THE CORDON HOME

BLOGS ARCHIVES
SELECT BLOG
September 14, 2008

Trivia - bowling

Good in isolation, great as a pair

Anantha Narayanan
A fine partnership: Glenn McGrath and Shane Warne leave the field, for possibly the last time in a Test in England, England v Australia, The Oval, September 12, 2005
 © Getty Images
Enlarge

To retain my sanity, I have to be away from ODIs and batting for some time. I do not want to hear the words "Richards" and "Tendulkar" for a few days. Hence my next article covers the forgotten species, Test bowlers.

Even though the batsmen always bat in pairs, it is while bowling that players are very effective operating in tandem. This article looks at Test bowler combinations who have bowled together most effectively, achieved more together than individually, won more and performed well away from home.

New table on % of Team wkts added.

Modified table covering % of Wins incorporated.

As normally done I have to set up some selection criteria. However this time there only two criteria, as explained below.

a. Both the bowlers should have captured a minimum of 100 wickets each. This means that the two bowlers would have played together for around 25 Tests which is a fair length for anybody's career. This will also ensure that the number of qualifying pairs will be kept to a managable size.

b. Both the bowlers should have bowled in the concerned Test. This ensures that in cases where an allrounder has played purely as a batsman, the Test will not be considered. A classic case is Imran Khan. He has played seven of his 86 tests purely as a batsman. In these Tests he has played alongside Sarfraz Nawaz and/or Abul Qadir. Without this condition, these Tests would dilute the Imran/Sarfraz and Imran/Qadir pairs.

Of course if Imran has played in a Test and bowled very few overs (such as the Bangalore Test of 1987, when the spinners Tauseef and Iqbal Qasim bowled 90% of the overs because of the pitch conditions) the Test will still be considered for inclusion.

Fleetingly I considered using an innings as a unit for determining the qualifying criteria and dismissed that in favour of a Test, which is the most acceptable and understandable unit of game.

With these dual conditions, the number of bowler pairs who qualify is 69. Let us look at this data in various ways.

1. By number of wickets captured:

The most basic measure of the bowling pair's performance.

Analysis of Test bowlers operating in tandem - Wickets captured

No Cty TstsWkts Player 1 Wkts Player 2 Wkts

1.Aus 104 1001 Warne S.K 513 & McGrath G.D 488 2.Slk 92 875 Muralitharan M 572 & Vaas WPUJC 303 3.Win 95 762 Walsh C.A 373 & Ambrose C.E.L 389 4.Pak 61 559 Wasim Akram 282 & Waqar Younis 277 5.Saf 93 547 Pollock S.M 377 & Kallis J.H 170 6.Ind 53 498 Kumble A 281 & Harbhajan Singh 217 7.Saf 63 490 Pollock S.M 235 & Ntini M 255 8.Saf 83 488 Kallis J.H 170 & Ntini M 318 9.Aus 58 484 McGrath G.D 274 & Gillespie J.N 210 10.Eng 60 476 Willis R.G.D 222 & Botham I.T 254

Warne/McGrath stand on top with 1001 wickets in 104 Tests, followed by Muralitharan/Vaas with 875 wickets in 92 Tests. One from a very successful team and the other from a team which has had mixed results. Muralitharan and Vaas captured the bulk of wickets secured by Sri Lanka while Warne and McGrath shared the spoils with, on an average, two bowlers. While Warne and McGrath are almost equal in their tally, Muralitharan has captured nearly double the wickets that of Vaas.

However, these are not true "bowling in tandem" pairs since these are two different types of bowlers. The next few pairs are similar type bowlers and headed by two truly great opening bowling combinations, Walsh /Ambrose and Wasim/Waqar. There is no doubt these two pairs would have acted in tandem in most of these Tests. Then there are a few fast bowling pairs from different countries.

However the interesting pair is Kumble/Harbhajan. This is the only pure spin bowling pair in the top 10. It is certain that they would have bowled together for long spells.

To view the complete table, please click here.

2. Bowling Average.

Here the individual bowling averages are analysed. In other words, how the individual bowler's bowling average improved or declined while bowling with the other specified bowler. Did the bowler gain or not.

