steve Waugh`s team beat `nobody`? I suggest you check the dictionary on that one. And just in case `nobody`s` mentioned it before - JUST BECAUSE YOU SHOUT REALLY LOUD DOESN`T MEAN YOU`RE RIGHT!!

  • Boll on May 24, 2011, 16:09 GMT

    Cont`d. I certainly don`t wish to denigrate any of the great batsmen I`ve had the privilege of watching; IVA Richards, G.Chappell, Sachin, Lara, Ponting...but the fact remains that these, and greats of previous eras, averaged 50-60 runs per innings. Has any of these greats averaged 5% more than his closest peer?

    Is it not the easier solution to accept that Bradman was quite simply, and by a fair stretch, the best player to pick up a bat? Or in the words of Walter Hammond `If I were choosing a side out of all the cricketers who have ever lived, I would put Bradman`s name down first. None of us had the measure of him and that`s the plain fact.`

  • Alex on May 24, 2011, 15:26 GMT

    @Arjun: I think the # groups and the thresholds which separate those is a matter of taste ... the ranking of batsmen in each group depends on the threshold values. In general, too large a number of groups is counterproductive since the sample space will be too small to derive meaningful statistical conclusions. Of course, 5 groups is OK on this analysis.

  • Boll on May 24, 2011, 15:15 GMT

    @Pallab. I must admit to experiencing some delight at being branded a fan/connoisseur. Not so sure about the `blind-sided` qualifier however.

    With regards `holy-cows`, particularly on cricinfo, I would suggest that Sachin fits the bill rather better than any Australian, be it The Don (Sir), Shane Warne, or indeed the oft-criticised Ponting.

    re. the myths you purport to be busting; a century in every 2.75 test innings is so far ahead of ahead of the nearest rival that comparisons become futile.

    Indeed, while perhaps not being the most graceful of players, he was the most rapacious and clinical run-getter of them all.

    28,000 first class runs, 117 centuries in 338 innings, at an average of 95. At the age of 22, both the first class (452*) an test (334) high scores. Boll, I am glad that I posted only a summary of Pallabh's comments. What you say is true. The holy cow in India is SRT. I myself consider SRT as the greatest, considering all forms of the game. But I would be the first to admit that certainly Bradman, and maybe one other batsman might be ahead of SRT in Tests cricket.Again another person might say only Bradman. I certainly can live with that. In ODIs, there is no way SRT can be conclusively considered as no.1 because of Richards' stupendous achievements. Again I can live with either of them anointed as no.1. But not anyone else. The problem is the hype that is made with the 100x100s. At the end of the day, if someone plays 170+ Tests and 450+ ODIs, and he is as outstanding as SRT, he is bound to get these 100x100s. This does not mean that someone could do it. No one has and no one is likely to. It is an Everest-like achievement. But no one should forget that that is at least partly related to longevity. This is not a record but a landmark. It will not be overhauled in my lifetime or your lifetime or your children's lifetime. One guy commented recently asking me how dare I say that Tendulkar was second to Bradman when he had scored 4 times the number of 100s. There was no way I could communicate to him of my lack of enthusiasm for a 100 per se, but could not also say that one guy played 52 internationals and the other guy 630 (granted quite a few of the one-innings variety). Unfortunately people are unwilling to perceive that the Average is a performance related measure. Runs are only partly so. 100s certainly are not. Ananth

  • Abhi on May 24, 2011, 14:24 GMT

    Also, these Bqtys are indeed starting to make more “sense. Lara and Tendulkar are virtual contemporaries. Their Bqty faced should be more or less similar, give and take a bit. And indeed Lara’s Bqty faced workd out to be about 5% better than Tendulkar’s (Undoubtedly due to the Bang. Matches).

  • Abhi on May 24, 2011, 14:15 GMT

    Ananth, Am totally fascinated by Arjuns new method of figuring out BQty. Would it be possible to put up a list of say the Top 50 bowlers with their "new avg" (calculated as per this new method) with their "traditional" averages side by side? This is not a new bowling average. It is only a method of consolidating such averages across a team, innings or career. The averages still remain the CTD values. Ananth

  • West Indies Follower on May 24, 2011, 14:10 GMT

    @Obelix. Clearly your comments were designed to elicit some angry response from me. . West Indies, whether in Packer, or normail cricket, between 1976 and 1982 beat a full strength Australia team, or drew with them IN AUSTRALIA. South Africa have never done that, only doing so after a particularly draining tour for the Aussies, and against a bowling attack whos spearhead was Garth Mckenzie, someone who would not be considered in the top 6 of West Indies fast bowlers of that period, or Australia for that matter, who had the Walker LilliThompsonPascoeHoggMallet. Australia in 1975, were no more talented than the West Indies, according to cricinfo WI vs Aussie timeline, just more organised and battle hardened. Whatever anyone says about South Africa, we just do not know. And remember, for all of Barry Richards greatness, he overaged 50.51 in England domestic, hardly better than other leading bats , and was never tested in different conditions. Sorry, I just cannot agree with your statement

  • Abhi on May 24, 2011, 14:08 GMT

    Arjun Thanks- am slowly beginning to see the light

  • Abhi on May 24, 2011, 14:04 GMT

    Ananth, I am kind of going through the above groups. In any case you have 5 groups in the main article. you could add one more group to the previous suggested list-

    The "super group" - Till -28 (Absolute best attacks ever) The "very good" group - 28-32 The "good" group - 32-36 The "good/average" - 36-40 The "pie chuckers" - 40 onwards

    Just a thought This certainly makes more sense. Ananth

  • Arjun on May 24, 2011, 12:42 GMT

    Alex

    Regarding grouping, shouldn't 3rd group(35.00-39.99) be considered as 'Test level' quality ? Likewise 1st and 2nd combined are 'above average' & 4th and 5th combined are 'below par'.

  • Alex on May 24, 2011, 12:19 GMT

    @Ananth: Congrats to Arjun for devising a simple and insightful idea!

    Again, it would be nice to have the "form" factor smooth out the "ctd" factor. For example, SRT's 136 & the Perth hundred don't make it to the top group but the 194* does even though these two bowling attacks were close to their peaks (even if we ignore the pitch conditions, which were difficult). Maybe these centuries got categorized in Group 4 in this analysis. So, another way to look at this analysis:

    1. Perhaps Groups 1&2 should be viewed together as the top bowling attacks. Here, the ranking is: Bradman, Viv, Sobers, Hobbs, Chappell, Hutton, Sanga, SRT, Ponting, Sehwag, Lloyd, AB, Lara, Miandad, S Waugh, etc. Sanga is the big surprise but makes sense otherwise.

    2. Likewise, Groups 3&4 should be clubbed together.

    3. Then, there are 3 groups in all. As far as I checked, all batsmen's averages increase across these 3 groups ... as is natural. [[ When we incorporate the Form factor, the Pitch factor and integrate these with the Bowling quality,we would have the mother of all analytical tables. Ananth: ]]

  • Arjun on May 24, 2011, 12:03 GMT

    Ananth

    The revised table is almost perfect, although on a later date we will see how 'form' related adjustments affect the current values. [[ And the Pitch factor. Ananth: ]]

    Abhi

    The current reciprocal method, automaticaly, gives more importance to strike rate. If you want to see results then i suggest you check the career BQTY table and compare it with previuos one posted couple of months back by ananth.

    You will notice player of yesteryears losing quite a bit and recent players gaining just becuase SRs of recent bowlers is very good.

    Sobers's career BQTY moves down form 33.40 to 35.04 Neil Harvey from 34.53 to 38.78 just because during those years bowlers had good ECO rates and poor StrikeRates.

    Since current bowlers have good Strikerates current batsmen's BQTY is improved. Moh Yousuf is up from 38.42 to 35.83 Hayden is up from 37.29 to 35.31

  • James on May 24, 2011, 11:09 GMT

    And what is this continued nonsense about the great West Indian attack being one-dimensional that it would eventually 'come back to bite them' and that they would eventually meet their 'Waterloo'.

    What part of this don't you understand: Other than when their stars were playing World Series Cricket, the West Indies only lost 1 very controversial series in New Zealand between 1976 and 1995, and were undefeated between 1980 and 1995. That is literally 20 years of unprecedented dominance. They were confronted with all types of conditions, every type of pitch, and were never beaten.

    So where exactly did the lack of a spinner ever 'come back to bite them'. It never did, because they had 4 of the greatest fast bowlers in cricket history who were able to adjust to different conditions.

    The modern Australian team lost to some very average teams along the way, and does not come close to having such a flawless record.

    The West Indies are clearly the greatest cricket team in history.

  • Ananth on May 24, 2011, 10:23 GMT

  • James on May 24, 2011, 10:20 GMT

    And these Australian fans who try and question the quality of opposition that the West Indies beat are seriously deluded.

    Most of these guys have probably never even heard of people like Imran Khan, Javed Miandad, Abdul Qadir, Sunil Gavaskar, Kapil Dev, Richard Hadlee, Martin Crowe, Ian Botham, John Snow, Derek Underwood, David Gower, Graeme Gooch, and the list goes on. Not to mention great Australian players like Dennis Lillee, Greg Chappell, and Allan Border who were all comfortably beaten by the West Indies.

    The irony of it all, is look at the mediocre opposition that Steve Waugh's team beat: A deplorable English team (widely regarded as their worst ever), a terrible New Zealand team, a chaotic Pakistan team that was often accused of match-fixing, a West Indian team in terminal decline, a one-man Sri Lankan team, an Indian team with no bowlers, and a mentally fragile South African team that always choked in big matches.

    Steve Waugh's team beat nobody. [[ Quite a lot of exaggerations as far as the Australian opponents are concerned. But these do not hurt anybody. However looking at Richards' and Lloyd's complete absence of runs against group 5, I can see that the overall quality during 1980s-90s was quite high. Ananth: ]]

  • James on May 24, 2011, 9:56 GMT

    @Obelix. You need to go and get your facts straight.

    The West Indies unmercifully hammered Australia in Australia in 1979, winning the series 2-0. They won in Adelaide by 408 runs, and in Melbourne by 10 wickets. That team had Dennis Lillee, Greg Chappell, Ian Chappell, Allan Border, Kim Hughes, David Hookes, Jeff Thompson, Doug Walters, Ashley Mallett, Rod Marsh, Terry Alderman, and Len Pascoe, while the West Indies best bowler in Malcolm Marshall hadn't even made his debut yet.

    So your claim that the Australian team of 1975 would have beaten Clive Lloyd's vintage West Indian team of the mid 80s is absolute rubbish.

    The only reason they won in '75 was because all of the West Indian greats were rookies the very start of their career. When they came back 4 years later in 1979 as hardened professionals, they gave Australia a royal thrashing.

  • Abhi on May 24, 2011, 9:17 GMT

    Ananth, Alright ,let me ponder on! Guess am just too dumb when there is extra math involved.

    Any ideas about the new groupings?

    The "super group" - Till -28 (Absolute best attacks ever) The "very good" group - 28-32 The "good" group - 32-36 The "pie chucker groups" - 36 onwards? [[ Too restrictive. As per the current BQI, 3450 are below 36. In other words nearly half would fall into the last group. That does not seem right. The Centurion Indian attack had a BQI of 41+. Like that many recent attacks. You should study the distribution pattern given in the main aricle. Ananth: ]]

  • Arjun on May 24, 2011, 5:21 GMT

    Ananth,

    I have feeling you have gone with the weighted average method in calculating career BQTY.

    result- 'Career BQTY' are on higher sides.

    I suggest calculate it with the same reciprocal method and re-post the tables.

    BQTY = (score1/bqi)+ (score2/bqi) + (score3/bqi) and so on...

    eg. take sachin's 2 innings 169 ag. SA (innings bqi-25.17) 203 ag SLK (innings bqi-41.90)

    According to weighted method career BQTY is 34.30.

    According to reciprocal method career BQTY is 32.18.

    that is why i belive reciprocal method is best as compared to weighted average method. [[ Yes, Arjun. I just replaced the NowQuality value with the Reciprocal method BowQty. I id not realize that even this summation can be done on a reciprocal basis. Will do and post the same. Ananth: ]]

  • Abhi on May 24, 2011, 4:53 GMT

    Arjun,Ananth: Just went through your excel sheet. Understood the explanation. However-you state: 1)New Inns bowQ takes both SR and ER into consideration. (But even a plain vanilla “bowling average” intrinsically does. 2)Since both ER-SR are considered there is no need to make extra adj. for SR( Why not? Again even the regular “bowling average” incorporates both qtys in equal measure) 3)A high ER will increase the Average as will high SR (Again this is automatically calculated when only “bowling average” is used) 4)The current method you have used is on assuming that Either All EcoRates are equal or All SRs are equal.( Could you explain this please?) Again, I like the idea of giving SRs more weight than eco rates. Also, as someone suggested the simplest way to arrive at a “useful SR “ figure would be to figure out the % of 40+avg. batsmen dismissed. [[ Abhi, don't miss the woods looking for the trees. Ignore the text and concentrate on the numbers. The key point is that Hooper's fifth-bowler-like figures impact the overall BowQ less using the reciprocal method than the weighted method. There are three calculations given to understand this. One is for the four top bowlers only. Second is one with Hooper, but based on weighted average method. And the third is the reciprocal method. Note now the third is closer to the first. Ananth: ]]

  • ObelixThe RomanBasher on May 24, 2011, 4:28 GMT

    @West Indies follower When you take a look at the WI sides of the 80's take a close look at the opposition. Australia were in transition in 1983 and it took them quite a while to get back a proper pool of players to be competitive. NZ, Pak and Eng were the relatively stronger outfits. NZ had basically a one man show with the bowling, Eng were good for a bit when Botham was on top but their forte was never consistency. Pakistan were the side with all of the weapons to compete with WI in the 80s and they did a very good job considering how unstable they can be. The rest were not worth mentioning. I will not even mention SA, they had players good enough to thrash the strongest WI side of that era, pity we didn't get to see that clash. However I have no doubt in my mind that had the 1975 Australian outfit been able to play the WI best WI sides of the 80's the Australians would have won. Would have been some close, but the one dimensional attack of WI would have come back to bite them.

  • Alex on May 24, 2011, 1:34 GMT

    @Ananth: I read the tables but did not double-check the method.

    1. The top 2 groups undoubtedly furnish the dataset for high quality bowlers. Fairly large (>40 innings) for most batsman who have played >50 tests. The average across these 2 may well summarize performances vs top attacks.

    2. So, I calculated the averages across the top 2 groups very roughly (without pen-paper/computer): Hobbs, Sobers, & Viv average in the 55-60 range. Chappell is 4th with 53. These are nearly their career averages (except for Viv)! SRT is 5th with 49. Then, a bunch follows (incl. Lara) at ave around 46-46. As a rule, I leave Bradman out ... he averages 88.

    The main caveat is the pitch quality. Sobers' 365*, Lara's 226, SRT's 194*, etc. were on belters (albeit one was a decisive test) and yet got rated as the top group performances. If the bowling quality has been accounted for well enough then this remains the only factor that needs a consideration. [[ Yes, AD was correct. Pitch quality is the only other basic cricketing measure which has relevance in the overall "quality" space. Once that is incorporated we can really sit down and argue. Ananth: ]]

  • Stark on May 23, 2011, 19:19 GMT

    Going by these statistics, Bradman and Sobers are easily the two best batsmen of all time!

  • Ananth on May 23, 2011, 16:53 GMT

    The following two files have been uploaded and the links are given here as well as at the end of the main article. 1. The explanation of the reciprocal method using a typical match # 1340. 2. The table of batsmen with their weighted bowling quality figures based on the most recent calculations.
    Er-Sr calculations: please click/right-click here.
    Weighted bowling quality table - ordered by WtBowQty value (gt 4000 runs): please click/right-click here.
    Ananth

  • somebody on May 23, 2011, 16:00 GMT

    [[ The mandatory rules of this blogspace are that you will not insult the author, the readers, the players and the countries. So even if you point might be a valid one, your comment will not be published ever if you violate these rules. Make your comment in an acceptable language if you want those words to see the light of the day. Ananth: ]]

  • shrikanthk on May 23, 2011, 15:32 GMT

    I think shrikanthk is "locked" on top 4 bowler idea much like he fixates on FC cricket data.

    Well. Fixations are good. It's Newton's fixation with the "Apple" problem that led to the theory of Gravitation.

    Anyway, here I am with another comment, though I told myself to stop commenting for a while. Just goes to show how addictive this blog is.

    Alex: I'm fixated because my heart goes out to Percy Chapman, the English captain in the '30 Leeds test.

    Poor fellow. Little did he know when he handed the ball to Maurice Leyland that July evening at Headingley that his decision will have such enormous ramifications on the way his bowling attack will be perceived eighty years later.

    Little did he know that this decision will spawn a heated debate in cyberspace after 81 years among a group of Indians!

    If he did, then I'm sure he'd have coaxed Tate and Larwood to bowl a few more overs that evening :)

  • Abhi on May 23, 2011, 15:06 GMT

    Ananth, Further to your earlier comment how about different groups - very very good attacks (upto 28) ,very good (28-32), good (32-36) etc. Possibly later. With the slight liberalization of BQIs, probably worth considering. Ananth

    h

  • A.Ali on May 23, 2011, 13:47 GMT

    Great work. This shows clearly why some batsmen like I.V.A. Richards are considered far better then the many others who have better average or more centuries to show. Are you planning to compute weighted batting average based on quality of runs scored?

  • Alex on May 23, 2011, 12:47 GMT

    @Ananth: I have no problems with this analysis. I only meant to say that while deciding All-Time Top 100 or so innings, masterpieces such as Hutton's 364 should not get penalized in any way because part-time bowlers (including a few who probably had never bowled before in tests) turned their arm over during the marathon. That project needs a different set of metrics.

    At any rate, any analysis will have loop-holes. For example, what really matters is the quality of bowling (& fielding) in a spell and not the bowler's stats preceding that spell. Bowlers routinely have off days and sub-par days. Just because the bowler's name is McGrath, the spell quality need not bear the stamp of a 22/60 calibre bowling.

    Also, one wicket-taking ball and the batsman is gone ... SMG took out a well-set Zaheer Abbas in his 90's _in_ Pak (his only wicket in test cricket)! The umpire's grand-mother might take Hutton out thanks to sheer chivalry from that gentleman cricketer.

  • Ali on May 23, 2011, 12:36 GMT

    Hi Ananth,

    I don't know if someone has already commented on this as there are too many to read now but I had a thought on Inzamam. I remember that quite a few of his innings, even against the top bowling attacks, came when the team when in a spot of bother. Is it possible to include the difficulty of the situation into the calculations to show which players excelled in a crisis situation. I am sure there will quite a few batsmen who perform better under pressure than others.

  • mohanlal on May 23, 2011, 10:59 GMT

    cont... in 1997 than in 2002 due to pedro collins.Another instance is Ponting against Ishant sharma. He was troubled thru out in 2007 tour by Ishant.All the cases above shows 3 modern greats troubled by specific bowlers.This shows that we cannot determine a BQI based on deliveries bowled by best bowlers only.Another case is of Sehwag vs NZL in last tour.When every one scored runs in tests he was a failure because his instincts to dominate which is his strength normally turned out to be his weakness there.These all implies that we cannot measure a batsman based on his runs and average made facing bowlers with high calibre only. [[ Mohanlal Please make your point in precise terms in no more than 2 comments. Otherwise people would not read those. These suggestions apply to all and sundry. Ananth: ]]

  • AD on May 23, 2011, 10:53 GMT

    Ananth: That is the actual email address.

    I had initially figured that there would not be any major changes to the lists irrespective of the manner in which you chose to calculate the Bowling Quality. The accurate appreciation of Bowling Quality is more important for individual innings lists.

    However, as it turns out minor adjustments lead to major changes.

