Captaincy January 9, 2009

Test Captains - an in-depth look

This article has been in the pipeline for long
61

This article has been in the pipeline for long. An analysis of Test Captains is not an easy task and will lead to many arguments and comments. However that cannot deter us from making an honest attempt. As long as the comments are positive in nature, it does not matter.

What are the requirements of a good Test captain? The measurable factors are on-field performance as a player, leading from the front, achieving good match and series results, both home and away. The non-measurable factors are man management, identification of talent, getting players to do their best and support of team members with selection entities. I will only concentrate on the measurable factors and stay away from the non-measurable ones. I am confident that this will be fair even to great captains whose on-field performance might be below par.

I have decided on a few yardsticks for this analysis. Readers should be happy with these since these reflect earlier reader comments.

The first is that there will be no longevity based allocation of points. I will set a fairly high bar for selection. Once this bar is crossed, all the selected captains will have an equal chance of achieving a high position in the table.

The second is that the team strength measures will be adjusted for the period during which the concerned Test was played. This will ensure a fair playing field.

The third is that the captains' individual performances will be weighted for the quality of opposition. The all-Test batting average of 29.92 and bowling average of 31.51 are used as reference. In other words, a 100 against a strong Australian team will be weighted at a much higher level than a 100 against a weak Bangladesh team. Similarly for bowling.

After a number of trial runs I have decided on 30 Tests as the minimum requirement for inclusion. This has been worked on various factors, not the least is the need to keep the number of qualifying cricketers, in this case 35, to a reasonable number. Also 30 Tests represents between 3 and 5 years reign, a fairly long one. Unfortunately this keeps out very successful captains such as Don Bradman, Richie Benaud, Wasim Akram, Jayawardene, Shaun Pollock (he was a very good Test captain) to name a few. To do proper justice to these great players, I have presented an additional table of captains who have led their teams in 20-29 tests at the end.

Now for the details, to be followed by the tables. I waited for the end of the wonderful Test match at Sydney to prepare these tables since the result there might have had a bearing on the final positions.

The measures for analysing Test Captains is broadly classified into the following four (measurable) factors.

1. Base unadjusted match results.
2. Match results adjusted for team quality and venue.
3. Series results.
4. Individual performances - Batting, Bowling and Fielding.

1. Base unadjusted match results.

These are the raw unadjusted results. A win is a win, whether it is against Australia at Sydney or against Bangladesh at Mirpur. Similarly a draw is not a loss and as such some credit has to be given. The measure of success is derived by the following formula and converted to points.

No of wins   +   (No of draws / 2)
Success Factor =  ----------------------------------
No of matches

A captain who wins all the matches (not that any one has done it) gets full credit.

2. Match results adjusted for team quality and venue.

The best way of explaining this measure, which carries the most weight is to show a table of imaginary match results. Let me take three teams. Australia, with a TSI of 75, England, with a TSI of 60 and Bangladesh, with a TSI of 45. All possible results and the winning captain credits are tabulated below.

Home win           Away win
Stronger team winning
Australia (75) defeats England (60)           60                 72
Australia (75) defeats Bangladesh (45)        36 (lowest)        45
England (60) defeats Bangladesh (45)          41                 50
Weaker team winning
Bangladesh (45) defeats Australia (75)        83                100 (highest)
Bangladesh (45) defeats England (60)          66                 80
England (60) defeats Australia (75)           75                 90

The summary is that the minimum points are allocated when the strongest team defeats the weakest team at home and the maximum points are allocated when the weakest team defeats the strongest team away. The limiting values have a factor of nearly 3 between themselves. Everything else is in between.

Captains in drawn matches get 50% of the adjusted TSI values.

The points for all tests captained by one player are summed and divided by the number of Tests captained. This ensures that longevity in captaincy does not play a part.

3. Series results.

A few comments on the Series calculations. Until now a total of 590 series have been played. Single Test series, 50 of these at last count, are not considered to be series. A minimum of two Tests have to be played. There have been three multi-team series (The 1912 Triangular tournament at England between Eng-Aus-Saf, the First Asian Test Championship of 1998-99 and Second Asian Test Championship of 2001-02). For these three tournaments, the home teams are respectively England, Pakistan and Bangladesh. The winner of these three tournaments, viz., England, Pakistan and Sri Lanka get the winner's credits. For the only series played in neutral locations, the 2002 matches between Pakistan and West Indies/Australia, all three teams are considered to be "Away".

The winning captain gets the average of the losing team's TSI as credit for winning the series. A bonus of 20% is given for winning an away series. If multiple captains have captained within a series they get proportionate credit. All these are subject to the above mentioned adjustments for relative strengths of the two teams. Aus/Ind/Saf will get least credit for winning at Bangladesh while Bangladesh will get maximum credit for winning at Aus/Saf/Ind, if ever that miracle happens.

The points for all series/part series captained by one player are summed and divided by the number of Tests captained. This ensures that longevity in captaincy does not play a part. Number of Tests rather than number of series is used to ensure uniform weighting. Also all series wins are treated the same. Of course, if a captain wins the series 5-0 he would have got substantial match win credit points as against one who wins 1-0.

It is possible that there is some overlap amongst the three Results based parameters. However each has a different objective and the overlap exists unformly across all captains. The across-the-board division by the "Tests captained" figure smoothes all variations.

4. Individual performances - Batting, Bowling and Fielding.

Finally on-field performance. I think it is essential to recognize the batting, bowling and fielding performances of the captain. Cricket is not any longer, and should never have been, a non-playing-captain game. We recognize the performances by converting runs/wickets/catches to points, adjust these for the quality of other team's batting and bowling, sum these and then divide by the number of tests captained. Wicket-keeper-captains' dismissals are given additional weighting.

Let me summarize these. I have kept in mind a figure of 70-75% for different Results related points allocation and 25-30% for Performance related points allocation. I also expected to achieve these allocations at the total level. At individual levels the allocations will vary considerably. A low-performing captain such as Brearley gets over 90% on Results related allocations, an average-performing captain such as Steve Waugh gets over 80% on Results related allocations while a Performing captain such as Imran Khan or Sobers gets over 40% in Performance related points allocations. The summary is given below.

For 62 Captains     For 290 captains
(20+ tests)           (All)
Results:          14.35%             14.76%
Matches:          45.55%             46.04%
Series:           13.86%             12.04%
Performance:      26.22%             27.16%

The summarized weighted percentages have come almost very close to what I set out at the beginning.

These summary figures meet the target I had set before beginning the analysis and we can now proceed to complete the table preparation work.

Top Test Captains - Minimum 30 tests as captain

SNo Player Cty (Runs Wkts C/S) Matches Ser Win% Win Mat Ser Perf CapIdx (Adjusted) M W D W Pts Pts Pts Pts

1 Imran Khan Pak (2438 198.6 17) 48 14 26 5 56.2% 11.2 36.9 11.1 43.6 102.88 2 Ponting R.T Aus (4698 1.1 53) 53 36 9 13 76.4% 15.3 49.8 17.8 18.9 101.71 3 Waugh S.R Aus (3854 3.1 34) 57 41 7 13 78.1% 15.6 49.1 15.4 14.6 94.69 4 Sobers G.St.A Win (3516 120.1 48) 39 9 20 3 48.7% 9.7 32.4 7.2 43.9 93.26 5 Illingworth R Eng (1262 53.1 24) 31 12 14 6 61.3% 12.3 41.5 13.8 22.6 90.25 6 Chappell I.M Aus (2463 6.0 46) 30 15 10 5 66.7% 13.3 43.5 13.8 19.6 90.14 7 Richards I.V.A Win (3093 18.3 49) 50 27 15 7 69.0% 13.8 44.2 12.9 16.3 87.18 8 Lloyd C.H Win (5114 0.0 71) 74 36 26 13 66.2% 13.2 45.1 13.3 14.8 86.34 9 Taylor M.A Aus (3292 1.1 84) 50 26 11 11 63.0% 12.6 41.3 15.5 15.0 84.50 10 Cronje W.J Saf (2912 39.4 27) 53 27 15 9 65.1% 13.0 40.8 11.6 17.4 82.93 11 Smith G.C Saf (5717 8.1 88) 67 33 15 14 60.4% 12.1 36.5 14.5 19.4 82.48 12 Chappell G.S Aus (3934 24.2 59) 48 21 14 6 58.3% 11.7 37.8 11.0 21.7 82.09 13 Vaughan M.P Eng (3132 1.1 27) 51 26 14 8 64.7% 12.9 43.0 12.5 13.0 81.49 14 Brearley J.M Eng (1060 0.0 41) 31 18 9 5 72.6% 14.5 46.5 12.2 8.2 81.37 15 Jayasuriya S.T Slk (2194 43.4 23) 38 18 8 8 57.9% 11.6 34.5 13.9 21.3 81.20 16 Kapil Dev N Ind (1351 118.7 26) 34 4 23 2 45.6% 9.1 28.5 6.7 36.6 80.91 17 Simpson R.B Aus (3677 41.7 62) 39 12 15 3 50.0% 10.0 34.1 6.6 29.0 79.65 18 Dexter E.R Eng (2393 35.7 18) 30 9 14 3 53.3% 10.7 34.8 7.5 26.1 79.04 19 Inzamam-ul-Haq Pak (2388 0.0 14) 31 11 9 5 50.0% 10.0 36.2 15.5 15.9 77.46 20 May P.B.H Eng (2971 0.0 28) 41 20 11 7 62.2% 12.4 38.9 10.2 15.2 76.72 21 Javed Miandad Pak (2433 0.0 32) 34 14 14 7 61.8% 12.4 36.9 12.2 15.3 76.65 22 Ganguly S.C Ind (2609 5.4 37) 49 21 15 9 58.2% 11.6 37.2 14.6 12.3 75.75 23 Border A.R Aus (6759 27.0 89) 93 32 39 9 55.4% 11.1 36.1 8.2 17.8 73.24 24 Gooch G.A Eng (3566 9.0 29) 34 10 12 3 47.1% 9.4 32.1 6.9 24.0 72.37 25 Howarth G.P Nzl (1449 1.1 21) 30 11 12 6 56.7% 11.3 36.3 13.4 10.6 71.70 26 Hussain N Eng (2362 0.0 27) 45 17 13 6 52.2% 10.4 38.0 11.0 11.1 70.52 27 Fleming S.P Nzl (5101 0.0 132) 79 28 24 13 50.6% 10.1 31.5 13.5 14.6 69.77 28 Gavaskar S.M Ind (3438 0.0 45) 47 9 30 4 51.1% 10.2 34.0 7.0 15.6 66.85 29 Azharuddin M Ind (3064 0.0 50) 47 14 19 5 50.0% 10.0 29.6 7.4 14.1 61.09 30 Ranatunga A Slk (3242 5.0 26) 56 12 25 6 43.8% 8.8 26.1 9.4 12.8 57.07 31 Atherton M.A Eng (3600 1.1 33) 54 13 20 3 42.6% 8.5 28.7 5.4 14.1 56.73 32 Reid J.R Nzl (2022 55.5 23) 34 3 13 0 27.9% 5.6 19.9 3.0 25.6 54.19 33 Lara B.C Win (4388 0.0 72) 47 10 11 4 33.0% 6.6 21.3 5.9 20.2 53.98 34 Mansur Ali Khan Ind (2446 1.0 26) 40 9 12 2 37.5% 7.5 23.8 5.1 13.1 49.47 35 Gower D.I Eng (2295 0.0 26) 32 5 9 2 29.7% 5.9 19.9 5.1 15.2 46.08