Analysis of Test bowlers operating in tandem - Bowling Average comparisons

No Cty Wkts Bowler 1 Wkts Avge CarAvg Diff Bowler 2 Wkts Avge CarAvg Diff

1.Aus 1001 Warne S.K 513 24.88 25.42 0.54 McGrath G.D 488 21.38 21.64 0.26 2.Slk 875 Muralitharan 572 21.80 21.96 0.16 Vaas WPUJC 303 27.87 29.31 1.44 3.Win 762 Walsh C.A 373 24.29 24.44 0.15 Ambrose C.E.L 389 21.12 20.99 -0.13 4.Pak 559 Wasim Akram 282 21.33 23.62 2.29 Waqar Younis 277 22.93 23.56 0.63 5.Saf 547 Pollock S.M 377 22.62 23.12 0.50 Kallis J.H 170 33.49 31.23 -2.26 6.Ind 498 Kumble A 281 27.79 29.33 1.54 Harbhajan Singh 217 32.22 30.87 -1.35 7.Saf 490 Pollock S.M 235 24.71 23.12 -1.59 Ntini M 255 27.67 28.22 0.55 8.Saf 488 Kallis J.H 170 32.65 31.23 -1.42 Ntini M 318 28.68 28.22 -0.46 9.Aus 484 McGrath G.D 274 20.52 21.64 1.12 Gillespie J.N 210 26.28 26.14 -0.13 10.Eng 476 Willis R.G.D 222 24.63 25.20 0.57 Botham I.T 254 25.67 28.40 2.73

Most of the bowlers in the top-10 pairs improved while bowling with the other bowler. The surprising result is with the spin bowling pair of Kumble/Harbhajan. Kumble gains significantly with 1.54 while Harbhajan drops with an equally significant 1.35. Why, I do not have an answer. Possibly because Kumble bowls quite tightly and the batsmen attack Harbhajan more. Note how Botham's average improves dramatically when bowling with Bob Willis.

To view the complete table, please click here.

3. Wickets per Test:

Here I have not considered the bowlers individually. Rather the two are considered as an integral unit and their performance together is compared with their career performance.

Analysis of Test bowlers operating in tandem - Wickets per Test as a pair

No Cty TstsWkts Bowler 1 Bowler 2 Pair Career Diff W/T W/T

1.Pak 61 559 Wasim Akram & Waqar Younis 9.16 (8.27) 0.90 2.Eng 49 367 Flintoff A & Hoggard M.J 7.49 (6.64) 0.85 3.Ind 33 251 Kapil Dev N & Doshi D.R 7.61 (6.77) 0.84 4.Pak 38 313 Wasim Akram & Mushtaq Ahmed 8.24 (7.54) 0.70 5.Ind 42 368 Chandrasekhar B.S & Bedi B.S 8.76 (8.14) 0.62 6.Eng 60 476 Willis R.G.D & Botham I.T 7.93 (7.37) 0.57 7.Win 60 418 Sobers G.St.A & Gibbs L.R 6.97 (6.44) 0.53 8.Pak 35 275 Sarfraz Nawaz & Imran Khan 7.86 (7.33) 0.53 9.Eng 40 243 Giles A.F & Flintoff A 6.07 (5.59) 0.48 10.Aus 27 263 McGrath G.D & MacGill S.C.G 9.74 (9.27) 0.47 ... ... 67.Aus 30 212 Gillespie J.N & Lee B 7.07 (7.90) -0.83 68.Pak 34 239 Imran Khan & Wasim Akram 7.03 (8.09) -1.06 69.Pak 36 265 Waqar Younis & Saqlain Mushtaq 7.36 (8.53) -1.17

Wasim Akram/Waqar Younis have performed as the best bowling pair operating in tandem in this measure. In their career they have captured a respectable 8.27 wickets per Test. However when they bowl together this figure increases to a very good figure of 9.16. Flintoff and Hoggard bowled very well together rather than separately as did Kapil Dev and Dilip Doshi. The spin twins of Bhagwat Chandrasekhar and Bishan Bedi come next.

At the other end, there are a few interesting combinations. The two Pakistani pairs Waqar/Saqlain and Imran/Wasim bowled much better separately than together. It is amazing that Wasim and Waqar are great together but fall off with two other bowlers respectively. Surprisingly Gillespie and Lee also lost out when bowling together.

To view the complete table, please click here.

4. Wins achieved:

This is an important measure. Did the pair do better than the team performance over the years.