    The reason for this seems to be that several innings may fall in the range of the sharp cut offs chosen; Bowling Qualitys may fall just on either side of the 30 or 35 range etc. With relatively few innings in any one particular group relative to a players full career the inclusion or otherwise of a few innings ,specially large innings ,is what seems to be the determining factor which causes the wide variations. [[ Do not worry about these changes. These changes are the result of many innings in the 28-32, 33-37, 38-42 etc sub-groups moving about. My feeling is that overall the values make sense. And just because Hunte added one more innings to 260%* he came on top in group 5. If I had limited that table to 1000 runs minimum, there would have been less churning. Ananth: ]]

  • mohanlal on May 23, 2011, 10:41 GMT

    cont... bickel had got Lara 2 or 3 times already, definitely the captain would have bowled him more in next tests with the aim of getting him out earlier.So as per this method statistically the presence of bickel in bowling line up reduces the BQI but practically it really increased the strength of bowling attack.Had bickel not there Lara would have found the rest of the bowling more easy to handle and would have scored much more than it actually turned out to be.If so Lara would have more runs against group 5 bowling attack.Another case is SRT in 97 WIN tour and in 2002 WI tour.In 97 he had an avg: of 57 against Ambrose,walsh,bishop & rose.In 2002 he had an avg: of only 41 in their absence but only 1 pedrocollins who troubled him thru out the tour and got him out 5 or 6 times.As per this method the runs scored by SRT in 2002 are of less quality than the runs he scored in 1997, but the reality is that he found run scoring much

  • mohanlal on May 23, 2011, 9:57 GMT

    Sir, In some of your earlier topics ,i had put a few posts and some of them were not displayed.And when I enquired about it you had replied that time that those were neglected to avoid further sharp reactions.Quite naturally this time too I thought on the same lines.But a thing that frustrated me a little bit was that this time only 1 from 3 of my posts were seen displayed before my last post.Now that you have clarified the reason there might had been some technical error that these posts didn't reach your end.I am sorry if my earlier comments frustrated you.Now I may narrate certains instances.In one of the Aus-Win series the bowler who got out Lara most was Bickel.Infact 6 to 7 times.Both Mcgrath & Warne were playing in the series thru out.So it is quite natural that when both Mcgrath & Warne weren't successful that bickel would have been asked to bowl.And when Bickel got Lara 2 or 3 times already the captain would have handed the ball to him earlier in the next matches.So as

  • Arjun on May 23, 2011, 9:42 GMT

    Here We go. 'Magic' method is as under....

    A = SUM of ('Ctd EcoRate per ball' multipled to number of balls bowled by bowler in concerned innings.) B = SUM of (balls bowled by bowler in innings divided by his 'ctd SR')

    BQI = A/B.

    Wighted avg. method distorts BQI.

    In this ER-SR method, both Ecorate and Strikerate are separated from conventional average and new BQI is derived.

    Hope this will clear doubts of lots of readers regarding methodology used. Readers can themselves enter hypothetical/real values in excel sheet and see the difference.

    Ananth, now correct tables have been posted. [[ Arjun, mere explanation would not do. I would try and do the match I suggested and post the Excel file. Ananth: ]]

  • shrikanthk on May 23, 2011, 9:29 GMT

    Who, incidentally is the fourth bowler. In the SA test, is it Harris with 98 wickets at 36 or Tsotsobe.. So you see there is no simple solution

    Had to come back from my self-enforced break to reply to the comments!

    There is a simple solution. Just pick the four bowlers who have bowled the maximum number of overs in that innings. That's all. So, Hammond and Leyland miss out while calculating the BQI for the 1930 Leeds test. [[ Pl see detailed response to this point in comment of Alex. Ananth: ]]

    I am not sure I understand the new "reciprocal" method completely. I'm sure it is sophisticated enough. All I have is a simple question.

    How can the Leeds Test Eng BQI be 37.xx when the mean bowling average (both career and CTD) of the top 4 bowlers (who bowled 140 of the 168 overs) is less than 30?

    This is happening mainly because of Leyland's very high average (CTD - 100+) according to me. Maybe I'm wrong. If I am, you can tell me why it is happening.

    We can take this offline if you're okay with it.

  • Alex on May 23, 2011, 8:57 GMT

    @shrikanthk & @Ananth: I think shrikanthk is "locked" on top 4 bowler idea much like he fixates on FC cricket data. On Don's 334 - Leyland-Hammond bowled 28 overs and that waters down the quality of attack in that particular innings ... plain and simple. That doesn't mean the Eng attack was that bad on any given day and does not degrade the Don's innings either. On the flip side, a poor batting side makes the bowling attack look too good on stats basis ... that is the way cricket goes.

    Incidentally, shrikanthk does have a point: in evaluating an innings, we should focus more on Top 4 bowlers and not on everybody who bowled that day ... I think 9 guys bowled during Hutton's 364 for example. After all, a Don in 334, a Sobers in 365* or a Lara in 400* might make the opposing captain wish for the umpire's grandmother to raise from her coffin and start bowling! [[ Alex/Shri If you open the door to selction of 4 bowlers, you are opening yourself to quite a few problems. If we select the fifth bowler based on overs bowled. What happens if the fifth bowler is the best in terms of average. What happens if the fifth bowler captures most wickets. What happens if the fifth bowler dismisses the top batsmen. On the othe hand if we select the four bowlers on average. What happens if the fifth bowler bowled the highest number of overs. What happens if the fifth bowler captures most wickets. What happens if the fifth bowler dismisses the top batsmen. I can go on. It will take me 5 minutes to get examples. Re your 903 for 7, the 5-7 bowlers bowled 53 overs. You cannot just ignore them. There is a good possibility that if the umpire's grandmother rose from her coffin and bowled 10 overs, these will be considered with appropriate precautions on the ctd (coffin-to-doosra) bowling averages of 150-year old grand dames. These solutions only create more problems than solve them. In general the BQI figures seem to be very well developed. Of course when you have 6837 innings, I can punch holes on these innings at the rate of one-a-day for the next year. Even that makes it only just above 5%. Ananth: ]]

  • Gerry_the_Merry on May 23, 2011, 7:30 GMT

    Ananth, have you considered limiting the maximum and minimum weightage foe each bowler in any innings and vary it on Adriana within these limits? I can advance two arguments invfvour of this 1) if (say) Aravinda and Ranatunga had put on a 150 stand and others had bowled also and not only two Ws, the bqi would have suffered relatively and an implicit penalty on better deserving batsmen 2) spinners bowl more balls and usually (except for Murali and O Reilly) have averages of 25+ as weighting as per overs bowled will weaken bqi. Consider Sobers making 74* in '66-67. He would have faced spinners largely, but been penalized thru weaker bqi. In general any innings on last day with only spinners bowling will suffer. I would recommend max and min weightage except for injuries. [[ I do not know why we should tamper with actual number of deliveries bowled. There should be no limit like this. Instead I have limits, lower and upper on the s/r and rpo. That is to prevent outrageous figures, that is all.

    Possinbly I have not understood what you really want.Ananth: ]]

  • Arjun on May 23, 2011, 7:01 GMT

    Ananth,

    I know one single composite figure hides many truths. Can you post a tables with career BQI faced for all batsmen?

    I konw you had posted it before but it was with weighted ave. method, now want to see the difference. [[ Will do. That is my favourite table. Now it should have many more acceptances. Ananth: ]]

  • Arjun on May 23, 2011, 6:54 GMT

    Some observations.... Martin crowe's avg. moves up from 19.0 to 33.0 in top group. quite correctly.

    Hashim amla who was 2nd previously only behind bradman in top gourp is probably 'Last' in this revised version; avgs only 4.0(only 32 total runs) as compared to 60.0+ previosly.

    Steve waugh, Miandad, M Taylor, Jayawardene's stats are now respectable in TOP group.

  • Arjun on May 23, 2011, 6:41 GMT

    Shrikanthk

    What you are saying(5th,6th bowlers distoring figures) makes sense if Weighted avg. method is used in calculating BQI. That was done previously.

    But Now Ananth uses reciprocal method by seprating 2 most important components of Bowling average; ie. SR and ER.

    I call it 'Magic Method' since it takes care of lot of things and normalizes high/low early ctd avgs, 5th-6th bowlers bowling high % of overs etc. Evenif a bowler has ER of 6.00 or SR of 200 balls this method dosen't distort the final BQI that much.

    You can see by yourself by entering some hypothetical figures in excel sheet and will be amazed at the 'Perfectness' of this ER-SR method. [[ Arjun, can you send me the workings for a match with a way-out fifth bowler, say MtId 1343 (Win-Aus at Sydney), 2nd innings where Hooper (51 @ 53.78) bowled 24 overs but the BQI is still 29.46. Then many people will be convinced of the working of the reciprocal method. Ananth: ]]

  • Adeel on May 23, 2011, 6:37 GMT

    Toooooo much information to digest.... Why some readers are continuous denial of the author...!!! Aanth has done this analysis by steping in his own boot if u r not getting him then fetch the data and make ur own calculations and analysis & keep admiring ur respective heroes. Its non of ananth's fault that ur desired celeb is not there where u wanted him to b...! Its a state thing & stats more often then not speak 4 themselves....

  • Arjun on May 23, 2011, 6:26 GMT

    Ananth,

    There is some mistake either in your revised article or revised tables. Please check.

    just to give an example. In revised article tendulkar's avg. agaisnt (30-35group) is 57.0 In revised tables it is 66.0. [[ I do not know whether I am coming or going. The wrong file was uploaded. However the selected batsman file was correct. Pl confirm taht it is okay now. Ananth: ]]

  • Abhi on May 23, 2011, 6:13 GMT

    Ananth, Guess I misunderstood. I had understood from your previous comments that you had not given SRs any extra weightage at all.

  • shrikanthk on May 23, 2011, 6:12 GMT

    What, incidentally, is the "right" thing.

    In this case (and in similar cases) it is very clear as to what is the best thing to do.

    We may have another case in Test history where the 6th bowler has an average of 400 (and he ends up bowling 10 overs in an innings of 150 overs. The whole BQI will suffer very significantly. [[ Again you seem to have very little condidence in my cricketing and analytical skills. Do you think I would do all these withput limiting values. Only a fool would take 400 or for that matter a bowling average of 5. Ananth: ]]

    Why is this so difficult to accept? Anyway, You know best. You're the one who has done all the hard work. Maybe I'm talking rot. Bye. Gonna take a break from this blog. Hopefully, will be back sometime in future. [[ Who, incidentally is the fourth bowler. In the SA test, is it Harris with 98 wickets at 36 or Tsotsobe with 1 at 53. So you see there is no simple solution. When I did the Win/aus work, it was on the best four bowlers. However this is a batsman analysis. As such real life bowling attack was far more important than a notional bowling attack. Anyhow I suggest read Arjun's comment. I have also asked him to send an example of the reciprocal method. You can rest be assured that extreme values would not affect the BQI as much as you have imagined. Ananth: ]]

  • shrikanthk on May 23, 2011, 5:54 GMT

    Finally, I'm not advocating this suggestion just to accomodate Bradman's 334 in Group 5 or 4.

    That's not my concern. I'm only interested in doing the right thing. Should the 5th and 6th bowlers in an innings of 168 overs be almost entirely responsible for moving the BQI of an attack from Group 1 to Group 3?

    My answer : No. [[ What, incidentally, is the "right" thing. This perception varies from person to person. Ananth: ]]

  • shrikanthk on May 23, 2011, 5:48 GMT

    In the match you are referring to, Hammond and Leyland bowled 28 overs for 90 runs and 1 wkt (3.3)). The other four bowled 140 for 476 runs (3.4). So the part-timers have not bowled that badly.

    It doesn't matter how well the part-timers have bowled. My point is that the bowling averages of part-timers are not very reliable. Even though Leyland bowled only 11 overs (his bowling avg of 100 odd) is a terrible outlier - significantly skewing the overall BQI.

    The parallel with Tendulkar's 111 is not ideal. Tsotsobe was the 4th SA bowler in that innings (he cannot be ignored). Leyland on the other hand is the 6th bowler (in an innings that lasted much, much longer).

    All I am saying is - Consider the top 4 bowlers (sorted by no of overs bowled). So Kallis will miss out in that 111* BQI. Hammond and Leyland will rightly miss out in the 334 BQI.

    The reason Leyland and Hammond bowled 28 overs was because Bradman clobbered Larwood! [[ Sorry, arguments on this line of reasoning are nothing but waste of time. Ananth: ]]

  • AD on May 23, 2011, 5:46 GMT

    I still have complications when importing the linked files. The data does not sort into separate columns. Cold you please advise how to resolve this issue?

    If it cannot be done I had previously requested an alternative- Would it be possible to place the average of each group next to the selected batsmen list? Fore example- Chappell G.S 42 2 1490 21.0 5 37.25 (AVERAGE BQTY IN THIS GROUP) Chappell G.S 49 10 2723 38.3 4 69.82 (AVERAGE BQTY IN THIS GROUP) Etc.

    We may then mentally adjust the batsmen averages accordingly. [[ It is very simple. Do one of the following. Double click on the link and the text document will open in a separate tab. Save the file as a .txt file. You can then import ithat into an Excel sheet with no problems. Alternately right-click on the link and save the file as a text file. I have just done that for two of the files and verified. Incidentally you seem to be sending your comments from an unreachable mailid. I sent a mail to you which has bounced. I suggest you send a comment from a valid mailid.You need not be worried that it would be misused. Ananth: ]]

  • AD on May 23, 2011, 5:34 GMT

    The fascinating part about this modified analysis as compared to the previous is that supposedly “cold hard stats” are anything but. They are for the large part outcomes of subjective inputs. As also relatively small changes in these subjective inputs may lead to large changes in end results.

    The current analysis is too simplistic to satisfy. Personally, I prefer an analysis where we take into account all possible variables that would actually occur in cricket. The debate should ideally be about the various weightages that should be accorded to the variables, although this too is necessarily subjective. Not whether these should be incorporated in the first place. [[ AD You are absolutely correct. Slight changes to the basis mean significant group changes. But that does not really matter since there is really no "right" solution. What matters is that there should be no major anomalies. Ananth: ]]

  • Abhi on May 23, 2011, 5:11 GMT

    Just going through that match score card. Steyn had an eco. of 3.4 (Tsotsobe 5.5) - But Steyn took out Sehwag ,Tendulkar and Laxman!!

    Would any captain anywhere mind that? [[ Is there a message in this other than that Tsotsobe should not have been selected !!! Ananth: ]]

  • Abhi on May 23, 2011, 5:07 GMT

    Ananth I guess you guys are fed up of me – but here goes. You state : “Let us take the 111 referred to by Abhi in his recent mail. Steyn, Morkel, Harris and Kallis form a decent attack by themselves. With the Home advantage they come below the 30.0 mark. However it is a fact of life that in this innings Tsotsobe had analysis of 24-3-98-1. Now how can anyone ignore his bowling. He was buffet lunch. “

    But that was because Tendulkar managed to survive in the first place. Wouldn’t Tsotsobe have been buffet lunch in the 1st inn also? But what happened? Steyn managed to STRIKE.

    That is precisely what I have been harping on and on about the critical nature of strike rates- Once Tendulkar was well set Tsotsobe or any other bowler would almost surely go for MORE than his long term career economy rate. In the 2nd inn. Tendulkar and others to a degree managed to survive the onslaught. Just another eg. Of the sheer importance of “strike power” which we ignore at our own peril. [[ I am not sure whether I understand you. Can you tell me who bowled the first 50 balls Tendulkar faced. Could well have been Tsotsobe and Harris. I suggest do not over-analyze these things. This was an excellent attack, standing at 30.80, despite Tsotsobe's 24 overs. I suggest please be happy with that. How can all of you take single situations and expect changes to be done to fit in with your perceptions. I have already told that in the revised method s/r is handled as a separate entity and gets some share of importance. Ananth: ]]

  • shrikanthk on May 23, 2011, 4:39 GMT

    Ananth: Here's a very interesting example.

    Test No: 196 Leeds Test of 1930. Bradman scored 300 runs in a day against Larwood, Tate, Tyldesley, Geary and Hammond. Pretty decent attack I must say. BQI : 38.73!

    Larwood career avg 28.35 (CTD avg at the end of this test: 32.63) Tate: career avg 26.16 (CTD: 25.58) Geary: career avg 29.41 (CTD: 27.38) Tyldesley : career avg 32.57 (this was his last test) Hammond : career avg 37.8 (CTD: 42) Leyland : career avg 97.5 (CTD: 114)

    So, this attack is ending up with BQI GT 35 mainly because of the 5th and 6th bowlers (Hammond and Leyland)

    In an innings that lasts 168 overs, the 5th and 6th bowlers will end up bowling a fair bit. This happened because Bradman got on top of the main bowlers! Why penalise him for that?

    Here's my suggestion: Please consider only the top 4 bowlers in each innings to calculate BQI (i.e the top 4 in terms of overs bowled). This would've ensured that Hammond and Leyland's outlier average doesn't skew the analysis. [[ I can show the counter example. Let us take the 111 referred to by Abhi in his recent mail. Steyn, Morkel, Harris and Kallis form a decent attack by themselves. With the Home advantage they come below the 30.0 mark. However it is a fact of life that in this innings Tsotsobe had analysis of 24-3-98-1. Now how can anyone ignore his bowling. He was buffet lunch. In the match you are referring to, Hammond and Leyland bowled 28 overs for 90 runs and 1 wkt (3.3)). The other four bowled 140 for 476 runs (3.4). So the part-timers have not bowled that badly. Also give up the requirement that the 334 has to be against very good quality. Does not happen often. Three of Bradman's 200s have come against the group 5 attacks. So why change a system just to get one innings in. Lara's 300s have been against two attacks either side of 40. That is the way it would be. In fact, in the selected player list, only Sobers' 365 and Sehwag's 309 have breached the 30.0 mark. Priobably right. Ananth: ]]

  • Pramod on May 23, 2011, 4:33 GMT

    This thread is not referring to this article btw...it's just a suggestion , could u come up with the bowling averages of bowlers against the top seven of the batting side, there are other factors such as a nightwatchmen,debutants et all. It would be interesting to assess the quality of bowlers through this parameter and the bowler's true average would come into picture. This is not to say the wicket of a tailender is cheap, but it would be better if one were not assess averages without them.Thanks. [[ Yes can be done. Somehere there I have to look at the forgotten species, by themselves, not as supporting actors bowling to these set of batsmen. Ananth: ]]

  • Abhi on May 23, 2011, 4:17 GMT

    Ananth, I was referring to the recent 111* not the '92 one- but anyway, i guess it is a similar Bqty. [[ The recent 111 was, lo and behold, 30.80. A pretty serious attack. Tsotsobe's 24 overs, in his fifth Test pulls it down quite a bit. But taht was also the case in real life. Virtually no pressure from him in this test. Ananth: ]]

    Srikanthk I guess you are referring to me. Take a look for yourself. The best "'90s" batsmen avg around the 40 mark. The best "2000s" batsmen avg. around the 50 mark. If you were looking through a balance sheet instead of cricket stats this should have immediately raised a "red flag".

  • Alex on May 23, 2011, 3:45 GMT

    @shrikanthk: That's fine. I was just trying to see if there is any way stats can show Kanhai to be a really special player (since the great players hold him in high regard but casual fans forget about him). Maybe there is no such way. VVS also averages only 48 with relatively few 100's and 50's but, to me, he is a very special batsman ... very few weaknesses and pure magic with willow.

  • Abhi on May 23, 2011, 3:40 GMT

    Also Tendulkar's 146 and 111* vs SA not in the new list. Still checking the lists out. Guess we just have to wait for a composite bqty/pitch qty etc analysis [[ 111 vs SA - 36.42 (Donald + 4 ordinary bowlers) 146 vs SA - 30.10 (just missed the group 5 bus). Ananth: ]]

  • Abhi on May 23, 2011, 3:33 GMT

    Ananth, I had mentioned in an earlier comment my surprise that the Tendulkar 169 wasn't in there but the NZ 160 was. You had posted a detailed reply. The SAF attack worked out to a Bqty of some 36. Now it is 25! Guess "intuition" often works better

  • shrikanthk on May 23, 2011, 3:22 GMT

    Now, why on earth do I hear voices seeking a return to the old approach? Is it because Tendulkar's average against Group 5 has dropped a notch? :) (Just kidding).

    Any methodology must be judged based on principles and not based on the results it throws up. The new approach is better, on principle.

    You HAVE to tweak the CTD average for bowlers with less than 50 wickets, such that they reflect career performance as opposed to very volatile early career performance. I suppose that's what you've done now.

    No wonder Bradman/Hutton have fewer innings in Group 1 (partly vindicating the points I had made earlier).

    This may not be the BEST methodology. But a definite improvement over the earlier approach. [[ Two schools of very rational thought. I am fine as long as the groups (not the bowling) keep their cool. Ananth: ]]

  • shrikanthk on May 23, 2011, 3:07 GMT

    but Kanhai has 37+ in a whopping 48% of his innings. This suggests a different batsman with an extremely solid game

    Alex: I'm not a huge fan of this line of reasoning. Read my earlier comment in reply to Abhi (May 22, 1:29PM) At the start of the innings, most batsmen are vulnerable regardless of how good they are. Not even Bradman is an exception. This is because when you are new to the crease, you are often rusty and also susceptible to nerves.

    The ability to kick on after reaching 20 is not purely a function of temperament as many people think. It's also about skill/tightness of technique.

    So, I find it hard to argue that a batsman who gets more starts than others is a more solid player.