Imran Khan is deservedly on top, both for his success as a captain and as a performer. He always led from the front. His average of 50 runs & 4 wickets per test as captain are testimony to this. His top position is due to his high level of consistent performances, And that is how it should be.

Ponting has been a very good and successful captain. People might say that this was easy with world class performers such as Warne, McGrath and Gilchrist playing under him. He still had to produce the results. Incidentally he was comfortably in the top position when I started this a couple of months back. The twin losses to India and South Africa have pushed him down. Ponting has averaged nearly 95 runs per test as captain.

Steve Waugh was as charismatic as Imran Khan. He inherited a good side from Taylor and handed over nearly as good side to Ponting. The changeover of the old guard under him was smooth and effective. His performance, however, has been average. Only 65 runs per test as captain.

Gary Sobers' results as a captain have been only average. He is the one of two captains to get a below-50% success rate in the top-20. However his performances on the field as captain have been the best by anyone. 90 runs and 3 wickets per test have pushed him into the fourth place. Overall a very deserved position.

Illingworth was again a successful captain with above average performance. His Ashes wins are legendary.

It can be seen that Mike Brearley, considered by many to be possibly the best captain ever is very well placed at the 14th place. Note his results scores and his performance score. He was a great captain but a mediocre performer. He scored a very low 35 runs per test.

If there was a bravery factor introduced, Greame Smith would be at the top. His performance at Sydney was heart-warming. However his achievements came much earlier, at Perth and Melbourne. There is no doubt that, by the time he finishes his captaincy career, he would be right at the top. A performing leader, Smith averaged 85 runs per test.

Lara is placed way down the table, justifiably so. One of the greatest batsmen who ever played the game, Lara was, at best, an average captain. These statements would also apply to the other great, Tendulkar.

Kapil Dev is the best Indian captain. Readers might say that Ganguly achieved more as a captain. However Ganguly's average performance (52 runs per test as captain) pushed him down a few places. Readers must also remember that this is an all-time best captain list and Kapil's 16th and Ganguly's 22nd places are reasonable rewards for their contributions to Indian cricket.

A few interesting captaincy related points:

1. 290 players have captained their teams in the 1905 Test matches, 41 of them having done so only once.
2. Alan Border has captained in most tests, 93, followed by Stephen Fleming with 79 tests.
3. Steve Waugh has won most tests, 41, followed by Clive Lloyd and Ricky Ponting (after the great Sydney win), with 36 wins.
4. The best result has been achieved by Steve Waugh with 78.1%, followed by Ponting with 76.4%.
5. With the great Australian series win, Greame Smith has won 14 series, alone at the summit he shared with four others. Lloyd, Steve Waugh, Ponting and Fleming have 13 series wins.
6. Imran Khan has taken most wickets, 187 in 48 tests, followed by Richie Benaud with 138 in only 28 tests as captain. Incidentally Benaud has performed in an outstanding manner as a captain. In 35 other tests he has taken only 110 wickets. 7. Border leads the run tally for captains with 6623 runs in 93 tests, followed by Greame Smith with 5633 in 67 tests as captain.

Top Test Captains - Addl report for those who captained between 20 and 29 tests

SNo Player Cty (Runs Wkts C/S) Cm Cw Cd Sw Win% WPts MPts SPts PPts CapIdx

1 Benaud R Aus ( 765 144.3 32) 28 12 12 5 64.3% 12.9 44.1 13.6 47.8 118.35 2 Pollock S.M Saf ( 946 108.7 22) 26 14 7 7 67.3% 13.5 41.0 20.3 41.6 116.36 3 Wasim Akram Pak ( 928 110.7 16) 25 12 5 4 58.0% 11.6 38.6 13.0 43.5 106.63 4 Bradman D.G Aus (3244 0.0 18) 24 15 6 4 75.0% 15.0 48.3 12.7 27.8 103.79 5 Jayawardene M Slk (2683 0.0 42) 26 15 4 6 65.4% 13.1 38.4 17.7 22.3 91.38 6 Walsh C.A Win ( 121 89.6 7) 22 6 9 4 47.7% 9.5 33.3 12.9 34.0 89.69 7 Hutton L Eng (1817 1.1 11) 23 11 8 4 65.2% 13.0 43.1 14.2 16.7 87.05 8 Hassett A.L Aus (1890 0.0 13) 24 14 6 3 70.8% 14.2 44.1 8.8 16.3 83.32 9 Dravid R Ind (1722 0.0 32) 25 8 11 5 54.0% 10.8 37.4 14.2 15.1 77.45 10 Richardson R.B Win (1260 0.0 20) 24 11 7 3 60.4% 12.1 43.1 9.9 11.3 76.34 11 Bedi B.S Ind ( 296 99.5 6) 22 6 5 1 38.6% 7.7 24.8 3.4 39.2 75.06 12 Woodfull W.M Aus (1495 0.0 2) 25 14 4 4 64.0% 12.8 39.0 11.0 12.0 74.84 13 Lawry W.M Aus (1921 0.0 17) 25 9 8 2 52.0% 10.4 37.2 8.0 16.0 71.61 14 Goddard J.D.C Win ( 658 28.6 14) 22 8 7 3 52.3% 10.5 33.5 10.4 17.0 71.42 15 Cowdrey M.C Eng (1663 0.0 22) 27 8 15 2 57.4% 11.5 35.3 9.7 13.1 69.60 16 Hammond W.R Eng (1792 3.0 28) 20 4 13 3 52.5% 10.5 28.4 9.7 20.5 69.11 17 Darling J Aus ( 863 0.0 15) 21 7 10 4 57.1% 11.4 36.1 11.7 8.9 68.08 18 Smith M.J.K Eng (1097 0.0 33) 25 5 17 2 54.0% 10.8 34.4 9.6 10.1 64.98 19 Kardar A.H Pak ( 873 19.8 15) 23 6 11 1 50.0% 10.0 32.5 5.0 15.1 62.57 20 Hooper C.L Win (1576 22.5 21) 22 4 7 2 34.1% 6.8 22.8 5.7 23.5 58.84 21 Tendulkar S.R Ind (1951 4.4 16) 25 4 12 2 40.0% 8.0 25.6 7.6 17.7 58.83 22 Streak H.H Zim (1101 54.3 8) 21 4 6 2 33.3% 6.7 15.2 4.8 31.6 58.20 23 Gatting M.W Eng (1555 1.8 13) 23 2 16 1 43.5% 8.7 28.6 5.0 14.7 57.05 24 Hughes K.J Aus (1668 0.0 18) 28 4 11 1 33.9% 6.8 23.0 3.0 12.6 45.34 25 Flower A Zim (1350 0.0 117) 20 1 9 0 27.5% 5.5 17.4 3.0 19.4 45.26 26 MacLaren A.C Eng (1311 0.0 23) 22 4 7 0 34.1% 6.8 23.6 0.0 13.0 43.41 27 Campbell A.D.R Zim ( 958 0.0 30) 21 2 7 1 26.2% 5.2 18.1 6.9 10.6 40.73

Richie Benaud was an outstanding leader and a great performer, averaging nearly 5 wickets and 30 runs per test as captain. The unassuming Shaun Pollock also achieved considerable success as a Test captain. These should not be forgotten because of the 2003 WC fiasco. Coupled with a high success rate he also averaged 38 runs and 4 wickets per test. Wasim Akram had slightly better figures as a performer and slightly worse figures under the results category. Don Bradman's success as a captain and performer is reflected in the fourth position. Note Jayawardene's performance. He is the only one, other than the Don, to average over 100 runs per test as captain.

If the cut-off had been lower at 25 tests, Benaud, Pollock and Wasim Akram would have taken the first three positions. I would not have too many problems with that list.

Douglas Jardine does not make the cut-off for this list also. He captained England 15 times and won 9 times, 4 of these, through the probably unethical body-line methods, against the strong Australian team.