    Similarly, not kicking on after getting a start need not always be an issue of temperament per se.

    Kanhai must have been a special player though. My comments are more generic. I don't mean to put him down in any way.

  • Alex on May 23, 2011, 2:58 GMT

    @Ananth: Abhi is right, IMO. I think the previous tables shed more light ... still not sure about SR but maybe the period adjustment should be kept in the analysis. [[ Alex, I am not going to anything now. If that is the consensus I will do it in the final composite version when I incorporate the bowling quality and pitch type pmeasures. Part of the s/r adjustment has been included in the revised method of working the reciprocal crd average sone separately for s/r and rpb. Ananth: ]]

  • Abhi on May 23, 2011, 2:22 GMT

    Ananth, Have the attached tables also been updated? [[ No, not yet. Some link problems. I will do it now. Ananth: ]]

  • Abhi on May 23, 2011, 2:21 GMT

    So some batsmen such as Tendulkar (39), Lara and Steve (43) , VVS (39), Border (36) G.Chappel (37) have changed Drastically ! The Pontings (48) ,Sangas (50), Sehwags (46), Jayawardene ( 40) (Though some of them have come off with lesser avgs) have apparently not only faced better bowling in the 2000s but have even done better! Ha!- bring back the period adjustments and strike rates to make things more “real”! [[ No, positively not. I will make changes only to correct errors if detected by me or pointed out by the readers. Ananth: ]]

  • Ananth on May 23, 2011, 1:44 GMT

    I have posted the major revision of the article with the following changes. 1. No S/R or Period adjustments. 2. Only H/A adjustment. 3. Tweaking of the below-50 ctd wicket situations. Details in the article. Ananth

  • Alex on May 23, 2011, 1:38 GMT

    @shrikanthk: Pollock seems like a refined version of Gilchrist. Also, like Gilly & Sehwag, all timing & balance with not much foot-work. So, his range of shots was probably limited. Barry Richards possibly was better both technically & on shot-range.

    Mad Max was unstoppable on his days (@20% innings ... more so in SL) but Kanhai has 37+ in a whopping 48% of his innings. This suggests a different batsman with an extremely solid game. Maybe he just lacked the will to score heavy.

    In an interview (can't find the URL now), Lloyd talked about Sobers, Viv, & Lara as follows: Timing & placement is Lara's forte. Esp., placement: he is unique there. Also, he scored big once he got in. Viv too was dominating but Lara has an edge in the spin department. Sobers, because of the way he played, could make a bad pitch look like a good pitch. He had good hands, good wrists, a genius really. Seems like Lloyd put Viv at #3 for sure with Lara at #1 on good tracks & Sobers at #1 on bad tracks. [[ Wonderful to hear something different. I would add Tendulkar for his all-round excellence. Ananth: ]]

  • Pallab on May 22, 2011, 18:49 GMT

    hardly a reflection of his real International worth unlike highly motivated Bradman and SRT (even Dravid maybe) who played almost every FC game with the same level of devotion/motivation and professionalism. Note that Laxman barely shows the same hunger now for 200s and 300s in FC cricket which he showed during his phase to regain his place back in Tests. Even Botham could on a switch turn on his performance at international level with hardly any FC form preceding it. Another thing to note is that Asian players were hardly paid handsome emoluments (unlike now) for FC unlike the professionals playing in OZ and Eng for long –so obviously different motivation levels to raise performances. Let’s just treat FC as a separate entity and not use weighted averages to pad players who might have a small sample size in Tests to highlight their place in Test history or not.@SHRI:read in continuation to Dravid etc. above (May 22, 2011 8:34 AM)

  • Pallab on May 22, 2011, 18:46 GMT

    “I strongly recommend a weighted average of CTD test avg and overall FC average” @SHRI: Ad-nauseum plea for using FC records to account for small sample size of Test players-batsmen or bowlers. What’s the point? Which analyst, pundit or fan will subjectively and objectively accept the superiority or otherwise of Eng, OZ (these 2 purportedly being acknowledged the best) India, Pak, Caribbean or SA FC cricket? I remember the clear dissing of Indian Ranji Trophy records by non-Asian fans when Barry Richards’s first class records were used to anoint him a Test legend ! in S. Rajesh’s column. Can we extrapolate Kambli’s (who I still hold a candle out for and feel that in the current well-managed Indian cricket structure of mentors and balanced coaches would have been a certifiable Test legend who was penalized simply because of 1 poor home series against WI) superb FC record to go with his small sample 17 Test returns and but equally great 54 average to place him among SRT, SMG,CONTD

  • shrikanthk on May 22, 2011, 17:19 GMT

    Alex: Thanks for the link! Had read it long back. Didn't remember it too well. I was just wondering whether Kanhai resembled Aravinda De Silva in some respects! James' essay reminded me of De Silva - the "Mad Max" of the nineties. Unstoppable on his day. But hell bent on self-destruction quite often.

    He went "Sobers is #1 ... you should have seen Graeme Pollock ... Viv's got to be there

    Chappell is as good a judge as any. Another reason why I strongly feel Sobers should be universally regarded as the greatest of post War batsmen.

    Re Pollock: I've seen a lot of him on videos and read about him too. I am not convinced whether he was as complete or as dynamic a batsman as Sobers. A tremendous off-side player, but perhaps not as fluent on the leg-side. I liked what I saw. But he gave me the impression of being "Ganguly on steroids"! The "stand and deliver" approach is somewhat similar.

    Sobers appears to have been a more orthodox player than Pollock.

  • Alex on May 22, 2011, 14:02 GMT

    @shrikanthk: Since Kanhai averaged just 38 vs the top group in this analysis, I just computed the # times he scored scored 37+ runs in his career. It is 65 in 137 innings ... a whopping 48%!! So, he probably didn't care much for 50's & 100's ... probably lacked the mental resolve to go on for bigger scores. He indeed converted 43 of his 50+ scores into only 15 centuries. CLR James has an interesting essay on him (http://www.guyanaundersiege.com/leaders/kanhai.htm).

    Regardless of what the stats say, we always know when we see someone special. During the '08 Oz tour in India, Ian Chappell was asked on TV to name 4-5 greatest batsmen he had seen. He went "Sobers is #1 ... you should have seen Graeme Pollock ... Viv's got to be there ... also, people don't talk about him but Kanhai was absolutely great ... Lara-Tendulkar also".

  • shrikanthk on May 22, 2011, 13:29 GMT

    How important was getting him out early relative to restricting him? After he was well set how valid would the long term bowler averages be

    I'd argue slightly differently. I think that when a batsman is new to the crease, he is vulnerable regardless of the bowling attack he faces. The likelihood of Bradman getting out for a single-digit score against Rangachari/Hazare is not too different from the likelihood of him getting out for a single-digit score against Larwood/Verity.

    Once he crosses 20, the likelihood of Rangachari/Hazare dismissing him for less than 100 drops very sharply. Whereas, Larwood and Verity still have a reasonable chance.

    Here's a simple test for this hypothesis: Likelihood of Tendulkar getting out for less than 20 in Tests: X Likelihood of Tendulkar getting out for less than 20 in FC : Y Average score of Tendulkar in Tests once he crosses 20 : A Average score of Tendulkar in FC once he crosses 20 : B

    Here's my bet: X = Y (atleast similar) A LT B.

  • Alex on May 22, 2011, 12:59 GMT

    @Abhi: Of course, you would rather take the Don (or any batsman) as early as possible. Also, cricket is a team game. The sight of Don batting would fuel the confidence of the other batsman and dampen the morale of the fielding side. So, best to get him out ASAP unless you are defending in the 4th and he is being painfully slow.

    I don't think you are posing the question properly. If we pose the following question, it may have meaningful answers. Say, there are 2 choices. Under the first choice the batsman will score 25 runs off 25 balls and then get out. Under the second choice, he will score just 8 runs off the first 60 balls but will continue to bat on. Now, which of these two choices is better?

    The answer to the above question depends on a # factors. No easy closed-form answer using mathematics either since those things are neither i.i.d. nor fully correlated.

  • shrikanthk on May 22, 2011, 12:41 GMT

    In discussions on great batsmen, casual fans often forget about Kanhai. As with Roberts, the career stats don't reveal Kanhai to be the supreme master that he was

    Right. In fact, when Kanhai first toured Australia in 1960-61, a lot of old timers were defiant that they had never seen such a commanding batsman in their country since Bradman's peak years.

    I think an average of 50 is a psychological barrier for a lot of critics. Batsmen who ended up a notch below 50 occupy less mindshare among fans and critics! Kanhai, Worrell, Harvey, Walters, Morris. Unlucky chaps.

  • Alex on May 22, 2011, 12:40 GMT

    @Abhi: I didn't mean that McGrath took SRT out regularly. However, he was a big factor on why SRT has only 2 hundreds (both small) in 8 tests vs top Oz attacks of '98-'04 despite crossing 40 in incredibly 50% of innings. I have thought a lot about it.

    1. Aussies overcame SRT on psychological terms really (see item 2).

    2. McGrath's strategy (Gillespie's also) was to deny SRT run-making opportunities and other bowlers now had a better chance since McGrath would put SRT out of his comfort zone ... SRT needs a 4 every now & then to feel good about his game.

    The '98 back injury had cost SRT 2 unorthodox money shots (lofted cover drive & lofted on-drive/pull) which could have broken this lock. Lara was better on patience & concentration (once he got in a zone) and had better traditional shots than SRT ... so, he could counter this strategy despite being inferior to SRT on technique. The more patient SRT (of 2010, e.g.) would certainly have fared better vs McGrath.

  • Naresh on May 22, 2011, 10:00 GMT

    You need to consider relative team strengths.

    e.g. australia in 1999 - strong batting and bowling. The oz team would pile up runs on the board, which would give more room for attacking the opposing batsman - thus making it more formidable. You could say thats the same as "strong bowling attack" - but, to the best of my memory, Pakistan was a strong attck in 1999 - and we all know what happened. It works the other way round too - it was easier for the australians to score runs then - but look whats happening to Ponting now, with a much weaker oz side.

    It goes back to getting runs against the best perhaps - but to consider what is the strength of the opposing team AS WELL AS your team. It is said a great team makes good batsmen appear great (many recent examples can be quoted but I won't)- this measurement will separate the merely good from the truly great.

    Its about...when u walk out to bat, its not just who you up against, but also who else is in your dressing room.

  • Pallab on May 22, 2011, 8:55 GMT

    right thru 90s and beyond. In a retrospective revision of the cricinfo all-time World XI in a decade or so, I foresee classically correct Sanga (in batting as well as keeping) displacing Gilchrist. I really feel bad for Aravinda de Silva who practically owned tough OZ attacks in ODIs home and away with a fantastic record and did well in Tests in NZ and OZ too in the limited opportunities he got. De Silva was in the Lara and SRT class of batsmanship right thru the 90s and was as good-at times better than – as SRT in ODIs while racing to 9000 runs.

  • Pallab on May 22, 2011, 8:53 GMT

    “Sanga`s great record against the best came as somewhat of a surprise. He seems to have been (very!) unfairly classed as a bit of a flat track/weak attack bully. I`ve been as guilty of that as any” @BOLL: Asians have long considered Sanga as world class. I had earmarked him as a great in 2002 itself after watching him closely the full NatWest series properly. Maybe if Aus, SA and Eng boards start giving 3 Test series (finally happening this year in Eng and a series in SA after 2002! and despite SL being competitive in Tests after 1995) to SL, many would start noticing Sanga and Mahela (finally making up for some of his poor away records due to longevity much in the same manner that due to longevity SRT has corrected his quite average record in SA) are certifiable Test greats (something which world-class Aravinda and Murali did not get a chance to do much in 90s in Eng and OZ). It is to the credit of BCCI and Pak boards that they kept inviting SL regularly for 3 Test series CONTD.

  • Pallab on May 22, 2011, 8:34 GMT

    Dravid, Sehwag and VVS in the pantheon of Test greats? Similarly, who is to say that 60s-70s Ranji Trophy cricket was not “tough” enuf with the Indian Test quartet spinners along with Shivalkar, Hans,Goel, Yadav,Doshi plying their wares ;which Indian batsmen would take quite unkindly to their records being slighted just cos OZ and Eng domestic circuit never produced such an array at the same time?I have it on good stead that even the unstructured domestic cricket set-up of Pakistan has so much depth in pace/fast bowling stocks which NEVER see the light of day and which would otherwise make Pakistan’s domestic cricket far more tougher and competitive than say OZ with like fast medium bowling resources. It is a wonder that Gavaskar,Vengsarkar, Jimmy,SRT,VVS,Sehwag raise/d the standard of their game SO MUCH in international against pace despite generally facing medium pace trundlers in their formative years in domestic cricket. A moody player like Sehwag’s first class record is CONTD.

  • Pallab on May 22, 2011, 8:25 GMT

    “Interesting. But I suspect most of those 9 innings were in the Bodyline series when he was up against Larwood, Bowes, Voce and Allen, when the attack he faced was downright immoral.” SHRI: “downright immoral “Again overstating and getting caught in the glare and narrative hype about the highly publicized, written about and mythologized Bodyline series (the dramatized docu of which was also shown in India in the 80s).Ball for ball, Holding (at 90 kmph faster than Larwood), Daniel against Indian batsmen in the 1976 series represented the most immoral attack as they were also bowling BEAMERS disguised as waist and above waist-high full tosses. Holding was the absolute master of intimidatory pace bowling in world cricket history. Gavaskar too like Bradman and co. did not wear helmets (and even chest pads of 70s were rudimentary). Larwood and co, were bowling conventional, back-of-the-length bouncers aimed at body/rib cage, the type of attack which Windies pace quartet and even CONTD.

  • Pallab on May 22, 2011, 7:43 GMT

    “S Waugh's 200 at Jamaica in 1995 (along with Mark's 126, these two innings are probably two of the most important innings played by ANYONE in the past 20 years).” SHRI: Overstating the case with respect to the period of 20 years and just cos of the toppling of the WI in WI-it’s OK to say most important innings(and I note that you have pushed your fav M. Waugh too to include the important innings.That’s neither here nor there.Every significant score of Sidhu in 1998 against OZ were also important as they busted the perceived threat of Warne for the latter Indian batsmen and maintain their vice-like grip on him throughout his career).In the noughties,I could turn around and say that Ganguly’s 141 and Dravid’s 241 in Aus in 2003-04 were the MOST important(which they were as Waughs in the 90s perhaps)SIMPLY because they slowly chipped away at the OZ citadel in Aus and showed the way to subsequent SA and Eng teams which finally began winning “live” Test matches in OZ and series too.

  • Pallab on May 22, 2011, 7:36 GMT

    Ananth: Just noticed that you quickly corrected that statement about the “one-dimensional “ attack of WI meeting their Waterloo when they never DID actually for 15 years in response to West Indies Follower’s post. Ananth: Honourable mention can be made of Manjrekar who made 769 runs in 13 innings @64 average(1 NO) during his annus mirabilis 1989 performances against WI and Pak (both away and am sure with high BQI for the bowling attacks) –much like M. Amarnath’s 1983 performance. Manjrekar was actually desperate (have interacted with him) to be the heir to Gavaskar’s Bombay school of batsmanship legacy but arrogance about his own purist technique and the refusal to seek advice from coaches/peers led to his downfall and subsequent overall poor Test returns. This is where a like for like batsman Dravid scored over him and became a Test colossus.

  • narinder on May 22, 2011, 7:28 GMT

    I wish someday I see SRT and BCL playing together in the same team even if its in IPL or any charity match.

  • Aparajith Shyam on May 22, 2011, 7:26 GMT

    “It can certainly move up or down with the way the player is performing. What we are saying is that the Bowling average is the most important measure for a bowler since it encompasses both strike rate and bowling accuracy.”

    Ananth: Agreed. Bowling average is important. My only point is to look at whether a Mike Hendricks (25/71) and a Simon Doull (29/61) are similarly effective bowlers. Both were fast bowlers, bowled in similar conditions and to batsmen with similar mindsets.

    I put my comments to see if the above was captured in the analysis and as you pointed out before, it has been included.

    I was referring to Anil Kumble to stress upon the point that, even when pitch conditions/batting mindsets in the 2000s changed, Anil Kumble remained the same effective bowler that he was in the 1990s, only that certain parameters of his stats changed. Anil Kumble = 27/70 = 29/65. At the end of the day, yours is a very fine analysis. I am just looking to contribute.

  • Abhi on May 22, 2011, 7:12 GMT

    Another (surprising!) eg- I was just checking out of curiousity. Bradman has 6 Ducks in 63 innings vs. England (almost 10% !) Yet his overall avg. vs England is 89.8 ! How important was getting him out early relative to restricting him? After he was well set how valid would the long term bowler averages be?

  • Abhi on May 22, 2011, 7:04 GMT

    Alex In matches "involving" Mcgrath ,Tendulkar has 7 50+ scores . Mcgrath has got Tendulkar just once in these scores. So, that doesn't gell with your comment.

    I feel that a similar trend would be discerned in most top batsmen vs. top bowler stats.

    After the top guys are well set a great bowler will certainly restrict them RELATIVELY more than a lesser bowler- but the ASSUMPTION being made in most stats is that the bowler will give away EXACTLY the same amount of runs as his longterm( or CTD) economy rate.

    This is what i find somewhat counterintuitive . Though it cannot be argued with using math.

  • narinder on May 22, 2011, 7:03 GMT

    Great article.I have tried to get my version.Hope people will like it.I have not seen much of Sir viv & other older players.My real batting heroes of the last 20-25 years have been SRT and BCL.There have been pontings and others but I think the 2 i refer to are shoulders above others.So i just compared the two of them only. I first made an analysis as to who were the bowlers with best bowling Avg during the period these 2 played (1.1.89 till date).I selected the bowlers with minimum 50 wickets during the SAID PERIOD.I think lowest bowling avg over 50 wkts is good enough criteria.Following is how SRT and BCL have done against bowles having avg below 30: M Runs Avg. 100s SRT 135 11398 56.14 40 BCL 117 10774 53.07 31 I think away avg is better indicator coz it shows how u did in the home of the opposition.Following is how thse 2 did agst them: M Runs Avg. 100s SRT 76 6103 53.06 20 BCL 59 5077 47.44 15

  • Ramesh Kumar on May 22, 2011, 5:44 GMT

    Aanth,

    superb work and appreciate the work which has gone into this. One observation--How do we treat short bursts of highs of bowlers which resulted in wins-Kasparowitz & Hirwani etc? My theory is that each team tries to put their best available 11 in the match and some of them shine for some and fade out. But they have outstanding matches and win matches for their country, why we should treat the bowling strength as inferior? It is understandable if it is a bowlers' analysis. But since this is batsmen analysis, I think we can think of a tweak to give higher weigtage to bowling if they win the match for their team or made their team perform better than the opponents. Facing Mendis in the first two years would be as difficult as facing murali at his best. Congrats once again for excellent work.

  • Alex on May 22, 2011, 5:09 GMT

    @shrikanthk: Walters (74 tests, 5000+ runs @ave=48) crossed 50 in 48 of his 125 innings. This outstanding percentage almost equals the one posted by Viv & SRT, the leaders in the post-'50 era in this category. However, he converted those into only 15 hundreds.

    In discussions on great batsmen, casual fans often forget about Kanhai. As with Roberts, the career stats don't reveal Kanhai to be the supreme master that he was. He has managed 39,48,48,50,53 in these tables. His contemporary Sobers has a whopping edge on him in Group 1. Kanhai's relatively poor record in Group 1 is a surprise.

  • Pallab on May 22, 2011, 4:44 GMT

    @BOLL: Since this blog is about myth busting (with some posters pointing out the poor performances of Miandad, M.Crowe, Waugh, Abbas and the like), time to also BUST some statistical MYTHS about “holy cow” Bradman, especially for a blindsided “fan-connoisseur” such as BOLL (using this term for BOLL who otherwise has been fairly balanced and neutral about other players on Ananth’s blogs).I was about to do this while S. Rajesh had dissected Bradman(but Rajesh did not REALLY critique him cos of FEAR of brickbats) in the Test Legends series but now feel this is better forum to do so. 1. Myth: Bradman was actually scoring a century every innings and a rapacious run-scoring machine when FACT is actually his humungous average of 99.94 is based ONLY on his 29 big-scoring 100 innings. Bradman’s entire Test success, mythology and reputation is purely built on those 29 hundreds which has seen him amass 5393 runs at an average of @ 207.42 (practically 24 innings as 5 were not outs) CONTD. [[ Pallabh has sent 14 comments. I have no idea what to do. I appreciate his commitment but this is not the way to do. You have to be concise and keep to 2/3 comments at a time. I have summarized his next 7 comments below in which he says that SRT is far ahead of Bradman. I have summarized these mails below. Ananth: ]] Bradman has failed in 38 innings (below 50 runs). SRT 100 centuries - 30% above next. Bradman would average 65-70 if transposed to now. SRT finished product at 16. [[ My response to this is given below. Anyone who knows cricket knows that all followers say that Bradman scored a century every THIRD innings, not every innings. What you say is applicable to all batsmen. They are all top-heavy scorers. They fail quite a number of times. The age at which they mature has nothing to do with their greatness. It is quite silly to project Bradman's average in today's condition as 35% below at the same time assuming that today's batsmen would have handled the conditions 50 years back nonchalantly. And a century is not a true measure of a batsman's ability. It is nothing but the 100th run. A 99 can and in many cases is as important as a 100. I understand the frenzy created by media and followers and sponsors on the 100th 100. But as a pure cricketing measure, it is nowhere at the top. These are my personal views. These long comparisons are counter-productive. Ananth: ]]

  • Gerry_the_Merry on May 22, 2011, 4:03 GMT

    Ananth, have you considered limiting the maximum and minimum weightage foe each bowler in any innings and vary it on Adriana within these limits? I can advance two arguments invfvour of this 1) if (say) Aravinda and Ranatunga had put on a 150 stand and others had bowled also and not only two Ws, the bqi would have suffered relatively and an implicit penalty on better deserving batsmen 2) spinners bowl more balls and usually (except for Murali and O Reilly) have averages of 25+ as weighting as per overs bowled will weaken bqi. Consider Sobers making 74* in '66-67. He would have faced spinners largely, but been penalized thru weaker bqi. In general any innings on last day with only spinners bowling will suffer. I would recommend max and min weightage except for injuries. [[ There are always limiting values for every one of my calculations. That is how I can ensure that outrageous values are not developed. Ananth: ]]

  • Sarosh on May 22, 2011, 3:17 GMT

    Couple of points: 1) The home/away factor may be appropriate for certain batsmen. But a large number of even very good batsmen have huge differences in their home/away averages in several countries. In some cases the difference is more than 50% !!

    @Arjun: 2)I disagree with the "top batsmen from 1-7" argument. Ananth's filter is much better. All batsmen with a batting avg. above 40 may be considered good.Then whatever position such a batsman bats in a bowler gets credit for getting him out. This eliminates the problem of some nations such as Bangladesh, Zimbabwe etc whose top order may also be very weak. [[ Sarosh, If you have seen recent correspondence I am going to stay away from all these things and serve a simple Margharita Pizza in my revision. Ananth: ]]

  • shrikanthk on May 22, 2011, 3:15 GMT

    Hutton also seems to confirm his standing as England`s greatest

    I've always thought so. The only competition is from Hobbs. Though it may be argued that Hutton faced better bowlers (though the BQIs may tell a different story).

    Also, Hutton suffered from a handicap (shortened arm thanks to an accident during the War) which probably constrained his play after the War.

    Hobbs, however, was a more attractive player. Definitely a more aggressive player than Hutton going by SRs. But then, it isn't fair to compare the SRs of the two. Hobbs played in a different era, in a less security-minded age.

    Hammond doesn't seem to have done enough against quick bowlers for us to be entirely convinced. Thankfully for him, he played in an era when England was the only side with a decent seam attack. [[ You can cut, slice and dice whichever way, there is no denying that Hammond's average was garnered against lesser bowlers. In my revised analysis, due to be posted shortly, over 75% against groups 1 and 2. Ananth: ]]

  • Alex on May 22, 2011, 3:07 GMT

    @Abhi: Re your argument (May 21, 2011 3:51 PM): it depends on many factors. If the top-order batsman is SRT/RD and the 25/70 bowler is a McGrath (well, he was more of a 22/60) then they will bat him out without scoring too much off such a potent tight bowler. So, those additional 200 or so deliveries don't mean additional runs. In fact, an SRT might lose his patience and get out ... that was McGrath's strategy vs SRT and it worked well.

    Batting is about scoring runs and a draw is the preferred option to a loss. So, SR is not _that_ important unless it converts a draw into a win. We can't fault the pre-1930 bowlers for poor SR if they were bowling in 6-day/timeless tests ... uncovered wickets might improve their SR but timeless tests would work against it. If not for anything else, people should avoid deriding the late O'Reilly for the sheer fear of running into him in the afterlife! [[ I may run into Bill sooner than you guys and I hope he remembers that I put my hand up for him. What a wonderful and exciting player. Ananth: ]]

  • shrikanthk on May 22, 2011, 2:59 GMT

    Doug Walters' fantastic record may also help to quiet those critics who have always seen him as little more than a talented, but deeply flawed, dasher

    Boll: Didn't Doug Walters score of century between Lunch and Tea once? Against WI in '73 I think. [[ Walters was a drinking, smoking free-wheeler. Hidden behind this could-n't-care attitude was a wonderful batsman, over 5000 runs at just under 50, 70% against the top three bowling groups. One of my all-time favourites. Then the mantle passed to Mark Waugh and then Lara, with the Indian greats always there just behind these guys. Ananth: ]]

    Don't know a lot about him. Probably a bit like Mark Waugh - a batsman destined to be forgotten by future generations. People continue to rave about Ian Chappell even today. While nobody ever talks about Walters - arguably a better batsman with a far better average.

    Many people of my father`s vintage have told me that Sobers remains the best batsman they`ve seen

    A view shared by the Chappell brothers, Boycott among others. I'm yet to hear from anybody who has seen both Sobers and Richards and yet concludes that Richards is a better batsman!

  • mohanlal on May 21, 2011, 20:45 GMT

    Sir,just as i am impressed with your analytical skill, i am disappointed in that you are not always welcome to criticisms though they are only facts and not blind sentiments. This is the reason I feel 2 of my posts were not published.Yes they were about the limitations of Lara, but only genuine facts and with in the context of the matter too. Yet you deliberately ignored them.On the other hand i went thru 1 comment in these lines as 'BCL got 1000 quality runs more than his nearest rival'.Aren't we not welcome to post criticisms? The only comment received from you was published with the following response from me. No other comments on this thread have been received. Yours is the 9th comment published. I would appreciate if , in future, you take the trouble of going through the comments published before making such unwanted and unwarranted criticisms. These do not speak well of you. Ananth Response [[ Lack of motivation cannot be termed as justification. If anything the lack of motivation could be understood and accepted against weak attacks. My suggestion is to look at the groups 1 and 2 and possibly 3 together. These represent very good to fair attacks. Ananth: ]]

  • Pankaj on May 21, 2011, 20:12 GMT

    Dear Ananth, Thanks once again for just being there and for being yourself. Appreciate the Nadkarni nugget on radio - recollect so many vignettes, not the least WI being 3 down for 1 run at Trinidad 1983, then both Lloyd and Gomes being dropped, both getting their centuries and drawing the Test. Just a nitpicking idea - would it not be better to also factor in the respective batmen's runs as % of the runs scored? Bowler imbalances etc come up so often in the feedback, that may be a balance. Rgds PS I have a sneaky feeling that any modification in the parameters would work to advantage of VVS. My wholesome bias. [[ This has to remain simple, Pankaj. What you are asking does not come into the analysis at all. It is more an innings rating area. Ananth: ]]

  • shrikanthk on May 21, 2011, 16:35 GMT

    For example, take Ntini and Keith miller. I will accept that they are from different eras, so that some will argue that a comparison is not valid

    I agree. It is not a valid comparison. Period. I can accept a comparison based on bowling averages (with some reluctance), but a S.R/economy rate comparison - NO WAY.

    Ofcourse, we can have all sorts of comparisons if the objective is to inject some fun to this thread. But the results of our arguments, won't necessarily be definitive.

  • shrikanthk on May 21, 2011, 16:25 GMT

    Gary Sobers’ stats would probably have had the same impact.

    Gary Sobers is a very, very interesting case. It is very hard to judge his bowling. Because he bowled in so many different styles (seam up, left-arm orthodox, left-arm chinaman)

    I've read both Fingleton and CLR James comment on his bowling. The takeaway is that Sobers wasn't a particularly good spin bowler. But as a fast medium seam/swing bowler, he was as good as anybody in the mid-late sixties. So, when we look at his SR, we don't know how much it worsened because of his relatively inferior slower stuff.

    If I'm not mistaken, he started bowling quick on a regular basis pretty late in his career. Once he started doing that, he was twice the bowler. Many regard him as one of the best, if not the best, quick bowler in the world of the mid-late sixties (after Hall slowed down).

    Back in the sixties, even great spinners like Gibbs had poor SRs thanks to pad-play. Sobers, a lesser spinner, was no exception.

  • shrikanthk on May 21, 2011, 16:13 GMT

    You cannot totally dismiss modern day bowlers as non-penetrative and certainly I DID NOT generalize that Tayfield and others lacked penetration

    Aparajith: I said no such thing. My bet is that they're no more or no less penetrative than bowlers of yesteryear. Nevertheless, certain types of bowlers are an improvement over their predecessors Eg: Offspinners thanks to the leniency of the ICC. Also, swing bowlers who are very adept at "reverse swing" - which may be regarded as a postmodern invention.

    But let us not reach any conclusion on the penetrativeness of the modern bowler or the docility of his 50s counterpart by looking at Strike rates!! It can be very, very misleading.

    Cricket has changed a great deal over the past 50 years. Pad-play is practically extinct these days among top-order batsmen. Maybe I'm exaggerating. But I hardly get to see stretches of several overs where batsmen deliberately pad away deliveries on TV these days.

  • Abhi on May 21, 2011, 15:51 GMT

    @Aparajith Shyam,West Indies Follower

    Aparajith- I am with you on this one. The "problem" is that the mathematics is clear. As Ananth says he Bowling average encompasses both strike rate and bowling accuracy.

    The issue is that it effectively uses EQUAL "weightages" for strike rate and economy.

    Again the math is clear- say 2 bowlers with a similar avg. of 25 (but SR 50 and 70) At the end of the inn. they will have both conceded 250 runs but the 2nd bowler would have taken (70-50)*10= 200 balls more. Can't argue with the math.

    But in "reality" if any top order batsmen was well in he would in all probability make those 200 balls count for MORE than the "average" bowling average of the 70 SR bowler. I guess that is the essential point you too are trying too make- unfortunately we are both operating on "feel" for the game-and have no real mathematical justification to conclusively argue our points home.

  • Aparajith Shyam on May 21, 2011, 13:44 GMT

    “Sobers had an SR GT 90. Had he been playing today, batsmen will take more chances against him and will also get out in the process. So, his average might still be the same. Economy rate will worsen, but the Strike rate will improve. Bowling average is a far more stable metric than the S.R when you compare players across eras.”

    Shrikanth & : I don’t totally agree with bowling average being more stable. Your contention can be tested with a bowler who spanned through the late eighties, nineties and 2000s. Anil Kumble is a shining example. His avg / SR in the 90s were around 27/70. Towards the end of his career and still very much performing at his best (say 2007 Pakistan series), his average had dropped to 29 but his SR had improved to 65. Gary Sobers’ stats would probably have had the same impact. No, Aparijith, we are not saying that the average is a stable measure. It can certainly move up or down with the way the player is performing. What we are saying is that the Bowling average is the most important measure for a bowler since it encompasses both strike rate and bowling accuracy. Ananth You cannot totally dismiss modern day bowlers as non-penetrative and certainly I DID NOT generalize that Tayfield and others lacked penetration. I just was pointing towards interesting stats

  • Arjun on May 21, 2011, 12:29 GMT

    Ananth, I agree with you that both period adj and SR adj. should be dumped. Why don't you use ER-SR method that incorporates both the components of Average, i.e. Eco-rate and Strike-rate.

    Few months back i had sent you email with excel file attached. That method according to me is best at deriving BowQ index of innings. I am resending it, please have a look and reply.

    (New BowQ) * (home/away) will be best and simple to understand. Excellent idea. Sometimes we tend to forget that the simple method is the best. This would not be a problem at all. No major change expected other than that created by tghe removal of the two other adj factors. Ananth

  • Boll on May 21, 2011, 12:29 GMT

    Does anyone twitter/tweet? Warney is in last place 3 times in a row! Then again, if you`ve managed to score more than 500 test runs against the best, second best, and third best rated test bowling attacks of all-time, you`re probably within your rights to tell a twitterer where to go!

    And he retired yesterday. Love you Sir Warney. Great career mate. Lucky this is only a batting analysis. Great player, one of the greatest ever. Ananth

  • Boll on May 21, 2011, 12:20 GMT

    As I`ve often said, it`s fantastic when the stats agree with your prejudices. So it`s been great to see the two men who`ve revolutionised their position in the last 50 years, Sehwag and Gilchrist, two blokes I`ve loved watching, coming out roses in this analysis.

    2 blokes I`d have in my all-time 11. Sehwag I simply couldn`t leave out and Gilchrist, is THE classic matchwinner at 7. And what this analysis doesn`t take into account is the fact that they score twice as quickly as the competition. Sehwag vs Hayden? Gilchrist vs Flower? no comparison.

  • Alex on May 21, 2011, 10:10 GMT

    @Ananth: Arjun's suggestion has loopholes and should not be implemented. If A is a great bowler but does not get to bowl at a top batsman in the early part of a top batsman's innings (as happens a lot, depending on the flow of the game), he cannot possibly take him out for less than 10 runs.

    A more meaningful alternative is to look at the average #runs scored by Top 7 batsmen against him ... you had a relevant article back in year 2009. [[ Again many alternate paths and opening up a few debates. Ananth: ]]

    However, all of this is too much. If Hadlee is the only great bowler in NZ team, batsmen will look to be conservative against him ... so, him managing a great SR will be that much more difficult. So, a bowler's SR is a great metric but need not be looked into in this analysis. The real important factor is to account for the nature of the track: placid/lively. [[ Finally I also get the feeling that this will only over-confuse issues. There are many alternate nuances to the bowlers' measures. All of those make sense. But these cannot all be incorporated. Who is a top order player. 1-6 or Batavge>40. The first is important from a match point of view since Bangladesh may not have great averages. The second separates good batsmen out. Then there would be arguments over this and we will go ad infinitum. Where do we draw the line. At 9 runs. I am sure one or more reader would come and say that dismissing Tendulkat at 10 is as important as dismissing him at 9. Some one will point to a dismissal of Tendulkar at 136 which won the match and so on. I am now very tempted to dump both s/r adj and period adj, keep only the base BQI and H/A adj and do the tables. After all that was what we started with. This was the essemce of Arjun's suggestion. In my opinion this is a batsman performance analysis and complicating the already complicated bowler measures is not the way to do this. I willl then do the Pitch type analysis, this time making sure to keep it simple. And then combine both to come out with a definitive study on batsmen runs across bowler/pitch combinations across the ages. Thanks for making me see clearly Ananth: ]]

  • west indies follower on May 21, 2011, 9:58 GMT

    Re the Bill O'Reily debat. Why I think strike rates can be misleading is because they do not take into account the economy rate. For example, take Ntini and Keith miller. I will accept that they are from different eras, so that some will argue that a comparison is not valid. However, Ntini's strike rate is about 8 balls higher, but his economy rate is 1 run per over higher. Whilst it would take an attack of Keith Millers longer to bowl out a batting line up, Ntini would concede just as many run as Keith Miller in the fewer overs it takes for him to bowl out an opposition. In the course of a five day test match, with 500 overs bowled, Keith Miller's strike rate would not mean that he was any less effective. Also, we must remember that spinners like Warne would be able to get through those 57 balls to take a wicket in a shorter time than it would take a Mcgrath to get through 51.

  • Arshad Zaidi on May 21, 2011, 9:45 GMT

    Very time consuming study and great work, I must admit that.

    However, such statistics can be viewed in different ways. When prople make comments and come up with anything negative, you just got to accept the critisim in a constructive manner rather than to defend everythging you have come up in your research. [[ Frankly I do not understand the comment. Just because I respond to almost every comment does not mean I am defending something or other. Many a time explanations have been sough and given. Ananth: ]]

  • Gerry_the_Merry on May 21, 2011, 9:35 GMT

    Arjun 1) SR - I too think it should be done away with. It is no more or less important than for batsmen. % of top order batsmen is a good idea but why for low scores? 2) on period adjustment, while I have till now advocated doing awY with this, i feel that another extreme case illustrates the need for it. Imagine a decade with exceptionally strong batting line ups. Bowling averages will rise and attacks will migrate to lower groups unless a period adjustment is done. Only then will batsmen not be penalized for their own excellence. Except for pre-1936, other adjustments are mild anyway. 3) bqi- shrikanthk beat me to it, i had the same thought. 2 Ws bowling out SL for 71 by themselves doesn't not mean that the attack was truly that strong. I suggest having upper and lower bounds to the weightages in an innings for each bowler. E.g. Innings weightage should not very outside of (say) 15 and 33 %.

  • Arjun on May 21, 2011, 8:57 GMT

    Ananth,

    How many % of career wkts(top order1-7 for sub-10 scores) has mcgrath/pollock has taken as compared to steyn/Waqar ?

    This single parameter could be given value in the range of 0.90 to 1.10. [[ Arjun Pl see detailed response to the comment made by Alex. Ananth: ]]

  • Boll on May 21, 2011, 8:50 GMT

    @Mahendran. If topping the first table by only 30% and the second by 50% casts Bradman in a poor light, you must be more of a fan than the great man than I am!

  • Boll on May 21, 2011, 8:41 GMT

    @WestIndiesfollower. First of all, great to see a few more WI fans on here. I`m following the current series with great interest, as I`m sure you are. re. your comments about Waugh`s Australians, `never able to beat them`, you seem to have forgotten the 3-0 home victory over India in 1999! Antigua was a test match, in a series which Aus won 3-1. 1999 was a drawn series in which Aus retained the trophy. WI vs Aus in the `70s, you also seem to forget the 5-1 drubbing in Aus.

    WI,less vincible perhaps, but Aus 16 consecutive wins, twice in a row, speaks volumes also.

    Nevertheless, you make a fair point. WI were unbeaten (in series) for so long that it`s hard to envisage their record ever being matched.

    re. the long awaited simulation series. No helmets and 70 overs in a day, WI win. Helmets and 90 overs, Aus win.

    Enjoy the rest of the Pak/WI series. I can only say that a resurgent Windies team would be a great boost for world cricket.

  • Boll on May 21, 2011, 8:16 GMT

    re. my comment regarding Hutton. I was looking at his figures in comparison to Hammond, and forgetting the other great `H` JB Hobbs, whose stats, not unexpectedly, are excellent.

    re.Michael Vaughan as the `posterboy` of this article. Surely it was the mighty Lara`s turn. 3,800 runs against the best-rated attacks. 1,000 runs ahead of his closest rival. An average of 48 over 81 innings (1 NO). Not sure if this is your choice or not Ananth, but I think BCL (if not Bradman, although that would become rather boring in batting-related stats analyses) deserves to be on one knee, flaying it through the covers, at the top of the page!

    @Arjun. You mention some of the less than flattering stats of some great Pakistani batsmen against the best-rated attacks. It would be unfair not to mention their prominence in table 2 however (5 of the top 15 averages). Younis again performs exceptionally, as do Inzy and Javed, not to mention Hanif and Mohsin, 4th and 8th respectively.

  • Ashwath on May 21, 2011, 8:12 GMT

    Great Article Sir. Is it possible to analyse the performances of players in adverse conditions. Like Zaheer took 7 wickets for about 180 odd in the India pakistan snoozefest at faislabad and how umar akmal scored almost 80 runs in the ball dominated 1st windies pakistan test?

  • Arjun on May 21, 2011, 7:35 GMT

    I think you should do away with SR since over SR could be misleading. Since this analysis is about top order batsmen, bowler who were good at dissimsing them cheaply should be given more credit. Better overall strikerate is often a result of bowler wiping out tails.

    (BowQ and %ToW) * (home/away)

    period adj is not necessary. [[ Arjun Give me a day to consider this. If I want to implement this, some preliminary work needs to be done. Ananth: ]]

  • Arjun on May 21, 2011, 7:31 GMT

    Ananth, My Suggetsion is as under... Strike rate alone does not reveal the actual quality of bowler. Another parameter will do and it is not that complex to calculate. It is Top order batsmen dismmised for low scores(ducks and single-digit scores).