Top Test Captains - Addl report excluding Performance data - 20 tests and above

SNo Player Cty Cm Cw Cd Sw Win% WPts MPts SPts CapIdx

1 Ponting R.T Aus 53 36 9 13 76.4% 15.3 49.8 17.8 82.82 2 Waugh S.R Aus 57 41 7 13 78.1% 15.6 49.1 15.4 80.13 3 Bradman D.G Aus 24 15 6 4 75.0% 15.0 48.3 12.7 76.01 4 Pollock S.M Saf 26 14 7 7 67.3% 13.5 41.0 20.3 74.77 5 Brearley J.M Eng 31 18 9 5 72.6% 14.5 46.5 12.2 73.21 6 Lloyd C.H Win 74 36 26 13 66.2% 13.2 45.1 13.3 71.55 7 Richards I.V.A Win 50 27 15 7 69.0% 13.8 44.2 12.9 70.90 8 Chappell I.M Aus 30 15 10 5 66.7% 13.3 43.5 13.8 70.59 9 Benaud R Aus 28 12 12 5 64.3% 12.9 44.1 13.6 70.53 10 Hutton L Eng 23 11 8 4 65.2% 13.0 43.1 14.2 70.37 11 Taylor M.A Aus 50 26 11 11 63.0% 12.6 41.3 15.5 69.46 12 Jayawardene M Slk 26 15 4 6 65.4% 13.1 38.4 17.7 69.13 13 Vaughan M.P Eng 51 26 14 8 64.7% 12.9 43.0 12.5 68.50 14 Illingworth R Eng 31 12 14 6 61.3% 12.3 41.5 13.8 67.64 15 Hassett A.L Aus 24 14 6 3 70.8% 14.2 44.1 8.8 67.03 16 Cronje W.J Saf 53 27 15 9 65.1% 13.0 40.8 11.6 65.49 17 Richardson R.B Win 24 11 7 3 60.4% 12.1 43.1 9.9 65.01 18 Ganguly S.C Ind 49 21 15 9 58.2% 11.6 37.2 14.6 63.45 19 Wasim Akram Pak 25 12 5 4 58.0% 11.6 38.6 13.0 63.16 20 Smith G.C Saf 67 33 15 14 60.4% 12.1 36.5 14.5 63.13

This table is provided only for information. Readers can draw their own conclusion. Only comment is on how high the position of Mike Brearley is in this list, fifth.

2. Match results adjusted for team quality and venue.

This measures the quality of the results achieved. Which team was beaten, what were the relative strengths of the teams, where was the match played (home or away). In general the winning team's captain gets credited with the losing team's Team Strength Index (TSI), already discussed in these columns. TSI is a composite of Batting, Bowling and Fielding strengths.

There are some adjustments. First the case of the stronger team winning. If the losing team is quite weak (TSI less than 75% of the winning team's TSI), there is a 10% reduction of points. (e-g) Australia defeating Bangladesh. Their captain will only get 90% of Bangladesh's TSI. On the other hand, New Zealand's captain will get full credit for defeating Bangladesh. If Australia defeats India/South Africa or vice versa, the winning captain will get credited with the losing team's TSI.

Now the case of the weaker team winning. If the winning team is quite weak (TSI less than 75% of the losing team's TSI), there is a 10% increase of points. (e-g) New Zealand defeating Australia. On the other hand, India/South Africa's captain will get Australia's TSI for defeating them.

Captains in drawn matches get 50% of the TSI values, subject to the above adjustments.

For away matches, an additional weighting of between 0% and 30% are given to the team winning away, depending on the relative strengths of the two teams.

The points for all tests captained by one player are summed and divided by the number of tests captained. This ensures that longevity in captaincy does not play a part.

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

  • neil retief on February 2, 2009, 14:43 GMT

    Ponting,s true grit as captain is only now being tested - Mcgrath,Warne ,Gilly and a few weeks ago,Hayden made a huge contribution to his record.Ponting,s recent record ,beaten in India and whipped by South Africa in their backyard is the beginning of a downward slide and i am interested to see what the future holds. It is early days,but already we see the effects of losing the big guns.It has a marked effect on Ponting,s record and i am convinced that there will be a big difference in the margin of victories with Warne and co and without.

  • Avi Singh on January 31, 2009, 2:03 GMT

    Hi,

    Enjoyed the analysis. However, when I was comparing Imran and Benaud's records with Dravid's in terms of performance indicators, it seems that the fact that they are all-rounders gives them a better chance in this section than someone like Dravid who is only a batsman. How did you account for this disparity?

  • David Barry on January 24, 2009, 0:05 GMT

    Run-outs, DVC, not not-outs! When you go to Statsguru bowlers and do the overall aggregate, it adds up all the bowling figures. Run-outs don't contribute to these.

    Wides and no-balls also have some effect.

  • D.V.C. on January 23, 2009, 15:53 GMT

    Let's say 11 Batsman score 220 runs. All 10(!) wickets fall. The bowling average is 220/10 = 22. The batting average is 220/10 = 22. They're the same. You might think it should be 220/11 for the batsman but if you are not out then that doesn't count as a wicket. How can a not out reduce the batting average?

    I think Tristan is right. Because, if in that same innings where the 11 batsman had scored 220 runs, the bowlers had conceded 20 extras in wides and no balls (note not leg-/byes because these don't count against the bowler) then the bowling average is then 240/10 = 24. The batting average is still 220/10 = 22.

    I feel I'm sticking my neck out here because both Ananth and David seem to think it's the not outs that make the difference. If I'm wrong though, I'd like an explanation.

  • David Barry on January 23, 2009, 4:50 GMT

    Just saw the comments on the difference between the overall batting and bowling averages. The difference is caused mostly caused by run outs, which reduce the batting average but not the bowling average.

  • Alex Fleming on January 21, 2009, 18:16 GMT

    I think one of the best Test captains ever was Adam Gilchrist, perhaps you could remind us of his figures. I also agree with many other ppl that Border, Lloyd and Fleming are the greatest captains ever, their longevity is testament to their superior leadership skills. I also agree with you Anath that we should reserve judgment untill the end of careers, because it now seems obvious to me that the truly greatest Captain of all time will be Graeme Smith. (being Aussie I still wish for a 3-0 away win) KP - what a shame, had the potential to rise to the top of this table/argument but now has only succeeded in conceding the Ashes 6 months early.

  • Tristan Hennessy on January 20, 2009, 23:26 GMT

    In reply to Ananth's answer to my question:

    [Tristan, let me explain the reason for the difference [in the overall batting average and the overall bowling average] in one word (okay two words), NOT OUTS.

    Ananth: Not outs should not make a difference, not only do they make the batting average rise but also the bowling average.

    But after some pondering over night I figured that the overall bowling average would be higher due to one word Ananth, SUNDRIES

  • Ananth (for Tristan) on January 20, 2009, 6:37 GMT

    For some inexplicable reason the following comment has not reached the approval platform. Hence I have taken the liberty of posting it myself and the reply. Name: Tristan Hennessy Comments:

    [The all-Test batting average of 29.92 and bowling average of 31.51 are used as reference]

    Out of interests sake, Can you please explain why the overall batting average and the overall bowling average are different?

    I can't understand why they would be different when a batting and bowling average is number of runs/dismissal [[ Tristan, let me explain the reason for the difference in one word (okay two words), NOT OUTS. Ananth: ]]

  • D.V.C. on January 19, 2009, 17:05 GMT

    I don't think that we should actually be including the captain's individual performance as a player in his captaincy rating. If the captain scores a century or takes 10 wickets he is already gaining benefit by helping his team to win.

  • Jeff on January 14, 2009, 8:59 GMT

    I agree that there’s not much more Ponting can have done so far as captain, but does continuing the success of his predecessor in captaining arguably the most talented team ever to consistent victories make Ponting one of the greatest captains ever? You can’t possibly say that with any certainty.

    In some ways Ponting has been cursed by leading such a great team. How much skill do you need to throw the ball to McGrath & Warne and say "bowl them out for me guys"? It’s far easier to demonstrate great leadership in a struggling team.

    I agree that we’ll now get a chance to see how good he really is. He might be a great captain – we just don’t know it yet.

    For the record, I think he might be a pretty good one – his declaration at Sydney was a bold move that paid off. And while it was technically in a “dead rubber”, if he’d lost he’d have had the stigma of being the 1st Aussie captain in 100+ years to be whitewashed at home, so it wasn’t without risk for him.

  • neil retief on February 2, 2009, 14:43 GMT

    Ponting,s true grit as captain is only now being tested - Mcgrath,Warne ,Gilly and a few weeks ago,Hayden made a huge contribution to his record.Ponting,s recent record ,beaten in India and whipped by South Africa in their backyard is the beginning of a downward slide and i am interested to see what the future holds. It is early days,but already we see the effects of losing the big guns.It has a marked effect on Ponting,s record and i am convinced that there will be a big difference in the margin of victories with Warne and co and without.

  • Avi Singh on January 31, 2009, 2:03 GMT

    Hi,

    Enjoyed the analysis. However, when I was comparing Imran and Benaud's records with Dravid's in terms of performance indicators, it seems that the fact that they are all-rounders gives them a better chance in this section than someone like Dravid who is only a batsman. How did you account for this disparity?

  • David Barry on January 24, 2009, 0:05 GMT

    Run-outs, DVC, not not-outs! When you go to Statsguru bowlers and do the overall aggregate, it adds up all the bowling figures. Run-outs don't contribute to these.

    Wides and no-balls also have some effect.

  • D.V.C. on January 23, 2009, 15:53 GMT

    Let's say 11 Batsman score 220 runs. All 10(!) wickets fall. The bowling average is 220/10 = 22. The batting average is 220/10 = 22. They're the same. You might think it should be 220/11 for the batsman but if you are not out then that doesn't count as a wicket. How can a not out reduce the batting average?

    I think Tristan is right. Because, if in that same innings where the 11 batsman had scored 220 runs, the bowlers had conceded 20 extras in wides and no balls (note not leg-/byes because these don't count against the bowler) then the bowling average is then 240/10 = 24. The batting average is still 220/10 = 22.

    I feel I'm sticking my neck out here because both Ananth and David seem to think it's the not outs that make the difference. If I'm wrong though, I'd like an explanation.

  • David Barry on January 23, 2009, 4:50 GMT

    Just saw the comments on the difference between the overall batting and bowling averages. The difference is caused mostly caused by run outs, which reduce the batting average but not the bowling average.

  • Alex Fleming on January 21, 2009, 18:16 GMT

    I think one of the best Test captains ever was Adam Gilchrist, perhaps you could remind us of his figures. I also agree with many other ppl that Border, Lloyd and Fleming are the greatest captains ever, their longevity is testament to their superior leadership skills. I also agree with you Anath that we should reserve judgment untill the end of careers, because it now seems obvious to me that the truly greatest Captain of all time will be Graeme Smith. (being Aussie I still wish for a 3-0 away win) KP - what a shame, had the potential to rise to the top of this table/argument but now has only succeeded in conceding the Ashes 6 months early.

  • Tristan Hennessy on January 20, 2009, 23:26 GMT

    In reply to Ananth's answer to my question:

    [Tristan, let me explain the reason for the difference [in the overall batting average and the overall bowling average] in one word (okay two words), NOT OUTS.

    Ananth: Not outs should not make a difference, not only do they make the batting average rise but also the bowling average.

    But after some pondering over night I figured that the overall bowling average would be higher due to one word Ananth, SUNDRIES

  • Ananth (for Tristan) on January 20, 2009, 6:37 GMT

    For some inexplicable reason the following comment has not reached the approval platform. Hence I have taken the liberty of posting it myself and the reply. Name: Tristan Hennessy Comments:

    [The all-Test batting average of 29.92 and bowling average of 31.51 are used as reference]

    Out of interests sake, Can you please explain why the overall batting average and the overall bowling average are different?

    I can't understand why they would be different when a batting and bowling average is number of runs/dismissal [[ Tristan, let me explain the reason for the difference in one word (okay two words), NOT OUTS. Ananth: ]]

  • D.V.C. on January 19, 2009, 17:05 GMT

    I don't think that we should actually be including the captain's individual performance as a player in his captaincy rating. If the captain scores a century or takes 10 wickets he is already gaining benefit by helping his team to win.

  • Jeff on January 14, 2009, 8:59 GMT

    I agree that there’s not much more Ponting can have done so far as captain, but does continuing the success of his predecessor in captaining arguably the most talented team ever to consistent victories make Ponting one of the greatest captains ever? You can’t possibly say that with any certainty.

    In some ways Ponting has been cursed by leading such a great team. How much skill do you need to throw the ball to McGrath & Warne and say "bowl them out for me guys"? It’s far easier to demonstrate great leadership in a struggling team.

    I agree that we’ll now get a chance to see how good he really is. He might be a great captain – we just don’t know it yet.

    For the record, I think he might be a pretty good one – his declaration at Sydney was a bold move that paid off. And while it was technically in a “dead rubber”, if he’d lost he’d have had the stigma of being the 1st Aussie captain in 100+ years to be whitewashed at home, so it wasn’t without risk for him.

  • Michael Brown on January 12, 2009, 18:45 GMT

    I would like to support Ponting, he has been given alot of flak but he has captained one one the most succesful sides ever while maintaining his position as one of the worlds top 5 batsman. What more can be asked of him so far.

    Of corse, now the real test of his abilities will begin

    [[ At last some support for Ponting. Mike's point about his pre-eminent position in world batting while being a captain is relevant. It is somewhat similar to the top player position held by Imran while he was captain. There are some interesting series ahead. If Ponting wins at Saf and wins the Ashes, quite a few voices will be silenced since he would have done with a diffent revitalized team. Similarly if Strauss wins at West Indies and wins the Ashes, he would prove Jeff's point that a less-performing but successful captain is back. Of course Strauss could suceed as a player also. Ananth: ]]

  • Jeff on January 12, 2009, 11:57 GMT

    I think it would take more than a few tweaks - I honestly think that captaincy is a bit too subjective/intangible to get something meaningful out of the stats available.

    On the subject of Strauss - I think that if he's still captain in the summer and leads England to Ashes victory then his place will be guaranteed for a good while, regardless of his own batting. Look at Vaughan - he averages 36 as captain (vs 50 in the ranks) but because he won consistently, and more importantly won the Ashes, his place was guaranteed while he was fit. And even now, despite him virtually not scoring a run in the past year, there are calls from journalists and ex-players to get him reinstated. This can only be for his perceived captaincy abilities.

    I think the difference between now and the past is that Brearley wouldn't get chosen in the first place. Selectors wouldn't take the "risk".

    btw - i'm only taking test cricket here - I agree that ODI/T20 captaincy is different.

  • Jeff on January 12, 2009, 10:09 GMT

    CONT… Conversely, if you average 20 with the bat and 100 with the ball but your team wins every series you play then you possibly have something to offer as captain (of course, as with the Aussies in recent years, it could be that the quality of the players is more important than the captains tactical/motivational ability.)

    Take Brearley as the prime example – he wasn’t worth his place as a batsman and yet his teams won more than you might expect. His strengths were his tactical thinking and his ability to get the best out of his players (eg, Botham & Willis) Would England have performed the same with another captain? Results before & after his tenure suggest not (although he did have Packer to help him for some of the results). Would most England fans want a new “Brearley” as captain? Someone who averages 20 but gets the best out of Flintoff & Harmison and wins back the Ashes? You bet.

    Thanks for the analysis – I might not agree with the conclusions but enjoyed reading it.[[ [[ Jeff, If you see the third table that completes takes away the performance factors bot based on only results. A few tweaks to that and you might very well get what you want. Although I personally feel that a captain such as Brearley would not last a0 matches today. If Strauss fails but delivers two series wins, won't there clamour for replacement of him. He could get by in Tests biut in ODIs/T20 I am almost certain that a non-playing captain would not last. Having said that I feel Mike Brearley was a once-in-100-years cricketer. Ananth: ]]

  • Jeff on January 12, 2009, 10:08 GMT

    Hi Ananth,

    What you have here is an analysis of the most successful captains of all-time, be it based on the record of the teams they captain, on their own personal performance as captain or a combination of all.

    Clearly, Ponting & Waugh are the most successful in terms of team performance and Sobers probably the most successful in terms of personal performance but do these facts make them the “best” captains? I’m not so sure.

    Would the Aussies have had the same record if Shane Warne or Ian Healy or Justin Langer had been captains instead of Ponting/Waugh? I’d say it’s probable. Did the Windies win a higher proportion of matches with Sobers in charge? The facts suggest not.

    I personally don’t think that individual performance should even be a factor in determining the best captain. You can average 100 with the bat and 20 with the ball, but if your team doesn’t win a match, then you are worthless as a captain…

  • Aditya Mookerjee on January 11, 2009, 8:31 GMT

    The facts and figures cannot be refuted. All what is written is undisputed. Perhaps, there is no need for anyone to refute. I would like to make an interesting observation. Mr Frank Worrell, was perhaps one of the greatest captains to grace the cricket field. I perhaps feel, that he was greater than Clive Lloyd. Sir Frank Worrell, intimated, that there was none better to captain the West Indies after him, than Sir Garfield Sobers. He did not take into consideration Sir Sobers cricketing abilities, but purely the estimation of Sir Sobers as a future captain. Facts and figures give security to forward one's views. But, perhaps, perceptions are not swayed by justification.

  • Sorcerer on January 11, 2009, 4:31 GMT

    I ardently followed Imran's ascendancy as a phenomenal player and skipper from '81 onwards and when he retired in '92 was not at all surprised by Benaud's stunning tribute that Pak had lost not one but an exceptional "3-in-1" player - one of the finest bowlers ever, arguably the greatest cricket skipper ever and a highly dependable batsman. After leading Pak to an astonishingly brutally fought 1-1 drawn Series in WI where he had captured 23 wickets in 3 Tests, Imran's bowling powers slowly waned (not a surprise since he was 37) but his batting got mightier and mightier.

  • Mick on January 11, 2009, 0:11 GMT

    I hold the unorthodox view, for an Australian, that Steve Waugh was a lousy captain, and that for one key reason: of the two jobs a captain has, managing his team, and development of strategy, he was wonderful at the former and inept at the latter. I'd love some way to separate these two components of the captaincy role. I think the 2-1 loss in India, and the draw in Australia was due largely to Waugh's inept strategy, e.g. enforcing the follow on in back to back tests therefore exhausting his strike bowlers. Then there was his miserable field settings without anyone straight forcing the bowlers to bowl wide of the stumps else give up a single on a ball defended back straightish each ball. I think his reputation is based on the team he inherited. Incidentally Vaughn is similarly inept. The Aussies were only in the 2005 Ashes due to his refusal to go for the throat when he should have. I laugh when the English hold him up as some wise old head. His is gutless captaincy par excellence.

  • Graeme Codrington on January 11, 2009, 0:07 GMT

    Thanks for an excellent article and some fantastic stats work. The comments - and your replies to them - have been insightful, too. Thanks so much.

    I agree with an early comment that any table the places Ricky Ponting high must be flawed. Not that your stats are flawed, but rather the assumptions about what it means to be a good captain are flawed. I'd like to join the list and add another possible weighting:

    * what was the expectation of success/failure compared to the actual success/failure of the team under the captain?

    Ricky Ponting has been lucky to be leading a team that has been unbeatable for a generation. That doesn't make him a good leader, it just means he was in charge of a good team. Compare him to Clive Rice who led South Africa after readmission - he led them to victories when no-one gave them a chance. A much better captain by far.