    Say, bowler A has SR of 52.0 and has dissmised 70 top order batsmen(1-7) for low scores (0-9) out of 183 career wkts(about 38%). Bowler B has SR of 43.0 and has dissmised 32 top order batsmen for low scores out of 183 career wkts(17%). I think Bowler A is better than B inspite of B having better overall SR; simply because he was able to dismiss cheaply top order batsmen; wheares bowler B may have better strikerate because of him dissmising tail too often. Bowler A could be Mcgrath, pollock, holding etc. Bowler B could be Wasim, Waqar, steyn etc. Insted of SR this parameter combined with Bow. Average could be used to determine bowQ. And only home/away adj. shoud be done. This Parameter could be named "%ToW" i.e. % of TOP order wkts

  • West Indies Follower on May 21, 2011, 7:10 GMT

    @Ananth 'An one-dimensional attack like Win of 1980s or Indian spin-based attack of 80s would find its Waterloo more often.'

    But funnily enough, the West Indies did not seem to find their Waterloo anywhere, considering that they won all over the world, and never lost, despite playing in more 5 match series than Australia, which would give the opposition more scope to rebound. Australia found their waterloos against India throughout the nineties, and Steve Waugh's great, mentally disintegrating team was never able to beat them. Indeed, the amazing seemed to happen more to Australia, (India 2001, Antigua 2003, West Indies 99, England 2005). Whenever there was a possibility of the West Indies losing throughout the 80s, they snuffed it out. Even i the 70s/80, West Indies beat and drew with full strength Australia, beat them in packer, and got the better of them in the drawn series in the carbbean in 1979. They did lose to New Zealand by 1 wicket. But otherwise, they were less vincible [[ Probably my comment does not apply to the all-conquering WI of the 80s. Ananth: ]]

  • Boll on May 21, 2011, 7:04 GMT

    As others have noted, this is a stunning, groundbreaking statistical collation. Just off the bat, forgive the pun, a couple of people stick out. Sanga`s great record against the best came as somewhat of a surprise. He seems to have been (very!) unfairly classed as a bit of a flat track/weak attack bully. I`ve been as guilty of that as any. Add these figures to his record as fastest (in terms of innings) to 8000 test runs and his career looks imposing indeed.

    Doug Walter`s fantastic record may also help to quiet those critics who have always seen him as little more than a talented, but deeply flawed, dasher.

    Less surprising are the records of Sobers and IVA. Many people of my father`s vintage have told me that Sobers remains the best batsman they`ve seen (quite apart from his all-round exploits) and I saw enough of Viv to know how good he was - the original Master Blaster, and the unquestioned great of his generation.

    Hutton also seems to confirm his standing as England`s greatest.

  • Ramesh on May 21, 2011, 6:59 GMT

    I hope after reading this article people will stop comparing Sehwag and Gayle. Both have similar figures in ODIs, but Sehwag is much, much better in Tests. Can someone say who's statistically better opener than Sehwag in Tests(Average and strike rate combined)?

  • Arjun on May 21, 2011, 6:11 GMT

    Ananth few readers have raised the issue of period adjustment. I think there is some truth; but my take on it is bit differnet.

    You will also agree that SRs of bowlers have improved since 2000s.

    You are making 3 adjustments to actual bowQ. Home/away is fine. the problem is with other 2 factors. What is happening is that, since 2000s BowQ is getting dual benefit of period index and SR. it is like BowQ * home-away * better period index * better SR. If you apply period index to Strikerate(as you are doing with BowQ) then only it is correct. or either you should do away with period index altogether.

    Some fine tuning needs to be done.

    are you re-doing entire tables? then i have one suggestion that can be used instead of Strike-rate. [[ Arjun Your follow-up comment did not come. Let me see what can be done. Is there really double-counting of strike rate. If so, my first inclination is to take away the period adjustment. After all this is a batsman analysis not a bowler analysis. So the period adj looks like can be dispensed with. s/r adj is essential since that is the only way two bowling attacks coming in at 30.00 on average basis with s/rs of 50 and 70 can be separated. Ananth: ]]

  • Abhi on May 21, 2011, 6:05 GMT

    Ananth Dead right! Srikanthk is particularly like a piranha. have to be as guarded and diplomatic as a politician with him around.

    Well ,for whatever reason LBW laws etc if these bowlers took relatively forever to get a wicket than this should be factored into the "period adjustment" The tables show: WW1-WW2 32.56 0.976 40s-50s 29.96 1.060 [[ Problem, Abhi, is that you are again over-stating things. 70 balls is not "forever". That is Shastri's or Emburey's or Hooper's 110+. 70 means, on an average, 8 wiclkets will be taken in a day and 40 in 5 days, ending in a result. So it is not at all bad. Ananth: ]]

    Reillys career from 1932-46. But the 40s -50s was apparently a bowler friendly period.

    We have a paradox. Perhaps SR should be introduced into the "period adjustments" too! i.e not just runs used. The

  • Boll on May 21, 2011, 5:28 GMT

    @Abhi re.[[If Bill O' Reilly had a strike rate of 70 - this means that he was a rather poor bowler.Period.]]

    Regarded by the greatest ever batsman as the greatest bowler he ever faced, and this is your conclusion?

    A test average of 22 (close to 800 first-class wickets at 16), an economy rate of 1.9 an over (averaged about 33 6-ball overs per innings), and a strike-rate of a touch under 70. Stunning statistics, particularly for a slow-bowler.

    Compare this to the oft-praised Bishen Bedi. Average 28, economy 2.1, SR 80. By your criteria, a pitiful bowler. Period? [[ Abhi, by now, must be regretting the full toss on leg stump he has bowled. To me the Bowling average is THE measure. Anything else is secondary. It is the most wonderful of cricketing measures since it incorporates the two measures, strike rate and accuracy in equal weight. You can ignore the bowling average at your peril. Ananth: ]]

  • Alex on May 21, 2011, 5:27 GMT

    @Ananth: I think the discussion on average and SR is assuming that everybody plays at the average level all the time. However, that never happens. O'Reilly may average SR=70 but in certain matches he will average SR=40 and in certain others he might average SR=120 ... unless you compute STD/variance in addition, we will never have a fuller picture on this. Also, quite a few times, Lara-SRT were taken out by a spinner within 20 deliveries ... you don't necessarily need a Donal/Steyn to do that.

    Spinners have the least thankful job in cricket since very few tracks give them an advantage (unlike fast/swing bowlers) ... when they get helpful tracks, usually media raises an outcry!!

  • Ananth on May 21, 2011, 5:16 GMT

    Antony I have uploaded the BQI table and the link is provided at the end of the article.

  • shrikanthk on May 21, 2011, 5:03 GMT

    The SRs of the most penetrative modern bowlers ARE infact the Best

    Agree. But you need to have a proper historical perspective. There is NO WAY you can compare the strike rate of a Gibbs with a Muralitharan! IT DOESN'T MAKE SENSE.

    Pad-play in the 50s/60s reached its zenith. Umpires were not as liberal with their LBWs as they are today. Batsmen could pad away anything posing the slightest risk. Moreover, the laws didn't liberally allow stuff like "doosra". [[ On 12 January 1964, when Nadkarni bowled his famous spell of 32-27-5-0, Barrington and Bolus padded up to more than 50% of the Nadkarni deliveries. Nadkarni was a wicket-to-wicket, non-spinning, non-flighting, deadly-accurate spinner. Today B & B would have been out lbw 20 times each. I remember, I heard the radio commentary, cutting college. Ananth: ]]

    To be fair to all eras, please consider averages. That is all. After all, averages do incorporate SRs. Isn't the Average a function of both economy rate and SR? [[ I responded to your earlier comment without reading this. Same request. Ananth: ]]

  • shrikanthk on May 21, 2011, 4:59 GMT

    If Bill O' Reilly had a strike rate of 70- this means that he was a rather poor bowler.Period

    Abhi: I don't like the finality of your tone. We're talking about a bowler here who was widely acknowledged by everyone as perhaps the greatest bowler of all time (of any kind) until the 70s (the only competition for that title is from Barnes). It's only post 70s that Lillee and the WI quartet changed the way we perceive bowling attacks by demonstrating the effectiveness of pace partly by slowing down overrates.

    It's not O'Reilly's fault that his SR is close to 70. The first half of his career was in an era of the Old-LBW law, when his googly was rendered TOTALLY ineffective as right-handers could just pad it away heedlessly. He was a great matchwinner. Renowned for getting great batsmen (like Hammond) out very early. He was perhaps the single most important reason why Australia retained the Ashes in 1938 (which was otherwise one of the weakest Aus attacks ever). [[ I am with you 100%. Bowling average is still the final measure of a bowler's quality. Strike rate is the more important component of this, that is all. Barrin g the pre-WW1 spinners like Briggs, Pates, Trumble, O'Reilly's average is bettered by only one spinner, Laker. Dismissing O'Reilly is like dismissing Trumper for poor strike rate or some thing similar. Ananth: ]]

  • Abhi on May 21, 2011, 4:29 GMT

    Srikanthk, That is a perceptive point. But again it is in a way a contradiction of your earlier stand on SRs and Ananths clarification on the importance of "balance" of bowling attacks (scroll up to check his reply to my earlier comment)

    If you have 2/3 great fast bowlers ( who get tired off early) and then force the "stock" bowlers to put in the yards...the Bqty will naturally suffer.

  • Abhi on May 21, 2011, 4:27 GMT

    Ananth, Thanks for the clarifications. All absolutely agreed. SRs have to be included in Btys (to what extent is debatable) Otherwise you get situations such as Steyn=Pollock. Don’t think too many batsmen would agree with that. [[ Yes that is correct. The s/r factor is only used to give Waqar edge over Pollock, Lee edge over Kapil, McGrath edge over Davidson and so on. Ananth: ]]

  • shrikanthk on May 21, 2011, 3:56 GMT

    My previous comment was intended to drive home an often overlooked point.

    In most innings, when a great batsman gets set to play a long innings, captains are FORCED to look beyond their best bowlers. Especially, in the case of batsmen who spend a lot of time at the crease like Hutton or Steve Waugh. This will obviously hurt the BQI of their innings.

    The BQIs of more aggressive players like Sehwag or Gilchrist won't be hurt to the same extent, since they don't spend the same amount of time at the crease.

    Another learning: Most clever players (like the Waughs) are very adept at taking on the best bowlers forcing captains to look elsewhere. A clever player doesn't look to face the "best" bowler all the time. He tries to take the best bowler off the attack. As a result, the BQI of his innings will suffer.

  • Abhi on May 21, 2011, 3:52 GMT

    Of course am using "scores"because they are more easily available than balls played.

    Here's Donald on Tendulkar : " As a bowler you have to have your gameplan ready before bowling to Tendulkar. Sachin looks to leave a lot of balls in the beginning, and as a bowler your best chance against him is in the first 20 minutes. During our team meetings, we often speak about the importance of the first 12 balls to Tendulkar. If you get him then you can thank your stars, otherwise it could mean that tough times lie ahead. "

    Essentially, what I am attempting to get across is a well acknowledged fact among the pros.

  • shrikanthk on May 21, 2011, 3:48 GMT

    Steve waugh's Total stats of 10000+ test runs @51.0 include only 1283 runs @34.0 against TOP group

    Arjun: Okay. That sounds revealing on the surface. But there's a LOT that it doesn't reveal.

    Lemme illustrate with an example. Steve Waugh's 200 at Jamaica in 1995 (along with brother Mark's 126, these two innings are probably two of the most important innings played by ANYONE in the past 20 years).

    Now, the WI attack comprised of Ambrose, Walsh, the two Benjamins and Hooper. Yet, this attack is NOT in the highest BQI group. WHY?

    The reason is that Carl Hooper bowled 43 overs in that innings! And we all know that Carl averaged about 50 with the ball!

    Now why did Richardson let Hooper bowl 43 overs? Because the two Waughs had the nerve to take on Ambrose, Walsh and Benjamins. Hooper was the only one who managed to stem the flow of runs. Now would you penalise the two clever Waughs for forcing Richardson to look beyond his "best" bowlers?

  • Abhi on May 21, 2011, 3:42 GMT

    So to round off!( regret so many comments- but typing as i think) If you had a bowling attack with Reilly and X- SR 70 and say 100 (avg 22) And another Steyn and Y - SR 39 and 50 (avg say 27) Which bowling attack would you chose vs. the Top batsmen?? No contest. By the time Reilly and co. get a wicket the Bradman's, Tendulkars,laras will be "IN" and then Sayonara. (Having faced more than 10 OVERS!!) [[ No, not at all. This is over-simplification. Steyn would bowl 7 overs and need rest for three hours. O'Reilly and others would bowl hours at a stretch and slowly but surely get their wickets. If that is the case, no top batsman should be out between 20 and 100. I myself know the value of striking early and have given an additional weight. However I would be a fool to think that an attack comprising of just attacking fast bowlers would always do well. If so, the West indian results of 1980s would have been far ahead of Australia of 2000s. That is not the case. A balanced attack will always do well in different places. An one-dimensional attack like Win of 1980s or Indian spin-based attack of 80s would find its Waterloo more often. Ananth: ]]

  • Abhi on May 21, 2011, 3:33 GMT

    If Bill O' Reilly had a strike rate of 70- this means that he was a rather poor bowler.Period. He may have done better in ODIs! If a candidate for "best bowler' (in any category) will on AVERAGE take 12 overs to get a wicket (compared to almost HALF for say a Steyn) this means that it WAS relatively easier to play him [[ No. Abhi I do not agree. Barring Murali, Warne and MacGill, most spinners have strike rates either side of O'Reilly's 70. Would you call Kumble, Harbhajan, Saqlain, Qadir, Bedi poor bowlers. 70 is fine and his average is great. He cost his team only 22 runs per wicket, better than Murali and far ahead of Warne. Let us not go overboard with this strike rate thing. 70+ is a problem only with an average of 32+ Ananth: ]]

    And as rgds. Modern day batsmen- If as you say if "It has everything to do with the less risk-averse approach of modern batsmen." then these guys should be willing to take on the spinners even more- Then why aren't the spinners strike rates on par with the traditional "strike " bowlers.

    Completely illogical arguments if you ask me.

  • Abhi on May 21, 2011, 3:25 GMT

    Further, in a previous comment.- What if you have say Steyn/Donald bowling just 12 overs in a 60 over inning? The rest bowled by tweakers – with a low average. Since the true strike bowlers will inevitably bowl short spells this means that the combination of them averaging slightly higher plus lesser amount of bowlers bowled REDUCES the Bqty of the attack. So, a normal situation leads to an abnormal Bqty. So, if you may get combinations wherein 2 great fast bowlers + 2 economical spin bowlers will almost ALWAYS have a better Bqty than say 3 fast bowlers+ 1 slightly expensive stock or spin bowler etc.

  • shrikanthk on May 21, 2011, 3:21 GMT

    I am very surprised to find the Pak attack against which Sobers scored 365 in '57 in Group 5 (the highest "quality" group). I know Fazal Mahmood and Khan Mohammed were world class. But surprising nonetheless. [[ Peculiar match. Good thing you raised it. Fazal then 82 at 21.56. Khan mahmood then 52 at 17.98. Both outstanding and full credit given. Two other bowlers at 37. But the key point was that Fazal and Kham Mohd bowled 140 of the 200 overs and this gave a good low value to the BQI which finally ended at 28.42. Ananth: ]]

    Sehwag's two triple hundreds fall in the top group. Having said that, both were on fairly placid tracks where you'd back him to score runs regardless of "Bowling quality". [[ Pitch related factors not there in this article. Ananth: ]]

    The other factor influencing BQI : Overrate Factor. Back in the 20s-30s, when teams were expected to bowl 110-120 overs in a day, the "best" bowlers couldn't bowl a large proportion of overs. The captain often had to rely heavily on the 4th and 5th bowler (resulting in BQI drop). [[ These are factors impossible to quantify. Ananth: ]]

    Post 60s, this has changed markedly. In an era where less than 90 overs are bowled in a day, the "best" bowlers (invariably fast bowlers with better averages and strike rates) can afford to bowl a greater proportion of overs in an innings.

  • Abhi on May 21, 2011, 3:20 GMT

    Srikanthk, There are several aspects to consider regarding strike rates; 1) The SRs of the most penetrative modern bowlers ARE infact the Best- Bond, Steyn, Waqar, Akhtar, Donald etc. Ridiculous to make statements “he better strike rates of modern bowlers has nothing to do with their "penetrativeness". It has everything to do with the less risk-averse approach of modern batsmen” These guys would take out the best anytime. 2) As mentioned several times- The critical aspect re. a top batsman’s innings is getting him out. Not merely restricting him .Once a Top class batsman gets in you are usually in for trouble. Look at the huge number of single digit/low scores for batsmen when playing grp5 players. After a Top class batsman gets “in” you may as well throw out the “average” (CTD or “career average”) of a bowler out the window. It will hardly matter. A bowler may give away an “average “ of say 25 runs/wkt for his “career” – But once the better batsmen are in, this number will in all probability change drastically. The toughest part of an innings vs. top class bowlers is the early part- then it gets progressively easier. 0-25 v.tough. 25-50 easier. After 50- another 50 very much on. For the guys with a knack of scoring big- then big trouble looms. 3) Would you rather have Steyn take out say Tendulkar, Lara and Ponting in his first 6 overs and in the process give away say 40 runs? Hell YEAH. Or would you have a Murali or Warne bowl 6 overs for 25 runs and maybe get one wicket? NO contest. Strike rate is a critical factor. 4) Am glad others such as Aparajith Shyam appreciate the value of this. [[ In one of my ODI Ratings analysis I explained to some one that a wicket-captured rating value has three facors attached to it. Basic wicket capturing value (30%), who was dismissed (30%) and when was this player dismissed (40%). He was flabbergasted. He said the player dismissed should get the highest weight. Then I explained that dismissing Tendulkar at 5 was far more important than dismissing him at 50 which in turn was far more important than dismissing him at 100. I think the "when" factor is ignored. A single situation answers this. Playing, arguably, his worst international innings, Tendulkar guided India into the WC Final and won the MOM. If he had been dismissed at a low score does anyone think India would have won. He being the batsman that he is made Pakistan pay. This defines the importance of dismissing a top batsman early in his innings. Abhi is absolutely correct. Dismissing a top player before he reaches 20 is very essential and this is where a fast bowler with a better strike rate is going to work well. How many time have we seen captains nowadays leaving gaps, conceding fours, with a view to get a dismissal. This in no way pulls down the spinners who have their own, often, patient ways of getting batsmen out. But we have to recognize and reward attacks with a penetrative edge. Ananth: ]]

  • shrikanthk on May 21, 2011, 2:51 GMT

    Purely based on strike rate a weight of 0.90 to 1.10 has been given. That is quite significant.

    Ananth: I think this partly answers my earlier question. I suspect this adjustment is playing a significant role in causing a lot of attacks to fall into the low BQI groups in the 30s/40s/50s.

    I'm not a fan of this adjustment. The better strike rates of modern bowlers has nothing to do with their "penetrativeness". It has everything to do with the less risk-averse approach of modern batsmen. Strike rates and economy rates have a strong negative correlation.

    Could you try simulating without this adjustment? I have a hunch this might improve the BQI of a lot of attacks from the 20s-60s.

    Take Bill O'Reilly for instance. A bowling average of 22. Strike rate of close to 70. This guy is a strong candidate for the title "Greatest slow bowler of all time". His strike rate tells you nothing about his aggressive approach as a bowler. To penalise him for his SR would be needlessly harsh. [[ No, Shri, this is not an adjustment as such. It is a recognition that the attack with an overall higher strike rates should be considered as a more difficult attack to face than one with lower strike rates. Indicentally this does not vary that much. Most values are around or above 1.00. The reason is that there is almost always a mix of bowler type and we are not talking individual strike rates. The first example I have given is an extreme. When do you have Wasim and Waqar bowl 28.2 overs. Even there the factor is 0.93. You will see in the other example, despite the presence of Steyn the strike rate factor is 1.05. So dropping this will have very little impact. Also I cannot make a change and present one alternate table. There are so may tables that I have to look at any change carefully. What I am going to do is to tweak the ctd below 50 wkt calculations to relate to the player's career average and re-do the tables. I would be re-posting the entire article at one shot. Ananth: ]]

  • shrikanthk on May 21, 2011, 2:41 GMT

    i just thought that a lot of bowlers of the yesteryears, although possessed with good averages were probably a little too patient in taking wickets

    Aparajith: Hang on. Nearly ALL bowlers had "ordinary" strike rates by today's standards prior to the 70s. This has mainly got to do with the careworn batsmanship of the earlier eras (when teams were often content with draws). This has nothing to do with bowlers being "too patient". Bowlers since Spofforth's day have always been too glad to take wickets as often as they can!

    Sobers had an SR GT 90. Had he been playing today, batsmen will take more chances against him and will also get out in the process. So, his average might still be the same. Economy rate will worsen, but the Strike rate will improve. Bowling average is a far more stable metric than the S.R when you compare players across eras.