    This would be tough to measure, and is probably too subjective. But it's an important component of "who's the best ever"! [[ Graeme, you guys keep on bowling wrong-uns at me (and I love you all for it). Virtually impossible to do an "Expected results" prediction in an objective manner. Let us take the return series between Saf and Aus. Both have approximately equal TSI values (SAF-72: Aus: 70). Home for Saf is a plus. Australia's recent form is another plus for Saf. Smith is a plus and Ponting is a minus.Australia, smarting under a defeat is a plus for them. With some simulation work thrown in I could predict, say, a 2.05-0.95 result for Saf. That is possible for this one-off instance. However doing it for 540 series over 130 years and then compare with actual results - it is mind-boggling. But thanks for an out-of-the-box suggestion. Ananth: ]]

  • BlueEarthCitizen on January 10, 2009, 22:36 GMT

    Ananth, I agree with your response to my comment, except, that is not what I said: {{BEC, I cannot accept that the sub-continent teams should be given credit for winning away. It is becoming as difficult for other teams to win in the sub-continent. During 1960s/1970s the emphasis in the sub-continent was to avoid losses.}}

    I am saying sub-continental CAPTAINS should be given credit for winning away from the sub-continent, and non-sub-continental captains should be given credit for winning ON the sub-continent, where I agree it is harder for them.

  • Nico on January 10, 2009, 22:27 GMT

    I think the analysis is good. As with all statistics you only get the numbers and yes captaincy is about more than numbers but the subjective things are just that. Also I think people want S Fleming higher up but maybe they remember his ODI performances for NZ and there "outperformance" in that format. There are to many factors in ODI's (luck/pitch/toss/individual performance...) too complie a similar list. Tx Nico

  • Kartik on January 10, 2009, 21:25 GMT

    Ananth,

    India was not such a weak team in Lloyd's era (spin quartet and later Kapil) and certainly not under much of Richards' era (many draws at home, win in England, draw in Australia).

    Sri Lanka played very few tests in Richards' era and almost none in Lloyd's era. Today, Bangladesh plays a lot.

    Maybe choose a better picture of Imran, perhaps from his playing days. He does not look like an inspirational leader and greatest captain of all time in that picture. hehe

  • parthy on January 10, 2009, 20:58 GMT

    Due credit at 27th position? I cant think of a test series where Fleming's NZ have been throughly outplayed as NZ are being today. In 2000-01 when Oz were at their peak , if not for some terrible umpiring at Perth , NZ might have done what SA have done today 8 yrs earlier tat too against Warne, Glenn , Gilly and Steve Waugh. 2003 - NZ drew a test series in India with the Fab 4 and Kumble at the height of their powers. In SL against Murali or against Steyn n Ntini at SA , Fleming's NZ were never out of place. The point is he never had exceptional talent under him unlike Ponting , Waugh , Imran n Llyod. Yet his team were always up n running n fighters to the core. Such things can never be brought about through statistics. It s like you have Gilly, Hayden , Dravid , Inzy n Ponting - But when you watch Sachin n Lara u know u r seeing something special.

  • namya on January 10, 2009, 20:19 GMT

    to summarize it, this is YOUR idea of a good/great captain, given on your statistical (?)analysis, which is great.. all the best [[ What you say applies to ALL analysis. If you agree with the analysis, fine. If you do not agree or you make a suggestion which is not accepted, it becomes the analyst's analysis. Unfortunately everyone cannot be pleased. Ananth: ]]

  • Ash Zed on January 10, 2009, 19:00 GMT

    Yet another excallent analysis. Very absorbing and interesting reading !!

    Since you have set qualification level at minimum 30 test as captain, lots of other interesting names obviously do not qualify. One such name is WW Armstrong of Australia who played 50 tests but captained in only 10, winning 8 and other two were drawn. I wonder if he is somehow analysed in your list, where will he fit? I am sure top 5 !!His performance with the bat and ball was quite OK as captain.

    Eagerly awaiting for your next masterpiece.

  • Srini, Bangalore on January 10, 2009, 17:24 GMT

    Re your response to Nishant's comments - "a wonderful draw with the last pair batting for an hour to save the match will get very little credit." Why should a batting captain get credit for saving the match unless he is actually batting ? Conversely, since you are not penalizing a captain for losing a winnable match, you should not penalize him for drawing one.

  • Pollock on January 10, 2009, 17:11 GMT

    An excellent analysis. A vital aspect of captaincy is the ability to build a team. Suppose a captain inherited a weak side ( very low TSI) but then went on to build a very strong one (very high TSI), I feel the above analysis might actually work against him since he spent most of his tenure leading a strong side, giving him an advantage in terms of TSI. However, he needs to be given credit for building a strong side in the first place. Therefore it would be great if you could post a separate table which lists captains by the difference in TSI for their teams at the start and end of their captaincy tenures. Since a captain rarely captains a side for 30 matches or so at a stretch, you could look at this difference for sets of 10 consecutive matches. I feel Allan Border might top the charts for this one since he inherited a very weak Australian side in the mid 80s and turned it into a really strong one. Thank you! [[ Shree, not a bad suggestion. Will consider it along with a few other similar ones made. Ananth: ]]

  • Brian on January 10, 2009, 13:30 GMT

    My thoughts on a good captain revolve more around tactical nous, and inspiration. Obviously these two are largely unquanitifiable, but one idea I would have is the following:

    For all players who have played a certain number of games (5 or 10) under a specific captain, add up their runs and wickets while playing under that captain to determine an overall average for all players. Then for the same players, add up their runs and wickets under any other captains, and compare the two. I would also suggest that the bowling difference here is more key as both tactics and inspiration come into the equation for a captain's relationship with his bowlers, whereas batting is a more individual effort, possibly only influence by a winning mentality.

    This measure would account for how a captain improves the performance of those around him and for all bar the recent captains, how they suffer with the loss of that particular captain. I suspect the Brearleys and the like would come up the rankings [[ Good, but complicated idea. I could look at it as a separate analysis as I have mentioned in response to another reader's comment. Ananth: ]]

  • namya on January 10, 2009, 12:23 GMT

    You would not give Pietersen negative points for losing the Chennai test? Why? That was exactly my point..It was bad captaincy wasn't it to let a side win chasing 387.. surely not good captaincy one would say. And statsitically how does it make sense to compare captains across if he is not penalised for losses.. a fundamental flaw on a purely statistical basis one could argue Captain's individual performance: no wightings for home/away means a tendency to favor captains who have capatained more matches at home.. One can take Ponting's example here.. look at his average as a capatin at home and away.. there's a stark difference. If match and series results are adjusted for venue for standardisation why not individual performance?

    cheers

  • Vijay Sharma on January 10, 2009, 10:44 GMT

    Ananth,a suggestion: this is not on the article itself but comes thru as an observation from reading your more recent articles.I am noticing that you seem to be getting into a mode of "justifying" your analysis by answering many (if not all) of the user comments.I understand that a lot of these comments pose questions and hence you feel the need to answer them.Having said that, I feel you don't have to (my personal opinion).Answer only the ones which are obviously different/meaningful/add to the discussion else let them be.Answering most of the questions comes across as 'you were not initially confident about the analysis yourself and hence need to justify them'.Maybe I am wrong but I have noticed this pattern over the last few articles and not right from beginning(been reading your articles for years!).You are an amazing analyst and believe me you don't need to justify your analysis.You may or may not want to post this comment since it does not refer to this article per se. [[ Vijay, many thanks. This is one comment I must respond to since you have reflected what I myself have felt. After finishing the latest series of comments I wondered to myself "What am I doing, responding to each and every comment. Lot of these are repeats. I must be selective". Ananth: ]]

  • Vivek on January 10, 2009, 10:20 GMT

    A very good article which brings out some very interesting rankings. I would offer one suggestion here.

    Captaincy should only be a measure of success achieved while leading the side and the individual contribution should be given minimal (or zero) weightage. I would much rather have a captain who wins all matches and contributes little as a player rather than someone who contributes say 100 runs per match but does not win as many matches. Brearley is an obvious example who was considered a top captain as he was successful without being a top contributor.

    A very good analysis regardless.

  • Kartik on January 10, 2009, 9:04 GMT

    So Ponting would have been #1 until Oct 2008, before the India and SA losses recently. Those two series (4 losses and no win) as well as his own modest performance in them, pushed him lower.

    There should be a way to filter our minnow series'. That is only fair.

    Note that an all-rounder like Imran or Kapil has an inherent advantage in this table over a batsman or bowler.

    Also, Lloyd and Richards being so low does not seem quite right. They played almost no tests against minnows, unlike Ponting/Waugh. [[ Kartik, Your comments on Lloyd/Richards don't seem to be right. Look back to many teams including India and Sri Lanka around that time and they were very average teams. I understand that the bowler valuation seems to be a bit high. When that is corrected, the all-rounder valuation will automatically be corrected. The minniow series need not be pushed out. On the other hand they carry a considerably lower weight. Ananth: ]]

  • Cliff Hannah on January 10, 2009, 5:43 GMT

    As usual, a fascinating and informative analysis. I was wondering, though, how your findings would change if the only criteria was charisma. In other words, who were the most charismatic captains, and why? [[ Cliff, how does one define "charisma" in a measurable form ??? Ananth: ]]

  • Thomas on January 10, 2009, 5:25 GMT

    "Stephen Fleming has been given due credit for performing reasonably well with a middling team. Ananth:". Are you serious?? You think that not having Stephen Fleming even in the top 20 test captains is due credit!? Regardless of statistical analysis being at about 30th best test captain of all time is nowhere near due credit!

  • David Barry on January 10, 2009, 5:23 GMT

    I think that there is a problem if Ponting and Waugh come out as the top two in terms of pure captaincy ability. The last table is closely with percentage of wins, so any adjustments based on team strength don't appear to have worked properly.

    I will have my own post on this soon, in which I compare results to what you'd expect based on the differences in team strengths. [[ David, as and when you do your own analysis and, say, Ponting is 13th and Steve Waugh is 11th, a few people might agree. That does not make your analysis correct and mine wrong. Moreover you would have your own interpretation of team strength and I have mine.The Bangladesh TSI is 39.27 and South Africa's TSI is 72. Your values might be different. Consequently your analysis would be different. We would then have had two interpretations of a certain important measure. There is no "correct" interpretation. Ananth: ]]

  • Vice on January 10, 2009, 5:23 GMT

    Interesting but ideally you should be measuring the impact a captain has made, I suggest you weigh in the effect a captain had on the team win % against what it was before and after their tenure. A captain like Waugh should have against him the fact that the team he inherited was already winning a high % of matches (you could call it Taylor's "legacy", although the argument could easily come down to the quality of players) but in turn he was able to maintain and improve on that % (Waugh's "legacy" to Ponting). The problem with the analysis as it stands is that it is biased towards captains who inherited established and flourishing empires (Viv, Waugh & Ponting) but punishes those that had to forge teams out of wreckage (Border, Ranatunga, Hussain, Fleming) or those that worked with nothing (Streak, John Reid, Atherton) [[ These comments have been answered elsewhere. The points made are valid. Ananth: ]]

  • Geoff on January 10, 2009, 5:21 GMT

    As with your previous analyses, this is again very interesting. I agree with others that bowling performances are too highly weighted (just being a bowler shouldn't give +20 to 25 for a captaincy rating)- perhaps you could redress this by increasing weight for catches as most non-bowling captains field in the slips. Drawn matches should not be anywhere near 50% of a win, certainly for strong teams where a mark of a good captain is a declaration leading to a gettable 4th innings run chase by the opposition and then bowling them out on the last day even if that leaves the possibility of a loss. A great 10th wicket partnership saving a match has nothing to do with captaincy. On the other hand for weaker teams a draw could be regarded as being almost as good as a win against strong opposition so there should be some weighting factor introduced for draws. Finally there could be some consideration of the improvement in a team's TSI during a captaincy reign to quantify man management skills. [[ Most of these have been answered. I accept the comments on Drawn matches and that would be one of the tweaks, the other one would be to value the wickets captured slightly lower. That would also take care of the all-rounder valuation. Ananth: ]]

  • Shafiuddin A Kazi on January 10, 2009, 4:58 GMT

    I like the analysis and it has succeeded in deciding who the best captain of all time is and its a worthy Imran Khan but, the subsequent positions (Ricky Potning) are not acceptable to me. Captaincy is not just about results and performing well, its more than that. Ricky Ponting is not even the best captain around today, leave aside a top position in the all time list. Captaincy is about leading by example and being an inspirational leader at the same time. The more important thing is inspirational leadership there. 1 man inspiring 10 others to give their best is more important than him doing exceptinally but failing to inspire. Imran Khan, Ian Chappel, Mike Brearly, Richie Benaud, Allan Border, Frank Worrel, Mark Taylor, Shane Warne (T20 & 1st class) are all great captains and dont need an analysis to support their position. Its how you feel more than what numbers say. Do you disagree that Allan Border is responsible for Australian success of the recent past & Sourav of recent Indian? [[ Shafi, I have provided many answers and these should address your comments. There may be many detractors of Ponting. However I will say that a lot of these comments are influenced by Australia's recent performances. That should not take away from earlier results. "Inspirational" is a very subjective term. Ananth: ]]

  • BlueEarthCitizen on January 10, 2009, 4:54 GMT

    Ananth, good analyses and clearly you must be pleased with the generally "positive" comments. One thing to add here to Muthu's point about "away" victories. I think you could give higher points for sub-continental teams' victories away from the sub-continent and England/Aus/SAF/WI higher scores for victories on the sub-continent. This would be a good factor to judge how well the captains/teams performed in unfamiliar conditions. So, thats my rational input. [[ BEC, I cannot accept that the sub-continent teams should be given credit for winning away. It is becoming as difficult for other teams to win in the sub-continent. During 1960s/1970s the emphasis in the sub-continent was to avoid losses. Dead pitches ensured a very high degree of drawn matches. Leter spin took over and now even the pace bowlers are flourishing. I feel away is away barring the very strong and very weak teams. (Reposted through myself). Ananth: ]]

    On the emotional/judgement end of things, I just cannot accept that Ricky Ponting is so high on the scale. I will reserve judgement until I see someone analyzing the impact of the captain in raising a team's performance vs earlier, and seeing how they performed "away", i.e. Sub-Continent or Elsewhere. Thanks for the analyses.

  • Ananth on January 10, 2009, 2:52 GMT

    This is a common response to some comments on the weight given to performance measures and the high placement of all-rounders. It should be remembered that the overall weight for performance (bat/bow/fld) is around 25% and around what I aimed for. There are obviously individual variations, ranging from Sobers at 45% to Brearley at 10%. That is bound to happen. If, as John has suggested, I halve the performance weight, Sobers/Imran will come down to 20+% and may be acceptable. However many others would have very low performance weights (Ponting 9%, Steve Waugh 7%, Brearley 5% et al). This might very well allow these results-oriented captains to move up the table. But the weightings will be too low. Also the overall weight of 12+%, in my opinion, will be too low. My suggestion to the readers is this. I have presented three tables. People can take what they want. They could even mix amongst these. After all I have only presented a table with Imran on top and Sobers at 4. That is all.

  • Sean on January 10, 2009, 2:03 GMT

    Hi Ananth, following on from my earlier comment (Jan 9, 7 pm), surely it would not be impossible to quantify the transformative impact the captain had: for example by comparing the win/loss/draw percentage of the team for a period of time before the captain under consideration took over, and the percentage after (with win/loss/draw percentage weighted as you already do, for home and away wins, and against stronger and weaker teams, etc). Thanks for the article, I am sure it took a monumental effort. [[ Sean, what you say is possible. However not as a part of this specific article. The whole thing would become too complex. However I may very well do it (a captain's progress/regression through his career) as a separate exercise since a few readers, including you, have expressed interest. Thanks for the idea. Ananth: ]]

  • Anand on January 10, 2009, 2:00 GMT

    Ananth excellent analysis once again. IMHO a good captain is one who can lead by example and change the attitude of the team for the better. Imran Khan is excellent on both fronts and I agree that he being in the top is a deserved result. As a qualitative measure alone I would have believed Steve Waugh to be second in the list but Ponting is not a surprise as well because but for the last one year, Australia were steam rolling other teams and Ponting also achieved 16 wins in a row including a 5-0 ashes. The top 10 in the list looks well justified. One can say that Ponting and Richards took over a team that was no.1 but maintaining a no.1 position for a decade and captain contributing dueing that time arenttrivial tasks. I have a suggestion. Is it possible to obtain something like how a team's ranking progressed during a person's tenure as captain? For WI (80s) & Aus (mid 90's to now) one can take the difference between them and the second best. I dont know how feasible this is though. [[ Anand, impossible to get a handle on world rankings. There have been multiple ranking systems of varying quality and not comparable. Ananth: ]]

  • parthy on January 10, 2009, 1:54 GMT

    Good effort. Cant agree though. Ponting at No 2? Sobers at 4? A great captain cannot b measured by statistics.The best captain of our times has been Stephen Fleming. Just look how the NZ team s today. With due respect , NZ never had the same level of talent that almost all other teams had. But stephen was d lynch pin who held them together. Leadng a team of Gilly , shane n Glenn or Marshall , HOlding , Greendige and Viv is a herculean task as against what Fleming had to conjure. [[ Pathy, then you should read the excellent books on Test Captains by Richie Benaud and Mike Brearley. They wriite like Cardus. I only analyze, I do not write books. Stephen Fleming has been given due credit for performing reasonably well with a middling team. Ananth: ]]

  • waqas on January 10, 2009, 0:02 GMT

    well researched article.i m not surprised to see Imran on top of ur tables.although i do respect whatever the Ponting achieved as a captain but his true test came in when he has no more of his stalwarts(Warne,McGrath, langer) and he has started showing his lack of depth as a captain without them while Imran never had such big names(except Miandad)in his ammunation.Wasim was growing and Waqar came at very end of Imran's tenure.Ananth is it possible to include white washes(or black washes) and bouncing back in a series to win it after losing 1st match(i remember Arjuna and Salim malik done it) as a separate measureable aspect in calcualtions.i also think that captains of low ranked teams like Zimbabwe ,NewZealand(never been world powers especially in start),India and Pakistan (in their early days)should be given extra points for their success against giants of cricketing nations.although i m sure it is hard to sort out who is below rank team at a particular time.but overal agood article [[ Waqas, you can be rest assured that the captains of lower strength teams DO get credit for defeating stronger teams. I feel we should measure Ponting at the end of his tenure. He might surprise everyone by defeating South Africa away and then win teh Ashes (strong possibility looking at the disarray England is in). Ananth: ]]

  • namya on January 9, 2009, 23:55 GMT

    A few observations about this analysis are:

    1.For individual performances, how is the quality of the opposition decided? On the basis of the team TSI? Given the fact that countries across the world have struggle to come up with good bowling line ups (excepting Asutralia) over the past decade maynot be reflected correctly. Also the venue seems to have been ignored in the performance point calculations. (It's not mentioned in the methodology2. Shouldn't one be giving negative points in match/ series results adjusted for quality of opposition and venues? Because if a captain has lost a match/ series to a hugely inferior (in TSI terms) team, he doesn't pay for it. All that happens is that the opposition captain gets more points. If this was a comparison of teams ranking playing at the same time it wouldn't have mattered. But to compare across eras think negative points would be more sensible. [[ Namya, one thing I have never done is to factor in negative points. It is very unfair to the people involved and very difficult to implement. Pietersen cannot be assigned negative points for missing the Chennai win (maybe for his subsequent coup attempt). If ever a top team loses to Bangladesh (in this century ???) credit the Bangladeshi team but do not penalize the other team. Ananth: ]]

  • Muthu on January 9, 2009, 23:32 GMT

    To add to a prior point, Dravid's points while higher due to the rating of away wins, may also not be necessarily high enough - I'm not sure for an Indian team, beating SA in SA is only 10 points higher in worth. Of Dravid's eight wins, five were outside India (WI, SA - first ever, Eng, Pak - first ever and Bang) which translates to 63% of wins or 50% excluding Bang, while in Ganguly's case, 52% were away wins but only 24% excluding Zim and Bang. While I guess consistency is important, I think the value of away wins for Indian (and Sri Lankan) captions should be higher than for Aus or English captains. [[ All valuations are based on adjusted TSI figures. So they have the weight of authenticity behind them. How do you say that Indian wins should be rated higher than those for English (?) teams. Ananth: ]]

  • Marcus on January 9, 2009, 23:05 GMT

    Interesting analysis as always! One thing I've noticed though is that you have have runs per game as a performance indicator, when that would seem to disadvantage captains like Waugh, Border, Imran, Ganguly etc. who'd generally bat in the middle-lower order, and so wouldn't geberally have as much chance as top order players to score runs in the second innings. Perhaps runs per innings would be a better indicator?