  • Aparajith Shyam on May 20, 2011, 22:30 GMT

    In reference to my above comments, would it be worth considering, giving the strike rate a tad more priority in test cricket. This probably might change the tables only a little, but i just thought that a lot of bowlers of the yesteryears, although possessed with good averages were probably a little too patient in taking wickets. (e.g.Hugh Tayfield, Bishen Bedi, EAS Prasanna, U'wood, Mike Hendricks, Chris Old, John Snow) [[ Purely based on strike rate a weight of 0.90 to 1.10 has been given. That is quite significant. Ananth: ]]

    Out of curiousity, how would you compare the below two bowling lineups using your current method

    Bowler (average\strike rate)

    Mike Hendricks - 25\71 Chris Old - 28\62 Tony Greig - 32\67 U'Wood - 25\73

    as against

    Dion Nash 28\66 Chris Cairns 29\53 Daniel Vettori 33\76 Simon Doull 29\61 [[ Who bowled how many overs. Assuming equal number of overs and other facors being equal, the first team would have a BQI of 27.75 (27.5 x 1.01 for sr). The second one will have a BQI of 29.90 (29.75 x 1.005). Both have similar strike rates (68 and 64). Ananth: ]]

    Please let me know if i have ignored anything in your calculations.

    P.S: Examples given above are just to emphasize on the strike rate & not to bring down any bowler.

  • Aparajith Shyam on May 20, 2011, 22:27 GMT

    Hi Ananth,

    A Very good compilation indeed. I guess a similarly good analysis was done by David Barry way back in 2008. i guess his analysis did not bring into picture the Strike rate of the bowler. Its nice to see this version include it. I have a question though...Does the metric, that is used to normalize the STRIKERATE of a bowler,ensures that bowlers with high strike rates are penalized enough. I focus specifically on the strike rate because, in test cricket,a modest average-good strikerate bowler (e.g Malinga) would probably be a better bet (under similar playing conditions) than a good avg-modest strike rate one(e.g John Snow). The opposite could be true for one-day cricket (e.g.Wasim Akram vs Dale Steyn).

  • DaGameChanger on May 20, 2011, 21:48 GMT

    I always thought Javed Miandad was average to decent player but not a great one or close to it. This proves FOR SURE proves it and also provies, Pakistan never produced any great batsman even Hanif Mohammad have terrible average against good top teams and even Inzamam has manage only 35ish. Also strange to see Martin Crowe there. [[ No. I suggest you do not generalize like this. As already explained in these columns the lack of runs against the top attacks may have nothing to do with a player but a sign of times. Inzamam has scored nearly 50% of his runs against the top 2 bowling attacks. Why is it strange to see Martin Crowe here. He is the best New Zealand batsman ever. Ananth: ]]

  • Pawan Mathur on May 20, 2011, 18:52 GMT

    This was a really good read. My comment however is not specific to this article. i do feel that when you refer to a particular test match (in most of the cases by No.), there should be a hyperlink directing to the scorecard page on cricinfo. and another thing that i want to know is that in one of your responses earlier in comments section, you had mentioned that you will do an analysis of an IPL player performance and his price. IS this coming shortly? [[ Pawan I have already mentioned this. There is no correlation between the Test match Ids and Cricinfo's scorecrd ids. So it is quite difficult for me to provide the link automatically. It may also be difficult for Cricinfo staff who publish my article to provide the link. Only soluttion is for me to upload all my scorecard files to my website and provide the links. Massive task, though. Ananth: ]]

  • Deep on May 20, 2011, 16:40 GMT

    I thought Sanga was a flat track bully and Miandad was pretty good. This is indeed an eye-opener. Sanga and Sehwag are far better batsmen than people give them credit for. Miandad's 27 average against top bowlers is pretty shocking. Viv's average of 29 against the weakest bowlers seems hilarious! [[ Probably a new insight into Richards. He could not really care unless otherwise the iopposition was good. Sangakkara's top-4 innings against group 5 attacks were 232, 192, 156* and 100*. Ananth: ]]

  • arch on May 20, 2011, 16:22 GMT

    @Alex 4. Viv's ave=29 vs Group 5 need not mean that he lacked motivation vs sub-standard bowling. He might have faced them a lot after '86, by which time he himself was no longer great ... maybe the pitches had something to do as well. Likewise for others.

    Viv's 29 average group 1 sample is only for ten tests. That is a small sample and what you say may very well be true, or not. Either way, his groups 2 and 3 averages (a large sample) are well below his group 4 and 5 averages. In fact, in this study, he is the only one who seems to progressively improve against better attacks. I think either Holding or Imran said here and there that he used to get bored and get out, and that he never scored a triple century because he did not need to and that he did not bat for his average. Perhaps this study illustrates that. In any case, his stats rack up differently from the rest and that is itself worth looking into.

  • Gerry_the_Merry on May 20, 2011, 16:14 GMT

    Ananth, fine. 1) I understand you have made some adjustment for career beginning averages and that you think some more improvement can be done later. That is fine 2) what about career ending - a 1991 attack of (hypothetically) Marshall Hadlee and Imran may have been preferable to facing 1991 Reid, Hughes and McDermott but would have scored very highly on bqi. Can you cover that impact through a recent form type of weightage? 3) on bqi, let me ask the question again more simply- why AT ALL do we need period adjustment? a bowler friendly period surely means tougher time for batsmen and more innings in level 4 or 5 4) can you please give home and away averages in all the above tables ? Typically for indian, paki and sl batsmen, there have been vast home away deltas. [[ If you see the correspondence with Abhi/David, dynamic recent form is a huge thing and is on the cards. Will incorporate that before the next major study of batsmen. Quite tough to give home and away averages since I have developed the same together, I have to change the method of calculations completely and the numbers would be quite small in some cases. The bowling quality is an independent measure and needs to have period adjustment built in. Ananth: ]]

  • shrikanthk on May 20, 2011, 15:53 GMT

    One of the things that's perplexing me... The mean bowling average of the inter war period : 32.56 as opposed to 34.94 for the present era (2005-2011).

    Yet, the proportion of innings accounted for by Group 1 is far greater for the inter-war batsmen than for the modern greats. How do you explain this? [[ Yes the New Zealanders, Indians, West Indians played many tests and were all placed quite low. That situation has not happened during the 2005-2011 period. even Bangladesh was 5 years old when this period started. And reasonably settled bowling sides now. During 20s/30s, only Australia and England had reasonably settled bowling sides. even South Africa was finding its feet. Ananth: ]]

    One reason is ofcourse the much-discussed CTD conundrum. The less you play, the more the volatility of CTD. The 32.56 figure is based purely on aggregate figures, I presume. Aggregate runs/Aggregate wickets? Whereas the BQI is based on CTD. I can't think of any other reason for this anomaly.

  • Antony on May 20, 2011, 15:14 GMT

    Great work. Is it possible to see the BQI lists to see which are the best rated bowling attacks of these eras were? [[ I will try and give the group 5 bowling attacks if not the full list. Just have to make the list readable. Now I print it in a cryptic manner for my verification purposes. Ananth: ]]

  • Alex on May 20, 2011, 15:03 GMT

    @Arjun: You are right on Waugh. Like a dyslexic, I took SR=42 to be SR=24. S Waugh managed only 2 50+ scores in the 4th ... one was a sedate 67 vs NZ when Oz could have won it and the other was his final innings of 80 that saved the match vs Ind. S Waugh became a much better batsman in his final 10 years, starting @1993 ... however he continued to perform poorly in the 4th!

    @Ananth: Oz attack of '00-'01 has the rating of 20.92 (Test #1527): it had McGrath, Gillespie, Lee, & McGill. The WI attack of '80 has the rating of 27.24 (Test #890): it had Clarke, Marshall, Garner, and Croft. I do think the Aussie attack was at its peak @'00 but dare say that the '80 WI attack was better. Yet your metric puts the Oz attack far superior to it ... any way to resolve this anomaly? [[ Australia: McGrath (21.95), Gillespie (20.22), MacGill (24.78) and Miller (25.07). This comes to 23.47. Multiplied by 0.947 (period adj) and 0.95 (home adj) and 0.96 (s/r adj) takes us to 20.21. West Indies: Garner (19.98), Croft (23.10), Marshall (<50wkt), Croft (<50w) leads to 26.10. Multiplied by 0.991 (period adj) and 1.05 (away adj) and 1.00 (s/r adj) takes us to 27.24. The same team, playimng at home, would have been .24.57. Not bad for a team with two novice bowlers, granted one of them is Marshall. So there is no anomaly. Ananth: ]]

  • bks123 on May 20, 2011, 14:46 GMT

    Still glancing. Need to spend more time. But at the first glance it seems sachin, lara, ponting, dravid, kallis the 5 grt batsmen of our generation have fared equally well against top quality bowling. dravid's ave 37 in grp 5 is well compensated in the next group with an ave of 57. All of them averages ~50 against quality bowling. some scored more runs as they played more innings. nothing much to choose from. Sehwag is exceptional. Some interesting features on the quality of Sehwag, steve wough and sunil gavaskar and the quality of bowling attacks they mostly scored against have come out. Intersting.

  • Arjun on May 20, 2011, 13:52 GMT

    Some observations

    M Taylor avegs. just 26.0 (top group) M Atherton 23.0 G Hick 23.0 may be they faced quality fast bowlers too often and got dissmised cheaply. Martin Crowe's aggregate of 292 runs in 15 innings includes single inn of 188 runs. In other 14 inns he scored 104 runs.

    C hunte's break-up stats across 5 groups are reversed.

    Chris Cairns is an intresting case. he scored 2000 runs @41.0 against top 2 groups combined. and 1200 runs @25.0 ag. bottom 3 groups. He excelled against top quality stuff, now we know what we missed because of his early retirement.

  • shrikanthk on May 20, 2011, 13:18 GMT

    The other thing I'm noticing is the clustering of innings in Bowling Quality groups, especially in the earlier eras.

    Eg: Bradman again. The 32-33 innings are most probably in Group 2. English attacks from '34 Oval test till '36-37 Melbourne test are in Group 5 (highest quality group). '36-37 Adeleide onwards - the English attacks get demoted to Group 4, mainly because of Bradman's heavy scoring in that test which hurt the averages of those bowlers.

    Now, I'm talking about the same set of bowlers throughout this period (Allen, Verity, Bowes, Voce). But they move in and out of Groups very often because of the CTD phenomenon and the small sample size. I'm not saying that's a bad thing. Just stating the reasons why Group 5 is so small for 30s-40s batsmen.

    The same set of bowlers (Allen, Verity, Bowes and Voce) today will have more stable CTD averages (possibly in Group4) because of the volume of test cricket played.

    The more tests you play, the less volatile the CTD.

  • shrikanthk on May 20, 2011, 12:52 GMT

    Unfortunately body-line series is an anamoly

    Ananth: I strongly believe that the Bodyline innings are falling in Group 2 (BQI : 40 to 45). I know why that's happening: [[ On the contrary, one test is in group 4, two are in group 3 and two are in group 2. I would personally put all these tests in group 4 because of the intimidation. Ananth: ]]

    - Bodyline was Larwood's last tour. His CTD average was much more than 30 at the beginning of the series. It kept improving throughout the series. Can't be helped.

    - Verity, Bowes and even Allen were at the beginning of their careers at that time, possibly with ordinary CTD averages. If you could use notional values for them (based on their very good career averages) like you did for Warne, I'm sure Bodyline innings of DGB will move to Group 2 if not Group 1.

    Just an explanation of things the way they are. I'm not asking you to do anything about it.

  • shrikanthk on May 20, 2011, 12:44 GMT

    At long last an article that shows the Don in poor light!

    !? Poor light? The man is averaging 81 (the highest against the "Group 5" bowling attacks. 121 against the Group 4 attacks! Is that poor?

    Now, let's look at some of Bradman's 9 innings against the "best" attacks.

    270 at Melbourne - an acknowledged masterpiece.

    244 at the Oval - one of his fastest double hundreds that clinched Aus the Ashes in '34. His SR for this innings was 90!! Moreover, he nearly died a few weeks after this effort due to appendix complications.

    82 : A second innings effort in '36-37 on a difficult wicket I think (not 100% sure)

    77 : In the same Oval test in which he scored 244

    38, 0 and 0 : All these innings were in the first 2 tests of '36-37 when Australia was caught repeatedly on wet wickets.

    So, it was not just the attacks. Even the conditions conspired against him in most of these 9 innings. Yet, the man averaged 81. Enough said. [[ Thanks, Shri for taking the trouble taken to the fifth table and extract the innings. Only after I posted my response to Mahender did I realize I should have done it. Pitch conditions are very important and I am moving towards it. Ananth: ]]

  • shrikanthk on May 20, 2011, 12:29 GMT

    you have to realise that back then each NZ first class team played 3 matches a year - so players like Badcock and Dunning who both played over 50 fc games did so largely on tour in England

    Good point. Infact, the SA cricketer Ian Smith I talked about even topped the SA bowling averages on a tour of England. We learn about a lot of early Indian/SA/NZ cricketers mostly by looking at their English tour FC records. Now, Ananth Sir. Don't get me wrong. Nothing you can do about it. All I'm saying is that Ian Smith is probably more of a 35-40-ish bowler as opposed to a 64 runs/wicket bowler!

    If you think I am expecting such, often shallow, appreciative comments, kindly have second and third thoughts on this

    I never thought so. But if all of us have to confine our comments to the realm of the "possible", this will be a poorer thread. Ofcourse, it may not be possible to incorporate FC records for Test cricketers with small no of tests. But what's the harm talking about it. [[ Only my limited one-person resources !!! As I told my wife the other day, normally we run after one train, now I feel like we are running after three trains. Ananth: ]]

  • victoria on May 20, 2011, 12:06 GMT

    Ananth, so why Cricinfo/ESPN did not hire some excellent statistical experts such as you and others, to first carry out some scientific researches like this, in order to pick the most creditable 'All Time World 11' in cricket; instead of selecting a few inappropriately informed guys, who were bound to base their assessments and judgments on sentiments and information of a rather subjective and dubious nature (eg. favouritism, friendship,insularity, numbers at face value, personal likeness/dislikeness, etc; in order to come up with 'their' team? If the result of the exercise was the product of a process such as this, it would have reduced by far, the tremendous volumes of worldwide criticisms, skepticisms and outrage that we experienced when that team was posted. This kind of wok must be commended; because it does not only help the world and fans not to be easily fooled by the bare outward appearance of figures; but it also gives true justice to every player who is up for scrutiny! [[ Maybe I was not included in the panel because I am only a simple cricket analyst with no name and public recognition !!! Seriously I would not blame Cricinfo. Their method brings in the peer-recognition factor which is very important. It is also assumed that most of the selectors would have enough analytical skills to know the true value of player performances. Ananth: ]]

  • Ananth on May 20, 2011, 12:05 GMT

    Shrikanth/Gerry I did a simple sum of the top 100 bowlers, based on the wickets captured. These are the bowlers who really matter since the number of wickets captured is 132 and above. The average number of wickets captured per bowler is 250.5. That means the early career tweaking is applicable only to 20% of all the tests played by these bowlers who matter. Even for these, I would take steps to strengthen to have a clear link to the career averages. However even now for 80% of the tests, the actual meaningful ctd values are used. Ananth

  • Mahendran on May 20, 2011, 11:14 GMT

    I do not know why First Class figures have to taken into consideration when we talk about tests. As Ananth said it is highly unreliable. As Sachin Tendulkar (in the capacity of having played the most number of tests) himself said that after playing one full season of 1st class cricket and was thrown into the Test arena and had to face Waqar, Wasim & Imran in his first test...he felt that he didn't belong. Test cricket is totally different according to him.

  • Arjun on May 20, 2011, 11:11 GMT

    Alex. Steve Waugh made only 613 runs @25.0 in 4th innings of a test match with SR of 42.39. Hence he lasted only 59 balls and not 100+. Infact he was equally bad in 3rd inn averageing just 35.0 with only 2 100s in 64 inns. His combined 2nd innings record is worst of all the batsmen who have made more than 6000 runs. 94 inns, 2369 runs, @32.0 with only 2 100s(other 30 100s in 1st Inn.)

    Abhi. you are right in saying "As with all stats it pays to use a pinch of salt and keep in mind that stats are like bikinis..." Steve waugh's Total stats of 10000+ test runs @51.0 include only 1283 runs @34.0 against TOP group. He averages 3 times against 45+ group as compared to TOP group. All are actual/unadjusted figures. These are facts, it cannot be changed. This article is not to downgrade any of the great players; infact it reveals everything and does not conceal anything. We can make our own observations/derive our own conclusion. Where else are we going to get this type of information?

  • Mahendran on May 20, 2011, 11:07 GMT

    At long last an article that shows the Don in poor light!! This may be rectified when you bring the pitch quality into equation - poor bowling quality in bad pitches is always better than good bowling quality in batting pitches. But there are some exceptions like the great Marshall for whom the pitch quality doesn't matter. But we have to admit that he will be a terror in bad pitches. So pitch quality does matter. Can you please conjure some trick to bring this factor into contention? You might be already having it up your sleeve. [[ Mahendran, please see Anshu Jain's comments. You cannot hold Bradman responsible for not having to face more of group 5 bowling attacks. It certainly does not show Bradman in poor light. rather it is a reflection of the not-so-great bowling strength of the other bowlers during the 30s-40s. Unfortunatelu body-line series is an anamoly. Ananth: ]]

  • AD on May 20, 2011, 10:41 GMT

    2) The Form factor would be more of a critical element in the Best individual innings lists which you compile from time to time. Over hundreds of matches this would even out and so have minimal bearing in this analysis. [[ That is a valid point, AD. Since we are dealing with groups, the impact will be less. Let me also say that you should not feel that I have relegated the Pitch type into the background. I have never underestimated the importance of that factor. The only reason why I did not want to look at that now was the complexity of this article itself. You can already see that this itself requires tweaking in the area of early-ctd values, recent form etc. But I will do one for Pitch quality later. Whether it is a composite one involving both or only for the Pitch type let me see. Ananth: ]]

  • AD on May 20, 2011, 10:40 GMT

    Ananth: 1) This piece is a Classic. And richly deserves the encomiums coming its way. I particularly like the nuances – period adjustments, strike rate incorporations, home/away factors etc. They bring the stats to conform more closely with cricket as we actually know it. Many of us have often been left scratching our heads when confronted with career weighted bowling quality indexes for batsmen with long careers.

    So, the acknowledgement of the finer points, skill and execution of this piece are what make it unique.

    There is an additional request. Would it be possible to place the average BqtyI per group faced for the individual selected batsmen? For example –Next to Sehwag’s figures for Group 5- an average BqtyI figure that he faced for that particular group. I tried to import the file to an excel sheet. However, I get several columns fused into one and so cannot figure this out individually.

  • Abhi on May 20, 2011, 9:59 GMT

    Gerry,Ananth Just struck me- Another reason SR absolutely must be given its additional due ( besides the obvious fact that getting a top batsman out is of far greater priority than conceding a handful of runs) is that in the BQTY model used the ACTUAL amount of overs bowled by the various bowler has been used (as is logical)

    So, obviously a fast bowler will in general bowl many fewer overs than a tweaker who may plod along all day long? As such if only averages are used the issue would become totally lopsided.

  • Anshu N Jain on May 20, 2011, 9:43 GMT

    contd...

    Will gladly invite comments/insights on this method of looking at performances, from readers (and you Ananth!) who have better knowledge of Statistics! Once again, superbly executed Ananth!

  • Anshu N Jain on May 20, 2011, 9:43 GMT

    • Only looking at players with at least 20% of their Completed innings against a Group 5 attack, Richards is tops with a ratio of 1.1, followed by Sehwag at 1.05 and Lloyd at 1.02. Sobers has a ratio of exactly 1. Underachievers here are Miandad (0.51) and Andy Flower (0.58). • Extending the above, and looking at players with at least 50% of their Completed innings against a Group 5, 4 or 3 attack (Hobbs, Hutton and Hammond miss out), Bradman is tops with 1.07, followed by Richards at 1.03. Flower and Laxman are at exactly 1.00. • Looking at a simple Standard Deviation of the ratios across the 5 Groups, Laxman has the lowest figure of 0.086, testament to his consistency across all Bowling attacks! Next is Lloyd at 0.096, followed by Sehwag at 0.105. • Richards and Tendulkar have similar figures of 0.20. • Vaughan (0.852), Hutton (0.772), Flower (0.695) and Waugh (0.600) are the only players with a figure of more than 0.5. contd...