    Also, I'd like to second John's comments on allrounders. While it's very impressive that Imran and Sobers managed to perform with bat and ball during their captaincies, they do seem to have beefed up their performance points and finished higher than captains with considerably better success rates.

    Anyway, thanks again.

    [[ Marcus, pl see an earlier response on the runs per test factor. Also I have sent a separate comment myself on the performance weighting. Ananth: ]]

  • Narayanan on January 9, 2009, 22:27 GMT

    The people management skills of a captain can also be factored in by comparing the average score of the team, individuals'(top x batsmen and bowlers in the team) during one's captaincy vs. career average.

    Field placement is another key and you can add # of catches and runouts to evaluate.

    It will be interesting if you can publish the data (excel) or Cricinfo exposes API to let us play with the data as well.

    [[ Narayanan, I do not agree. Is Dhoni to be held responsible for Dravid's run of low figures or Ponting for Hayden's run of failures. If that is so Dhoni should get credit for Gambhir's successes and Ponting for Clarke's run. Ultimately there should be only two serious contenders for measuring a captain's skills. How his team performs and how he performs. Ananth: ]]

  • Shaival Shah on January 9, 2009, 21:44 GMT

    It's very difficult to come up with such rankings purely on numbers. The rankings that you have come up with don't look out of place. That is the very reason why this effort needs to be lauded. I do agree with Geico Caveman about Ponting given a higher rank than he deserves. Two criteria that in my opinion should also be used to judge how good the captain is, are: 1. Team Quality: How good the team is 2. Improvement Factor: How much the team performance improved under the captain as compared to previous one(s).

    1 is very difficult if not impossible to quantize. 2 if quantized could be a very useful factor For example, Ponting has a higher ranking but he inherited an excellent team from Steve Waugh under whom the improvement in performance of the team improved much more as compared to that under Mark Taylor, than it did with Ponting as compared to Steve Waugh. If this could somehow be factored in to the rankings, it might make the rankings a little more comprehensive IMHO. [[ These points have been answered in other responses. Ananth: ]]

  • John Boon on January 9, 2009, 20:48 GMT

    What i find most interesting is the position of stephen fleming. Fleming was always thought of as a very thoughtful and inventive captain and also led NZ to a drawn test series in australia when they were at or near their peak. Great article.

  • Nishant on January 9, 2009, 20:32 GMT

    overall a good list, but I feel you gave too much weightage to drawn matches. A captain's ability comes through when he goes for the win. I feel you should weight a draw at 33% rather than the 50% you use now. [[ Nishant, I must confess that giving a weightage of 33% for drawn matches did not strike mel. This is the equivalent of the EPL method of 3 for a win and 1 for a draw. This is a good suggestion. My only concern is that a wonderful drwa with the last pair batting for an hour to save the match will get very little credit. Let me look at it. Thanks Ananth: ]]

  • Dr Manish Agrawal on January 9, 2009, 20:18 GMT

    I really commend the effort taken by you in compiling these stats.. however, i feel that for the performance ratings, the bowlers have got much more points than batsmen, e.g. B S Bedi - 4.5 wkts per match - 39.2 points, Walsh - 4 wkts per match - 34 points while Bradman - 135 runs per match - 27 points, J'wardane - 105 runs per match - 22 points. i dont agree that 2 wkts per innings is better performance than 65 runs per innings. If you go through the entire lists, all the bowler-captains have performance ratings in excess of 35, while the highest points for a batsman-captain is 27.8 (Bradman). [[ Manish, this is to adjust for the fact that there have been very few bowling captains and the total number of wickets captured by captains is far less, in comparable terms, to the number of runs scored. I agree that there is a slight bias towards the wickets measure. Pl note that this additional weighting is only for this analysis. Ananth: ]]

  • Gurudatt on January 9, 2009, 20:15 GMT

    Thank you for very interesting analysis, I have been reading this blog for a long time but had never posted any comments

    Though we can't account for emotions, momentum, imagination etc in Statistics.

    Is there a way to include what was the strength of the opposition at the time of the series (based on the then "form"). For this may be we can consider the trailing 12 or 18 months data.

    e.g. beating Australia today may be relatively easier than beating them one year back. Same way, beating WI in 70's and 80's was almost unthinkable.

    Also, it is anyway possible to find out, how many wins were series turning/winning or drawing efforts? That may make up for interesting statistics.

    Somehow I feel that to win for India in Perth was much tougher than it was for SAF to win in Perth.

    But, it may be tougher for SAF to win in Melbourne since Aus was 1-0 down, the same way it was tougher for India to win in Adelaide after the Perth win.

    Would be interesting to know if possible !

    Thank you

    [[ Gurudatt, it would be possible to factor in the current for. I have the required data. It would make the already complex analysis much more complex. For every test, I would have to look at the recent form of every player and build this in For instance Hayden would pull Australia down now as Dravid would India. Whether it is worth that or not is a moot point. Thanks for asking the query. Ananth: ]]

  • Shriram on January 9, 2009, 20:13 GMT

    Good analysis overall. However, does the TSI assigned to a team vary with time? For e.g. the West indies teams from 1980 - 1993 should have a much higher TSI than the West Indies teams post 2000. Therefore a win against the WI in 1985, for e.g. should count for a lot more than one in 2005. [[ Shriram, the TSI is a fairly accurate index, already discussed in these columns, of how strong a team was/is. In this current analysis the TSI is also weighted by the period values. Ananth: ]]

  • Manojkumar TN on January 9, 2009, 20:10 GMT

    While measuring the captains permormance, using the runs per test rather than per innings would be unfair for players like waugh. Austrailia won many matches in the Warne Mcgrath era when a player at no.5 or 6 would not have 2 innings per test regularly. [[ Manoj, I have not used runs per test per se. Each run scored by or wicket taken by a captain is weighted by the iopposition bowling/batting quality, summed and divided by the number of tests captained. The number of tests is a common factor running through. It would be incorrect to suddenly change that to innings or series, as I have mentioned. Ananth: ]]

  • Uday Chatterjee on January 9, 2009, 20:06 GMT

    Rather than revealing 'good captains' the analysis reveals 'captains who performed well on very successful teams.' A captain's role is to marshall limited resources to deliver better than expected results. However, these results are biased towards extraordinarily successful teams.

    A more meaningful (the only meaningful?) approach to the question is through subjective surveys / reviews.

  • Amarta on January 9, 2009, 19:27 GMT

    Hi Ananth..Great work..but have you taken into consideration the teams that these captains have captained i.e. the quality of the teams. For example, when ricky ponting took over, he had a team of champions under him so captaincy for him at that time would have been much easier that for him now, when he has a team of inexperienced new comers. So the true mettle of his captaincy should be judged by how he captains and how much his teams win now. Similarly when sourav ganguly took over the reins, indian cricket was in a state of turmoil with the match fixing issues, regionalism in team selections and hardly any test wins outside India. He mould a bunch of talented but non united people into a TEAM and India started winnign not only in india but also outside. Don't you think he deserves a upper ranking on captaincy terms rather than a ponting? Also I am astounded to see Allan Border so down in the list when we all know the foundation of the strong aussie teams of today was created by him. [[ Amarta, Some of the factors you mention have been covered by the Results analysis. If a captain wins with a weaker team and/or against a stronger team , he gets substantial extra credits. Ananth: ]]

  • Sean on January 9, 2009, 19:00 GMT

    And what about adjusting for what the results were BEFORE the captain took over, the inexperience of the squad that is typical during rebuilding.. basically how much of a TRANSFORMATION the captain achieved? A captain leading a team that has been consistent world beaters without him has less to do than someone who pulled the team to greater heights than previously imagined. Without this factor, I would consider this analysis, as any based purely on numbers, to be failry lacking. [[ Sean, I have always stayed away from areas on which I cannot back up my work. I will never be able to quantify some of the measures you have talked about. These are very relevant but belong in a book on captains written by Brearley or Benaud. Ananth: ]]

  • Ravi Gupta on January 9, 2009, 17:14 GMT

    Well, if we consider 25 matches and above, Dravid's index is better than Ganguly's....I am not surprised...in our country undemonstrative performers rarely get their due recognition... [[ That is certainly a good point. Although you will see that their results indices are almost the same but the difference of 2 points or so is in the performance. Ganguly's has been below par at 52 runs per test against Dravid's 75 runs per test. Ananth: ]]

  • Dom on January 9, 2009, 16:23 GMT

    That Imran is nowhere to be seen on the last table suggests they wouldn't have had the success they did without him,and conversely,the Aussies would without Ponting. [[ A very valid point. All the more reason why there has to be a significant weight for the captains performances as players. Ananth: ]]

  • Geico Caveman on January 9, 2009, 16:14 GMT

    I think this article, while well motivated by a desire to quantify captaincy, is a singular demonstration of how stats cannot tell the whole story, and in one glaring instance above, actually lie.