  • Anshu N Jain on May 20, 2011, 9:41 GMT

    For each player, I looked at the ratio (% of Runs scored)/(% of Completed innings) across each Group of BQI (5 through 1). That shows some interesting results; a few examples below: • Tendulkar's ratios trend - (Group 5 through 1)- 0.85, 0.86, 1.03, 1.15, 1.34 - is the ideal trend for a typical test batsman; Run making gets progressively easier as Bowling quality declines. Surprisingly, no other batsman in your selection has the same uniform trend! This could, in part, be attributed to Tendulkar having had the most widespread exposure to varied quality of Bowling attacks over his 21 year career. [[ This shows a very high level of consistency. Probably rounds off Tendulkar's career. Ananth: ]]

    • Only looking at players with at least 20% of their Completed innings against a Group 5 attack, Richards is tops with a ratio of 1.1, followed by Sehwag at 1.05 and Lloyd at 1.02. Sobers has a ratio of exactly 1. Underachievers here are Miandad (0.51) and Andy Flower (0.58). [[ This is almost the same as the ratio of group 5 average to their career average. Sobers 57.77 to 57.78, Richards 55 to 50 and so on. Ananth: ]]

    contd...

  • Anshu Jain on May 20, 2011, 9:40 GMT

    Great work as always Ananth! Well done!!! I find it erroneous to make comments like Player x scored 30% of his runs against a Group 2 attack and only 5% against a Group 5 attack, and thereby implying his batting quality. This figure has to be weighed in against the % of innings the Player has played against each Group (5 through 1). After all, its not Hutton’s fault that he came up against a Group 1 bowling attack in 60% of his test innings! [[ A very valid point. Ananth: ]]

    I tried using your table for select Test players' performance against each Group to work out some numbers of my own. Notwithstanding errors due to small sample sizes, and my rudimentary knowledge of statistics, I have taken the liberty to draw out some numbers, and then their implications.

    Contd...

  • Gerry_the_Merry on May 20, 2011, 7:37 GMT

    Ananth what about the other question? your bqi adjustment based on whether the period was a bowler or batsman friendly period is in my opinion incorrect. If from (say) '92-96 Ambrose, Warne, McGrath, wasim, waqar, bishop, Walsh, Donald, et al made life tough for batsmen, why should they be penalized and a sort of mean reversion done? [[ My Period factor is for periods as long as 35 years between 1914 to 1939 or as short between 2000 and 2004. Within these years all performances will be adjusted down or up. There does not seem to be any other way to adjust for the periods. Ananth: ]]

  • Alex on May 20, 2011, 6:42 GMT

    @Ananth: Great article. Form factor is important, easily determined and hence should be included. A few observations. [[ "Easily determined". Ha !!! Not really. It has to be a rolling value and has to be dynamically determined. So a new database segment has to be created and limnked to the main database. Fixed periods or number of tests is no great problem and is already available. Ananth: ]]

    1. Not clear whether runs vs Group 1-2 were on placid/lively tracks ... an important factor. [[ When I bring in thye Pitch type more light will be thrown on this. Ananth: ]]

    2. Stewart (15 100's & 45 50's at ave=40) has 62% of his runs vs the top 2 and a sensational 13% vs the bottom 2. Add wkt-keeping load ... this man earned every bit he got.

    3. Hobbs-Hammond-Hutton scored >66% of their runs vs the bottom 2 (76% for Hammond!).

    4. Viv's ave=29 vs Group 5 need not mean that he lacked motivation vs sub-standard bowling. He might have faced them a lot after '86, by which time he himself was no longer great ... maybe the pitches had something to do as well. Likewise for others.

    @Arjun: S Waugh might have averaged only 28 in 4th but those were at SR = @24. So, he still batted for 100+ balls on average, which is still decent as a defensive effort ... a bit similar to what RD has been doing for a while now vs good attacks on decisive pitches.

  • craigmnz on May 20, 2011, 6:39 GMT

    As a New Zealander the problem I have with the discussion of our attack in 1933 is the comment that it was below the level of a college 1st XI. I appreciate that the very limited number of tests we played in those days (14 between 1930 and 1945) presents a statistical problem. However you have to realise that back then each NZ first class team played 3 matches a year - so players like Badcock and Dunning who both played over 50 fc games did so largely on tour in England. So raising questions about the quality of our fc game doesn't really help either. Even today NZ fc cricket (with 6 teams) comes from a total population base the same as Sydney - suggesting that it should be as good as Sydney grade cricket. I've no idea what the standard was like in the 1930s but arguably the NZ team of the 30s was at least a match for fc counties in England, if not really of test standard. Might I suggest that pre-war performances by NZ be treated the same way as 19th century matches and ignored? [[ Craig, my apologies if I have in any way put down New Zealand cricketers, for whom I have the utmost regard. The only problem was the magnitue of Hammond's score and the fact that the New Zealand bowlers were so thin on the surface. I underatnd what you say. However having started Test cricket these problems are bound to be there and my apologies a second time if I have offended you guys. Hammond scored 217 against India. The Indian bowling attack had a BQI of 47.7 as compared to the New Zealand BQI of 51.3. So there is not much difference and my program has treated all newcomers in the same manner. We shoudl not ignore any Test match but accept and underatand the fact that the starting teams would have no great bwoling attacks and consequently the batting performances have to be downgraded. Ananth: ]]

  • Arjun on May 20, 2011, 6:36 GMT

    Every batsman plays very well against certain teams eg. gooch ag. WI, Laxman ag. Aus, for reasons unkown. Similarly every batsman has a certain bogey teams. Before this article we made our assumptions based on teams they played against. eg. for gooch, atherton we assumed they played very well ag. tough bowling attacks. But now we can make our own observations by looking at performance by different bowling groups instead of diff. teams.

    If a batsman avgs. over 60.0 in 40+ and 45+ group then it is ok but his performance in TOP group is the real deal. I believe Batsmen averaging above 40.0 in top group(atleast 30-35 inn. played) is serious stuff.

    Just to get an idea how good is the TOP group, look at bowling Index of SAF in the match at capetown in which sachin scored 169 and added 200+ in partnership with Azhar. It comes into 3rd group(35.0-40.0) inspite of donald/pollock bowling 47 overs(out of 92) in the innings; other bowlers were Klusner, B.Mcmillan, P Adams and McMillan bowled the rest.

  • rizwan on May 20, 2011, 6:17 GMT

    Ananth

    The definitive article on batting stats.Take a bow.

    I have been waiting for such an analysis for a long time and thank you for obliging.

    The analysis clearly seperates the men from the boys and the myth from reality.

    I am glad that special guy Laxman gets noticed and hope the rest of the media acknowledge VVS's outstanding contribution to India. The surprise package is Vaughan, where did he come from ? It must be those 190 s he scored against the aussies.

    The tables also indicates that Waugh and Miandad were flat track bullies ( an accusation levelled by Imran Khan on Miandad)

    It will be interesting to see the bowling equivalent of the above analysis.I am certain it would not be easy to crunch the numbers but it would be useful to know the bowlers who tamed the big guns.I am sure it would be quite an eye -opener and I doubt Murali and Warne would be top of the pile.

  • Abhi on May 20, 2011, 4:52 GMT

    Srikanthk Frankly, I honestly think Ananth has put up an excellent article. And truly busted some long cherished myths.

    There will always, always be tweaks here and there.

    The whole point is that borderline differences in averages tell us something. And 5/10/15 % differences in averages tell us something different altogether.

    THAT is what this article is essentially about.

    The CTD stuff, form etc etc is frankly Old Hat..and an ongoing process.. HOw about you use your incisive mind to glean something from the above tables?

  • Abhi on May 20, 2011, 4:49 GMT

    Ananth, Yup,Thanks you've told us that before. No hurry - frankly i dont think it will make an earth shattering difference, just an additional tweak. [[ Batting form is rather easy. Bowling form is tough since there are two factors. But will move the whole thing moving forward. Unfortunately the work I did on streaks was done in a straight-jacketed manner, allowing few derivatives. Otherwise that could have served the purpose. Ananth: ]]

  • Matt Jane on May 20, 2011, 4:43 GMT

    Ananth, I have always wondered what kind of records would Gavaskar and Tendulkar, Dravid,etc. have owned if they played against India? Is it possible to create such hypothetical data for top batsmen using this analysis? If not, is it a worthwhile exercise? [[ No simple solution to this other than the whole Simulation works. Ananth: ]]

  • shrikanthk on May 20, 2011, 4:22 GMT

    If you thought that I would have valued Hirwani's ctd value at 8.5 or Massies's at 8.51 in their second tests, you have a poor opinion of my cricketing and analytical abilities. That is where people like Alex, Abhi, Arjun et al differ. They KNOW that I would have taken care of such situations. Hence even when I do not give the explanations, they accept the valuations

    I never asked for a FC database! Just jotted down some thoughts on some areas where purely using Test figures may not work optimally. [[ Then why bring in FC when you know that it cannot work. By now you know very well the rigid framework within which I work, which is the Test database. Period. And this is by choice, for purposes of verification and credibility. This has been established over 15 years of data creation and maintenance. Anything and everything within this framework is fine. Half of your recent comments have been on FC figures. Somehwere I have to respond. You are correct to say that Warne's early-ctd figures must bear a relationship to his career figures. Fine. I have done that in a nominal manner. Can be improved to make it more algorithmic. Concentrate on that. But you are wrong in saying that Ian Smith's test average is 64 but he is a much better bowler than that since his FC average is 22. How can this be true. How many Indian FC players are there with 50+ FC figures and Test failures. Tell me, what do I really take Ian Smith's test average. You have a great insight into the game. However you must learn to know what can and cannot be done. Concentrate on the Warne/Murali early career situation since that is a very valid point. Forget about the Dunning/IanSmith situation since there is no answer, other than what has been done, at least in this blogspace. Ananth: ]]

    Why should every remark of mine be taken personally? I'm not the sort of chap who'll chip in with pointless inane comments like: "Great work Ananth. Brilliant stuff. Shrikanth". [[ I am sorry to say, your comments are bordering on the nonsensical. If you think I am expecting such, often shallow, appreciative comments, kindly have second and third thoughts on this. Ananth: ]]

    Maybe that's what I should do more often in order to make friends! If I had a poor opinion of your work, I wouldn't be here in the first place.

  • Gerry_the_Merry on May 20, 2011, 4:07 GMT

    One more issue is that your bqi adjustment based on whether the period was a bowler or batsman friendly period is in my opinion incorrect. If from (say) '92-96 Ambrose, Warne, McGrath, wasim, waqar, bishop, Walsh, Donald, et al made life tough for batsmen, why should they be penalized and a sort of mean reversion done? Also, don't spinners get penalized due to the strike rate factor? Thanks. [[ Let us agree that on a normal pitch the batsmen would rather face Warne/Murali than Waqar/Wasim. This effectively takes part care of that. Ananth: ]]

  • Abhi on May 20, 2011, 4:00 GMT

    David, All very good points and put in a much better manner than i normally manage. You are effectively looking at a "form" factor which has been in the pipeline for a while now to judge the quality of bowling a batsman was "actually" facing. [[ Abhi/David One thing I can assure you is that it WILL be done. I need a clear lengthy period to see how it can be dynamically done. Easier sai, but creation and an on-going maintenance of the database is a nighmare. I have done some work on fixed segments but to make it more dynamic requires some more tough work. It has never been forgotten. Ananth: ]]

  • shrikanthk on May 20, 2011, 3:31 GMT

    Until a bowler reached 50 wickets he does not get credit for ctd average, if it is low

    Very good. By the same token, he mustn't be credited with the CTD average, if it is very high!!!

    That's because CTD is bound to be volatile in the early tests of ANY bowler.

    It's always better to factor in the career average in some way in conjunction with the CTD average. Ideas anyone?

  • shrikanthk on May 20, 2011, 3:24 GMT

    Bradman has a 81+ average but has scored only 737 runs.

    Interesting. But I suspect most of those 9 innings were in the Bodyline series when he was up against Larwood, Bowes, Voce and Allen, when the attack he faced was downright immoral.

    He might have topped 1000 with an average of over 100 against the same bowlers if they had bowled conventional stuff.

    Or am I wrong? Were those innings in a different series? Going by averages, the 8 bodyline innings he played must be either in BQI LT 30 bucket OR BQI : 40-45 bucket.

    If it is the latter, there's something missing. An attack of Larwood, Voce, Verity, Bowes and Allen can't have a BQI of 40-45! It can happen only because Verity and Bowes were starting out then, though they rank among the great bowlers of the 30s. Hope you've used notional averages that are close to Career figures instead of CTD averages for those two in the Bodyline series. [[ You could refer to the accompanying table. Most of these innings were after body-line. Body-line attacks were not valued high. The numbers were quite average. But I cannot help it. Ananth: ]]

  • Gerry_the_Merry on May 20, 2011, 3:19 GMT

    Ananth, great work again, though I have some problems with the treatment of bowlers at the beginning and end of their careers. Am on the road for 20 days and can't look at the detailed tables. One request. What would Viv Richards break up (categorywise % runs and avg) have been if you had considered his career only up to March 1989?

  • Abhi on May 20, 2011, 2:30 GMT

    Srikanthk You will never be able to get an "exact" bowling quality figure. If it is OK within 5 % or so i think the job is done. If someone averages say 35 vs grp5 bowlers and someone else 40/45 then we can deduce something.

    I too was a bit surprised by some of the figures (just one eg. I expected Tendulkars 169 vs SA to have a better Bqty than the 160 vs NZ (though a classic)). [[ More than half the overs in the SA innings were bowled by four average bowlers (not Donald/Pollock). That lowered the quality to 36, still pretty good. Of the 5 Nzl bowlers, two were below 30 and three just above 30. No real weak spot. Both home. Nzl-better period adjustment. Ananth: ]]

    But,then overall we get a pretty good idea- and thats what it is about. This is not exact science...it is most definitely part ART as well..and we need to appreciate that.

  • Abhi on May 20, 2011, 2:25 GMT

    @Ashish Mehta I think you are making a big mistake on the RPI vs Average issue. I had posted this on the other blog: "Getting a top batsman out early is the most critical part of his innings. Not restricting him. After a top flight batsman is "in" he will inevitably make you pay- averages of bowlers notwithstanding. For eg with Tendulkar and Bradman. Once Tendulkar gets past 25, he has 13470 @ 98. (72 % better than his career avg.) Bradman 6824 @ 145. (45 % better than his career avg.) This similar trend applies to practically all top batsmen."

    If a great batsman is around after he scores around 20/25 you are almost always in for trouble.

  • shrikanthk on May 20, 2011, 2:07 GMT

    Who did Dunning bowl to. What was the level of New Zealand cricket. In New Zealand what is 26.

    Let's not assume the worst just because we are ignorant of something.

    Until they reach 50 wickets I do not take the CTD average. i take a notional average which also depends on how their career turned out

    Ok. This is more interesting. All I am suggesting is a way to use a methodology for the same based partly on FC averages instead of using a purely judgmental notional figure.

    Thanks

  • shrikanthk on May 20, 2011, 2:02 GMT

    a career of 132 tests will get full credit. Even earlier they will get more credit for facing Warne than Blewett even if Blewett's ctd is better than Warne

    Ananth: Could you elaborate on this? You might have explained this in an earlier article. I simply based my comments on this line - [[ I am not going to elaborate on the methodology more only because it will lead into more discussions on the numbers used. Suffice to say the following. 1. Ctd is used as it is only when a bowler crosses 50 wkts. 2. Lower than 50 wickets, there is a graded notional values such as 40/50 etc. 3. The bowler who finishes very well will be treated different to the Dunnings. Re explaining more, I suggest you appreciate the following. 1. This is a 5000 word article. The longest I have done and possibly the longest anyone has ever sent Cricinfo. I wanted to keep some of the obvious explanations down while going out of the way on important factors like BQI determination. 2. If you thought that I would have valued Hirwani's ctd value at 8.5 or Massies's at 8.51 in their second tests, you have a poor opinion of my cricketing and analytical abilities. That is where people like Alex, Abhi, Arjun et al differ. They KNOW that I would have taken care of such situations. Hence even when I do not give the explanations, they accept the valuations. 3. The figure of 50 wickets has been arrived at based on playing between 10 and 15 tests. Ananth: ]]

    The Bowling quality index (BQI) is based on Career-to-date values

    So, for the batsmen who faced Warne in the Sydney test of 1992, which average will you consider? [[ Probably 35. Ananth: ]]

    You keep on hapring back to FC. Frankly I would not touch FC with a 100-foot pole.

    Ananth: I'm sorry I don't agree. I wouldn't want to touch the career average of someone who played in 4 tests over 10 years (for no fault of his). [[ How do I get through to you. I DO NOT HAVE FC AVERAGES, NOR ACCESS TO THE SAME. I WOULD NOT TRUST FC OF MOST COUNTRIES. I WOULD NEVER CONTEMPLATE BUILDING A FC DATABASE. Resolution of names would take an year and I would go mad. In Pakistan Habib Bank vs PIA is First class. In India Tripura vs Assam is FC. There was a time when Moin-ud-dowlah was FC. God knows what in Slk or Bng.All your arguments are based on Australia and England. If a guy played 4 tests over 10 years, his ctd figure would probably be 50, justifiably so. I am not faulting the poor sod only playing in 4 tests over 10 years. But he will be ranked very very low in my valuations, that is all. And the batsmen who faced him would have had, as Boycott would say, a buffet lunch. Why don't you accept that for a batsman of Hammond's calibre, the Nzl bowling of #226 was buffet lunch and more. Hammond's 85 against Australia couple of tests before is placed at 25.86. Look at that attack. Ananth: ]]

    Re Ian Smith - I was referring to this chap who played several tests against Hutton. http://www.espncricinfo.com/ci/content/player/47213.html

    You don't have to use FC average as it is. But what's the point using the Test avg of bowlers who simply didn't get enough opportunities due to War? [[ I think I am spending too much time on this. If you are going to repeat the same arguments, albeit in a different method, I will publish without comments. Ananth: ]]

  • David on May 20, 2011, 1:58 GMT

    I'm an absolute layman when it comes to statistics, so I'm probably missing something basic, but here goes!

    I understand the rationale behind using CTD - facing Lee early in his career would have been more difficult than facing him in the middle of his career (although I'd suggest that the end of his career was his best period ... when he combined skill with pace and didn't have Warne/McGrath leading him). I also agree that before 50 wickets, the methodology must be different.

    But I still wonder if CTD after 50 wickets is too much of a blunt instrument. CTD tells you the broad trend-lines of a bowler's career, but not the smaller peaks and troughs of form, which is more significant to the question of the quality of bowling the batsman is actually facing.

    What would be the pros and cons of taking 11-test slices (current test + 5 tests on either side) instead of CTD? One small bonus would be that as well as form, it would capture something of pitch type. [[ If I wanted to incorporate a dynamic past-10-tests figure, probaby correct, this article would have come out three months later. Ananth: ]]

  • Ravi on May 19, 2011, 19:18 GMT

    A very insightful article Ananth. A good deal of thought going into it and also a great deal to get out of it! Some myths shattered, some reinforced. Aside from the numbers I have a question reg. Tendulkar who has a particular propensity to fall victim to debutantes and/or novices (Panesar, Paul Harris, Paul Adams, Aquib Javed, Giles, Olonga etc). How are debutante's c-t-d numbers incorporated into the BQ? And can they fall into a high BQ group eg. Panesar together with Flintoff, Hoggard, Harmison? [[ Ravi, it does not matter who dismissed SRT. What matters is who was part of the attack. I am very severe on early Tests. Until a bowler reached 50 wickets he does not get credit for ctd average, if it is low. Otherwise we would have some Lohmannesque numbers in thje second Tests of Hirwani and Massie. After 50 wickets, the bowlers get credit for whatever is their ctd average. Ananth: ]]

  • shrikanthk on May 19, 2011, 18:26 GMT

    The other problem with CTD average is the volatility factor.

    Take Bowler X who makes his debut in Test Y. I've shown X's stats in his first 3 tests.

    Test Y: 4/40, CTD avg - 10 Test Y+1: 1/200, CTD avg - 48 Test Y+2: 0/80, CTD avg - 64!

    We're not being fair to either Bowler X or the batsmen facing him in these tests by using these CTD averages. For all you know, this Bowler X may go on to take 500 test wickets at an average of 25!

    Batsmen like Hammond and Hutton are unfortunately bearing the brunt here because the WI/SA/NZL bowlers whom they faced had very volatile and unreliable CTD averages due to the TINY sample size.

    A suggestion following from all my previous comments:

    - If Test Y is bowler X's Zth test where Z LT 10, then USE ONLY Career average!

    This should greatly reduce volatility in BQIs and will be fair to batsmen of all eras. [[ I think I have said enough. Pl start looking for cheese rather than holes. Ananth: ]]

  • shrikanthk on May 19, 2011, 18:16 GMT

    Another gripe against using pure CTD averages is that it is unfair to a lot of great batsmen who came up against young great bowlers at the fag end of their careers.

    Take Bradman vs Bedser for instance. Now, Bedser is an all-time great. But between '46 and '49, Bedser was single-handedly shouldering the English attack against very, very strong Aus batting sides. No wonder his CTD average in his matches against Bradman won't be too special. That doesn't mean he wasn't a good bowler to begin with. Bradman always regarded Bedser as the best medium pacer he ever faced. So, it is unfair to both Bradman and Bedser to use Bedser's CTD average in those early games.

    Another example: Shane Warne had a terrible CTD average at the end of his first test. So, strangely Shastri and Tendulkar are getting no credit whatsoever for having played him so well in that Sydney test!

    I strongly favour using a weighted mean of the CTD average and Career average (50-50 weights). [[ The batsmen who faced Shane Warne after he reached 50 wickets (a period of 132 tests) will get full credit. Even earlier they will get more credit for facing Warne than Blewett even if Blewett's ctd is better than Warne. There is no problem with warne and like. The problem is Dunning and Babcock. For some reason you think Hammond should get higher credit. Why I do not know. And what is this "get no credit whatsoever". A sweeping statement when you do not know the methodology used. Ananth: ]]

  • shrikanthk on May 19, 2011, 18:02 GMT

    Insufficient sample sizes for NZL/WI/SA bowlers of the inter war years are primarily responsible for the large proportion of BQI GT 45 innings for Hobbs/Hammond and Hutton.

    I cited the instance of Ian Smith in the previous comment. Consider Lindsay Tuckett. 9 tests between '47 and '49. First-class average of 23. Test average of 52! Again, a decent fast bowler whose test-average is meaningless thanks to the small sample size.

    Even two of the NZL bowlers murdered by Hammond in '32-33 - Badcock and Dunning have FC avgs of 24 and 28 respectively. Test avgs of 38 and 99. Again both of them played in less than 10 tests.

    I'm not saying that Dunning and Badcock were world-beaters. But they're certainly better than their Test-averages.

    What's the point taking the CTD average of someone who has played in 4-5 tests? For bowlers with less than 10 tests, I strongly recommend a weighted average of CTD test avg and overall FC average (I suggest a weightage of 30-70 for Test vs FC). [[ I get the feeling that you are looking for holes rather than the cheese. What ypu are referring to is a very small part of the overall scene. Who did Dunning bowl to. What was the level of New Zealand cricket. In New Zealand what is 26. Until they reach 50 wickets I do not take the CTD average. i take a notional average which also depends on how their career turned out. Dunning might have 45 and McGrath (in his early tTests) might have 35. Just examples. Ananth: ]]

  • arch on May 19, 2011, 17:54 GMT

    Look at Viv Richard's average against Group 1, and compare it to his performance against Groups 4 and 5. I guess he really did get bored playing second rate bowlers!

  • shrikanthk on May 19, 2011, 17:50 GMT

    Very intriguing piece.

    Are you telling me that Hutton had 81 innings against teams with Bowling Quality index GT 45? (I suppose the BQI is basically the CTD bowling average of the attack weighted by the number of overs bowled)

    That figure stunned me! 81 innings? I was flummoxed for a while, but eventually figured it out. A lot of English players pre WWII who played a lot against WI/SA/NZL have a lot of innings with BQI GT 45. Here's the reason:

    Several NZL, WI and SA bowlers of the inter-war years played VERY LITTLE test cricket (LT 10 tests mostly). Many of them had abysmal averages in those brief careers, on perfect pitches against great batting sides like Eng and Aus.

    That doesn't mean they're as bad as the average suggests! Take this guy - Ian Smith, an SA leg break bowler who played 9 tests between '47 and '57. He averaged 64 in tests. But 22.5 in First-class cricket! So, obviously a pretty decent bowler. His test average of 64 is a misleading result of a small sample. [[ Shri, please see the bowling attack against which Hammond scored 336. 4 bowlers who had a total of 26 ctd wickets at 40+. Not that these were McGraths or Lillees. They finished their combined careers with 36 wickets at similar or worse averages. What is this attack other than one of the poorest. You keep on hapring back to FC. Frankly I would not touch FC with a 100-foot pole. Anyhow I do not have FC records. I do not even know who Ian Smith is. Cannot trace him. But if he averages 64, he is a 64-level bowler. Pl see response to Ravi. Ananth: ]]

  • Abdullah on May 19, 2011, 17:46 GMT

    Hi ananth

    Amazing article. One if the best if not your best piece of analysis to date.

    Following on from opemp's suggestion, can you do a table of the top performances aagainst group 5 bowlers by any batsman. I'm trying to get to a top-10 top-20 innings analysis based on this measure. There would be some subjectivity in it but I trust your judgement and it will be a nice one to debate about [[ There is already a table of top batsmen against group 5 bowlers. You could import that into an Excel sheet and do whatever you want. Ananth: ]]

  • OpEmp on May 19, 2011, 16:23 GMT

    Great analysis, as usual. I have particularly enjoyed your previous two articles as well.

    I was scanning the extra tables you posted and was wondering whether you could order them by BQI faced by the batsmen? I am curious about individual innings where batsmen performed admirably (centuries, half-centuries) against particularly potent bowling attacks (as measured by BQI). Perhaps some unheralded innings, hidden gems if you will, will emerge from such an inquiry. [[ The table of the group 5 performances of all key batsmen is already availeble. The problem is the huge size of tables. Let me see whather I can do a table of all 50+ performances of all top batsmen. Ananth: ]]

  • Abhi on May 19, 2011, 15:58 GMT

    Still just glancing through. But VVS seems to be one guy for whom it just doesn't matter who is bowling! (Or what the match situation is either!) The Iceman just goes out and does his thing- nevermind bowling quality, match situation,first inn, last inn etc etc! [[ In terms of the sum of the last two groups (the really weak bowling groups), the following are the lowest. Stewart: 13.7% VVS: 16.0% Lara: 16.4% Sehwag: 17.1% Ponting: 18.5% Richards: 18.5% This tells something. Ananth: ]]

  • Abhi on May 19, 2011, 15:30 GMT

    Ananth, FAB effort. One of your best.

    Arjun, I somewhat disagree with your comments.As with all stats it pays to use a pinch of salt and keep in mind that stats are like bikinis....

    I disagree with your one-on-one comparison of Gavaskar and Sunny. I know Ananth "Warned" us about it above - but you seem to have forgotten that pitch quality is not mentioned. Numerous ppl have mentioned Sehwags record in SA,Eng etc.

    But,anyway it certainly is an absolutely WONDERFUL effort- even though we use a single parameter (bowling quality)...It does throw some light on so many issues..

    More observations to follow!... [[ By Sunny you mean Sehwag. I think it was hust a set of observations from Arjun. But intriguing and deserving further look-in. I may very well do a separate article on the Pitch type groups. I like this concept of getting into groups rather than absolute values which hide a lot of things. Ananth: ]]

  • AN on May 19, 2011, 14:46 GMT

    @David: I see what you are getting at, but there is nothing wrong if somebody uses data like this to show that indeed batsman A is better (or sometimes much better) than B? It bears repetition because many (including cricket writers) use one match or series to anoint a batsman in broad terms. Everybody is equally guilty on this and need to be shown reality often, to hammer the point home. On the other hand, denial is a strong force and may not make a difference to people's emotional assessments. [[ I think David is referring to the fact that often the whole thread veers off to lengthy discussions on one batsman. I have no probelms with any comment subject to the three criteria all of you are familiar with. Ananth: ]]

    Nice analysis by Mr. Ananthanarayanan as usual.

  • mohanlal on May 19, 2011, 14:45 GMT

    Hi Ananth sir ,As usual a real x-ray of stats from your part.But even though this shows the strengths and weeknesses of various batsmen against these different groups of attacks this does not imply that the batsman with the best avg: against the best group has a huge advantage against one with a fairly low average. For eg: take ViV. His avg: is worst against the worst group.That does not imply that he deliberately got out against these attacks for low scores.It might be that he couldn't motivate enough(which is his weakness indeed) or that he was troubled more by certain bowlers in this group.Similarly with Stevewaugh.Though low avg: against best group, he was the best bat to come at say 30/3 or 40 for 4 etc.Any way this analysis reveals a lots of facts. [[ Lack of motivation cannot be termed as justification. If anything the lack of motivation could be understood and accepted against weak attacks. My suggestion is to look at the groups 1 and 2 and possibly 3 together. These represent very good to fair attacks. Ananth: ]]

  • Arjun on May 19, 2011, 13:49 GMT

    Steve waugh avg. 34.0 ag. top group 111.0 ag. 45+ group.

    Jimmy Adams avg. 29.0 ag. top group 102.0 ag. 45+ group.

    Javed Miandad avg. 27.0 ag. top group 65.0 ag. 40+ group. 68.0 ag. 45+ group.

    percetion - they were tough cookies, men of steel, fighter extra-ordinaire etc. reality- They crumbled against top quality stuff and cashed in against poor attacks. Infact, steve waugh has worst average in 4th innings of a test match with no centuries and has scored almost 1800 runs @ 111.0 against 45+ attacks. don't want to downgrade any of these great players but this article will be real eye-opener for the readers.

    [[ I can see sounds of the hornets moving about since their nest is being stirred. Let the fun and games start. Ananth: ]]

  • Arjun on May 19, 2011, 13:16 GMT

    Against Top group (only pak batsmen) Haneef Moh. aveg. 19.0 Javed Miandad 27.0 Saeed anwar 32.0 Inzamam 35.0 Amir Sohail 23.0 Moh Yousuf 30.0 Only Younis khan averages respectable 44.0 Since Pak have better record against top sides, this data alone indicates how much more contibution pak bowlers have made in their victories. [[ That is an interesting observation. Fazal/KhanMohd/Sarfraz/Imran/Qadir/Mushtaq/Qasim/Wasim/Waqar/Saqlain/Akhtar is a formidable set over 50 years. If we do it for India, it might be probably be the other way around. Ananth: ]]

  • Arjun on May 19, 2011, 12:30 GMT

    The purpose of this article has been served, it is to seprate perception form reality.

    Perception A-gavaskar scored most of his runs against fearsome westindies quatret and mostly against tough attacks. In reality he scored 1482 runs @ 41.0 against best of the best, which is about 14% of his career runs.

    Perception B- Sehwag has scored most of his runs 7000 odd runs against average attacks and on battig tracks. In reality he has scored 2302 runs @ 56.0 against best of the best, which is about 30% of his career runs. I am not saying sehwag is better than gavaskar. Although most cricket lovers will select both of them as openers in India's all time Test XI.

    Against top group Martin crowe aveg. just 19.0 Atherton 23.0 jayawardene, strauss, saeed anwar just 32.0 Steve waugh, greame smtih, zahher abbas just 34.0, Ganguly, attapatu, S fleming just 28.0 hard to believe but it is reality. [[ Arjun The other common perception is that both Sangakkara and Jayawardene are similar in their batting deliveries. When you see their own group-wise figures, barring the share between group 5 and 4, see how similar they are. I have not thanked you separately in the article. But many many thanks. Ananth: ]]

  • dhanush on May 19, 2011, 11:56 GMT

    Hi Ananth,

    An Excellent analysis to compare the technicality of the batsmen!!! Group-wise distribution of Players based on average would result in more analytical fight between readers

  • Ashish Mehta on May 19, 2011, 11:32 GMT

    Dear Anantha, Really very interesting post and the one for which I was desperately waiting for. I was surprised on BC Lara not making to the all time world XI. This proves how great he was when it comes to batting. One suggestion I would like to add: Instead of Average we should take into account the RPI (Runs Per Innings) that would give more clear picture. Because personally I have seen so many times batting greats like Sachin and others exposing the tail-enders to remain "not out"... But has never been the case with BC Lara. Any way just an personal observation. Apologies to the readers if I am wrong. [[ RpI by itself is only as good or as bad as the Average. While the Average benefited by the number of not outs, the RpI is unfairlly weighted down by small unbeaten scoes. In another context, I adopted the simple idea of considering single-digit dismissals as completed innings, in other words penalize such dismissals. Works very well in practice. But it would still be a derived figure. On balance the Average seems the most accepted one. Amongst top batsmen only Lara had not outs below 10. But that is the way he batted and should be accepted . Ananth: ]]

    One more subject I feel would be interesting to work on. i.e. ICC provides ratings to every individual batsman after every inning played so the list of top 25 test batsman on the basis of average of all the individual career ratings achieved till date.

    Once again this is a very good post as others posted by you.

  • archit on May 19, 2011, 11:26 GMT

    Great Analysis Ananth, as usual. the modern greats - lara, sachin, sehwag, ponting, kallis, dravid - don't seem to have drastically different figures.

  • KnowWho on May 19, 2011, 10:11 GMT

    Hi Ananth, Quite an interesting article.However it is like comparing oranges to aplles if we take across ages. But it brings out few similarities too between top batsmen. For example Lara and Sachin have same averages. pOnting kallis dravid are down. Near abt 50. Sehwag and Sanga have same averages near around 56. [[ What fun is there if we restrict ourselves to comparing players only within a range of 20 years. Also partly this has been taken care of with the adjustment for periods. You would understand if you read the article in depth. Ananth: ]]

  • David on May 19, 2011, 10:07 GMT

    Can I just get in with a quick pro forma comment, which will satisfy everyone who feels the need to waste space with a "I told you my favourite batsman was the best" type of comment. Just print it out, fill in the blanks and cross out the options you don't want, then pin it up on your wall at home so you can get the full benefit of your insight:

    "What Anantha's analysis shows is that __________ is light years ahead of __________ because __________ scored more/fewer/a different number of runs against quality/average/poor attacks and his total percentage of runs against quality/average/poor attacks was higher/lower/not even vaguely the same as __________'s. So there!" [[ I appreciate the spirit behind this statement. But I am used to handling all types of comments well now. Last article had over 350 comments. Ananth: ]]

  • jay on May 19, 2011, 9:13 GMT

    Awesome, Ananth. Don't worry, even after this much hard work, people will surely find some faults with it.

  • No featured comments at the moment.

  • jay on May 19, 2011, 9:13 GMT

    Awesome, Ananth. Don't worry, even after this much hard work, people will surely find some faults with it.

  • David on May 19, 2011, 10:07 GMT

    Can I just get in with a quick pro forma comment, which will satisfy everyone who feels the need to waste space with a "I told you my favourite batsman was the best" type of comment. Just print it out, fill in the blanks and cross out the options you don't want, then pin it up on your wall at home so you can get the full benefit of your insight:

    "What Anantha's analysis shows is that __________ is light years ahead of __________ because __________ scored more/fewer/a different number of runs against quality/average/poor attacks and his total percentage of runs against quality/average/poor attacks was higher/lower/not even vaguely the same as __________'s. So there!" [[ I appreciate the spirit behind this statement. But I am used to handling all types of comments well now. Last article had over 350 comments. Ananth: ]]

  • KnowWho on May 19, 2011, 10:11 GMT

    Hi Ananth, Quite an interesting article.However it is like comparing oranges to aplles if we take across ages. But it brings out few similarities too between top batsmen. For example Lara and Sachin have same averages. pOnting kallis dravid are down. Near abt 50. Sehwag and Sanga have same averages near around 56. [[ What fun is there if we restrict ourselves to comparing players only within a range of 20 years. Also partly this has been taken care of with the adjustment for periods. You would understand if you read the article in depth. Ananth: ]]

  • archit on May 19, 2011, 11:26 GMT

    Great Analysis Ananth, as usual. the modern greats - lara, sachin, sehwag, ponting, kallis, dravid - don't seem to have drastically different figures.

  • Ashish Mehta on May 19, 2011, 11:32 GMT

    Dear Anantha, Really very interesting post and the one for which I was desperately waiting for. I was surprised on BC Lara not making to the all time world XI. This proves how great he was when it comes to batting. One suggestion I would like to add: Instead of Average we should take into account the RPI (Runs Per Innings) that would give more clear picture. Because personally I have seen so many times batting greats like Sachin and others exposing the tail-enders to remain "not out"... But has never been the case with BC Lara. Any way just an personal observation. Apologies to the readers if I am wrong. [[ RpI by itself is only as good or as bad as the Average. While the Average benefited by the number of not outs, the RpI is unfairlly weighted down by small unbeaten scoes. In another context, I adopted the simple idea of considering single-digit dismissals as completed innings, in other words penalize such dismissals. Works very well in practice. But it would still be a derived figure. On balance the Average seems the most accepted one. Amongst top batsmen only Lara had not outs below 10. But that is the way he batted and should be accepted . Ananth: ]]

    One more subject I feel would be interesting to work on. i.e. ICC provides ratings to every individual batsman after every inning played so the list of top 25 test batsman on the basis of average of all the individual career ratings achieved till date.

    Once again this is a very good post as others posted by you.

  • dhanush on May 19, 2011, 11:56 GMT

    Hi Ananth,

    An Excellent analysis to compare the technicality of the batsmen!!! Group-wise distribution of Players based on average would result in more analytical fight between readers

  • Arjun on May 19, 2011, 12:30 GMT

    The purpose of this article has been served, it is to seprate perception form reality.

    Perception A-gavaskar scored most of his runs against fearsome westindies quatret and mostly against tough attacks. In reality he scored 1482 runs @ 41.0 against best of the best, which is about 14% of his career runs.

    Perception B- Sehwag has scored most of his runs 7000 odd runs against average attacks and on battig tracks. In reality he has scored 2302 runs @ 56.0 against best of the best, which is about 30% of his career runs. I am not saying sehwag is better than gavaskar. Although most cricket lovers will select both of them as openers in India's all time Test XI.

    Against top group Martin crowe aveg. just 19.0 Atherton 23.0 jayawardene, strauss, saeed anwar just 32.0 Steve waugh, greame smtih, zahher abbas just 34.0, Ganguly, attapatu, S fleming just 28.0 hard to believe but it is reality. [[ Arjun The other common perception is that both Sangakkara and Jayawardene are similar in their batting deliveries. When you see their own group-wise figures, barring the share between group 5 and 4, see how similar they are. I have not thanked you separately in the article. But many many thanks. Ananth: ]]

  • Arjun on May 19, 2011, 13:16 GMT

    Against Top group (only pak batsmen) Haneef Moh. aveg. 19.0 Javed Miandad 27.0 Saeed anwar 32.0 Inzamam 35.0 Amir Sohail 23.0 Moh Yousuf 30.0 Only Younis khan averages respectable 44.0 Since Pak have better record against top sides, this data alone indicates how much more contibution pak bowlers have made in their victories. [[ That is an interesting observation. Fazal/KhanMohd/Sarfraz/Imran/Qadir/Mushtaq/Qasim/Wasim/Waqar/Saqlain/Akhtar is a formidable set over 50 years. If we do it for India, it might be probably be the other way around. Ananth: ]]

  • Arjun on May 19, 2011, 13:49 GMT

    Steve waugh avg. 34.0 ag. top group 111.0 ag. 45+ group.

    Jimmy Adams avg. 29.0 ag. top group 102.0 ag. 45+ group.

    Javed Miandad avg. 27.0 ag. top group 65.0 ag. 40+ group. 68.0 ag. 45+ group.

    percetion - they were tough cookies, men of steel, fighter extra-ordinaire etc. reality- They crumbled against top quality stuff and cashed in against poor attacks. Infact, steve waugh has worst average in 4th innings of a test match with no centuries and has scored almost 1800 runs @ 111.0 against 45+ attacks. don't want to downgrade any of these great players but this article will be real eye-opener for the readers.

    [[ I can see sounds of the hornets moving about since their nest is being stirred. Let the fun and games start. Ananth: ]]

  • mohanlal on May 19, 2011, 14:45 GMT

    Hi Ananth sir ,As usual a real x-ray of stats from your part.But even though this shows the strengths and weeknesses of various batsmen against these different groups of attacks this does not imply that the batsman with the best avg: against the best group has a huge advantage against one with a fairly low average. For eg: take ViV. His avg: is worst against the worst group.That does not imply that he deliberately got out against these attacks for low scores.It might be that he couldn't motivate enough(which is his weakness indeed) or that he was troubled more by certain bowlers in this group.Similarly with Stevewaugh.Though low avg: against best group, he was the best bat to come at say 30/3 or 40 for 4 etc.Any way this analysis reveals a lots of facts. [[ Lack of motivation cannot be termed as justification. If anything the lack of motivation could be understood and accepted against weak attacks. My suggestion is to look at the groups 1 and 2 and possibly 3 together. These represent very good to fair attacks. Ananth: ]]