    Any ranking that places Ponting so high up, let alone above captains like Waugh, Taylor, Benaud or Imran, is a deeply flawed ranking. IMO, Ponting is a captain in Azhar's mold (without the match fixing scandals of course). When he comes up against hard opposition (India, South Africa, and England 2005), he visibly and literally wilts. His field placements are often wrong, late in devising, and he is very slow in reacting to a changing situation. Ponting, thus, would struggle to be even an average captain, let alone a good or a great one.

    I do not doubt your calculations. It is the results of those calculations that conflict with rather obvious reality. You need to refine your measures in some way. And many, including myself, even have doubts that that can be accomplished. [[ 1. Ponting is not placed above Imran, as you have suggested. 2. If 25 tests was the criteria, Benaud would have been at no.1. On these two points you are wrong. Coming to Ponting's captaincy, my analysis is based on the numbers, in other words, the team results and captain performance. It would be churlish to deny Ponting his team's successes when he has also contributed as a player. Whether he should be rated above Steve Waugh and Mark Taylor is a matter of personal opinion. The numbers suggest the order of the table. Individuals have the right to have their own choices. Incidentally nothing is obvious. Unfortunately if you see Ponting through Australia's current mixed results, he would seem to be below par. As Steve Waugh was when he let slip a potential 3-0 win in 2001 into a 1-2 loss. Ananth: ]]

  • John Fosbrook on January 9, 2009, 15:57 GMT

    Very interesting article, and an excellent analysis.

    My first comment is that the individual performances seem to count for rather too high a percentage of the overall score, and particularly favour all-rounders. I'd imagine that if the individual performance scores were all halved, then you would get a cross between the main table and the one excluding performance data, resulting in a list I would personally agree with the most. (I think Sobers is much too high in the first table, given that he won less than 25% of matches he captained.)

    My second topical point is that I think you should include a final score for "ability to get on with the team coach, the press and bureaucratic administrators". -50 for KP, methinks.

    Many thanks for a most interesting article. [[ Good suggestion. However please remember that the overall weight for the individual performances is only around 25%. Only Imran Khan and Sobers have somewhat higher figures, that too because of what they did on the field during their captaincy tenure. Ananth: ]]

  • No featured comments at the moment.

  • John Fosbrook on January 9, 2009, 15:57 GMT

    Very interesting article, and an excellent analysis.

    My first comment is that the individual performances seem to count for rather too high a percentage of the overall score, and particularly favour all-rounders. I'd imagine that if the individual performance scores were all halved, then you would get a cross between the main table and the one excluding performance data, resulting in a list I would personally agree with the most. (I think Sobers is much too high in the first table, given that he won less than 25% of matches he captained.)

    My second topical point is that I think you should include a final score for "ability to get on with the team coach, the press and bureaucratic administrators". -50 for KP, methinks.

    Many thanks for a most interesting article. [[ Good suggestion. However please remember that the overall weight for the individual performances is only around 25%. Only Imran Khan and Sobers have somewhat higher figures, that too because of what they did on the field during their captaincy tenure. Ananth: ]]

  • Geico Caveman on January 9, 2009, 16:14 GMT

    I think this article, while well motivated by a desire to quantify captaincy, is a singular demonstration of how stats cannot tell the whole story, and in one glaring instance above, actually lie.

    Any ranking that places Ponting so high up, let alone above captains like Waugh, Taylor, Benaud or Imran, is a deeply flawed ranking. IMO, Ponting is a captain in Azhar's mold (without the match fixing scandals of course). When he comes up against hard opposition (India, South Africa, and England 2005), he visibly and literally wilts. His field placements are often wrong, late in devising, and he is very slow in reacting to a changing situation. Ponting, thus, would struggle to be even an average captain, let alone a good or a great one.

    I do not doubt your calculations. It is the results of those calculations that conflict with rather obvious reality. You need to refine your measures in some way. And many, including myself, even have doubts that that can be accomplished. [[ 1. Ponting is not placed above Imran, as you have suggested. 2. If 25 tests was the criteria, Benaud would have been at no.1. On these two points you are wrong. Coming to Ponting's captaincy, my analysis is based on the numbers, in other words, the team results and captain performance. It would be churlish to deny Ponting his team's successes when he has also contributed as a player. Whether he should be rated above Steve Waugh and Mark Taylor is a matter of personal opinion. The numbers suggest the order of the table. Individuals have the right to have their own choices. Incidentally nothing is obvious. Unfortunately if you see Ponting through Australia's current mixed results, he would seem to be below par. As Steve Waugh was when he let slip a potential 3-0 win in 2001 into a 1-2 loss. Ananth: ]]

  • Dom on January 9, 2009, 16:23 GMT

    That Imran is nowhere to be seen on the last table suggests they wouldn't have had the success they did without him,and conversely,the Aussies would without Ponting. [[ A very valid point. All the more reason why there has to be a significant weight for the captains performances as players. Ananth: ]]

  • Ravi Gupta on January 9, 2009, 17:14 GMT

    Well, if we consider 25 matches and above, Dravid's index is better than Ganguly's....I am not surprised...in our country undemonstrative performers rarely get their due recognition... [[ That is certainly a good point. Although you will see that their results indices are almost the same but the difference of 2 points or so is in the performance. Ganguly's has been below par at 52 runs per test against Dravid's 75 runs per test. Ananth: ]]

  • Sean on January 9, 2009, 19:00 GMT

    And what about adjusting for what the results were BEFORE the captain took over, the inexperience of the squad that is typical during rebuilding.. basically how much of a TRANSFORMATION the captain achieved? A captain leading a team that has been consistent world beaters without him has less to do than someone who pulled the team to greater heights than previously imagined. Without this factor, I would consider this analysis, as any based purely on numbers, to be failry lacking. [[ Sean, I have always stayed away from areas on which I cannot back up my work. I will never be able to quantify some of the measures you have talked about. These are very relevant but belong in a book on captains written by Brearley or Benaud. Ananth: ]]

  • Amarta on January 9, 2009, 19:27 GMT

    Hi Ananth..Great work..but have you taken into consideration the teams that these captains have captained i.e. the quality of the teams. For example, when ricky ponting took over, he had a team of champions under him so captaincy for him at that time would have been much easier that for him now, when he has a team of inexperienced new comers. So the true mettle of his captaincy should be judged by how he captains and how much his teams win now. Similarly when sourav ganguly took over the reins, indian cricket was in a state of turmoil with the match fixing issues, regionalism in team selections and hardly any test wins outside India. He mould a bunch of talented but non united people into a TEAM and India started winnign not only in india but also outside. Don't you think he deserves a upper ranking on captaincy terms rather than a ponting? Also I am astounded to see Allan Border so down in the list when we all know the foundation of the strong aussie teams of today was created by him. [[ Amarta, Some of the factors you mention have been covered by the Results analysis. If a captain wins with a weaker team and/or against a stronger team , he gets substantial extra credits. Ananth: ]]

  • Uday Chatterjee on January 9, 2009, 20:06 GMT

    Rather than revealing 'good captains' the analysis reveals 'captains who performed well on very successful teams.' A captain's role is to marshall limited resources to deliver better than expected results. However, these results are biased towards extraordinarily successful teams.

    A more meaningful (the only meaningful?) approach to the question is through subjective surveys / reviews.

  • Manojkumar TN on January 9, 2009, 20:10 GMT

    While measuring the captains permormance, using the runs per test rather than per innings would be unfair for players like waugh. Austrailia won many matches in the Warne Mcgrath era when a player at no.5 or 6 would not have 2 innings per test regularly. [[ Manoj, I have not used runs per test per se. Each run scored by or wicket taken by a captain is weighted by the iopposition bowling/batting quality, summed and divided by the number of tests captained. The number of tests is a common factor running through. It would be incorrect to suddenly change that to innings or series, as I have mentioned. Ananth: ]]

  • Shriram on January 9, 2009, 20:13 GMT

    Good analysis overall. However, does the TSI assigned to a team vary with time? For e.g. the West indies teams from 1980 - 1993 should have a much higher TSI than the West Indies teams post 2000. Therefore a win against the WI in 1985, for e.g. should count for a lot more than one in 2005. [[ Shriram, the TSI is a fairly accurate index, already discussed in these columns, of how strong a team was/is. In this current analysis the TSI is also weighted by the period values. Ananth: ]]

  • Gurudatt on January 9, 2009, 20:15 GMT

    Thank you for very interesting analysis, I have been reading this blog for a long time but had never posted any comments

    Though we can't account for emotions, momentum, imagination etc in Statistics.

    Is there a way to include what was the strength of the opposition at the time of the series (based on the then "form"). For this may be we can consider the trailing 12 or 18 months data.

    e.g. beating Australia today may be relatively easier than beating them one year back. Same way, beating WI in 70's and 80's was almost unthinkable.

    Also, it is anyway possible to find out, how many wins were series turning/winning or drawing efforts? That may make up for interesting statistics.

    Somehow I feel that to win for India in Perth was much tougher than it was for SAF to win in Perth.

    But, it may be tougher for SAF to win in Melbourne since Aus was 1-0 down, the same way it was tougher for India to win in Adelaide after the Perth win.

    Would be interesting to know if possible !

    Thank you

    [[ Gurudatt, it would be possible to factor in the current for. I have the required data. It would make the already complex analysis much more complex. For every test, I would have to look at the recent form of every player and build this in For instance Hayden would pull Australia down now as Dravid would India. Whether it is worth that or not is a moot point. Thanks for asking the query. Ananth: ]]