Analysis of Test bowlers operating in tandem - Wins achieved

No Cty TstsWkts Bowler 1 Bowler 2 Wins % Win % Diff

1.Aus 25 227 MacGill S.C.G Lee B 21 84.0 (46.8) 37.2 2.Aus 45 371 McGrath G.D Lee B 36 80.0 (46.8) 33.2 3.Aus 45 421 Warne S.K Lee B 35 77.8 (46.8) 30.9 4.Aus 27 263 McGrath G.D MacGill S.C.G 20 74.1 (46.8) 27.2 5.Ind 37 287 Harbhajan Singh Zaheer Khan 18 48.6 (22.6) 26.1 6.Win 29 263 Marshall M.D Ambrose C.E.L 17 58.6 (33.7) 24.9 7.Win 33 291 Holding M.A Marshall M.D 19 57.6 (33.7) 23.9 8.Ind 29 249 Kumble A Pathan I.K 13 44.8 (22.6) 22.3 9.Win 36 322 Garner J Marshall M.D 20 55.6 (33.7) 21.8 10.Aus 104 1001 Warne S.K McGrath G.D 71 68.3 (46.8) 21.4 ... ... 67.Ind 79 394 Kapil Dev N Shastri R.J 10 12.7 (22.6) -9.9 68.Aus 30 211 Lindwall R.R Benaud R 11 36.7 (46.8)-10.2 69.Aus 41 346 Benaud R Davidson A.K 15 36.6 (46.8)-10.3

This part of analysis really stumped me. The reason is that the Team win % is an average of over or under 100 years of Test cricket. Australia has played 696 Test matches during this period and won 326 Tests leading to a Win % of 46.8. This is a misleading figure since this includes umpteen Australian teams (Pre-WW1 years, late 1930s, early 50s, 70s and 80s) which were quite weak. These more than offset the other successful teams of 1920s, 1940s, 1960s and recent ones. But I have no pat solution. Australia and England have played for 130 years while Pakistan for 50-odd years, Sri Lanka for 20 years and Bangladesh for 10 years. Where is the common dividing point. Maybe the solution is to look at a quartile as a comparison unit. Readers are welcome to offer their suggestions. Let us now look at the results.

MacGill and Lee form the most successful pair in this regard, achieving 37% over the team numbers. Lee pairs with three bowlers and takes the first three places. The Indian's recent resurgence is shown by the high placing of Harbhajan/Zaheer and Kumble/Pathan. These two pairs have achieved 26% and 22% over and above India's overall results.

At the other end there are Kapil/Shastri who were bowling in tandem during the very unsuccessful 1980s period for India. Similarly Lindwall, Benaud and Davidson were part of moderate Australian sides of the 1950s.

Just to give the complete picture I have given the Team performance summary here.

Cty <-----Total----->   <------Home----->   <-----Away------>
Tests Wins    %     Tests Wins    %     Tests Wins    %

Aus 696 326 46.84 361 202 55.96 335 124 37.01 Bng 53 1 1.89 25 1 4.00 28 0 0.00 Eng 877 306 34.89 447 175 39.15 430 131 30.47 Ind 421 95 22.57 213 64 30.05 208 31 14.90 Nzl 342 65 19.01 166 41 24.70 176 24 13.64 Pak 335 103 30.75 141 54 38.30 194 49 25.26 Saf 336 115 34.23 181 74 40.88 155 41 26.45 Slk 180 54 30.00 90 38 42.22 90 16 17.78 Win 448 151 33.71 201 74 36.82 247 77 31.17 Zim 83 8 9.64 44 6 13.64 39 2 5.13 Icc 1 0 0.00 0 0 0.00 1 0 0.00

To view the complete table, please click here.

5. Away wickets captured:

Analysis of Test bowlers operating in tandem - Home / Away wickets

No Cty Wkts Bowler 1 Wkts Awy Hme Bowler 1 Wkts Awy Hme

1.Aus 1001 Warne S.K 513 249 264 McGrath G.D 488 248 240 2.Slk 875 Muralitharan M 572 203 369 Vaas WPUJC 303 138 165 3.Win 762 Walsh C.A 373 182 191 Ambrose C.E.L 389 201 188 4.Pak 559 Wasim Akram 282 170 112 Waqar Younis 277 166 111 5.Saf 547 Pollock S.M 377 162 215 Kallis J.H 170 70 100 6.Ind 498 Kumble A 281 80 201 Harbhajan Singh 217 65 152 7.Saf 490 Pollock S.M 235 96 139 Ntini M 255 75 180 8.Saf 488 Kallis J.H 170 77 93 Ntini M 318 100 218 9.Aus 484 McGrath G.D 274 173 101 Gillespie J.N 210 117 93 10.Eng 476 Willis R.G.D 222 84 138 Botham I.T 254 104 150

Warne/McGrath have captured nearly as many wickets at home as away leading to great results for Australia everywhere. Muralitharan has captured far more at home than away. Surprisingly Wasim and Waqar have captured majority of their wickets away from home, possibly relishing the more favourable conditions outside the subcontinent.

To view the complete table, please click here.

6. Away wins achieved:

Analysis of Test bowlers operating in tandem - Away wins achieved

No Cty Tsts Bowler 1 Bowler 2 <-------Away-------> TstsWins % TWin % Diff

1.Ind 29 Kumble A Pathan I.K 15 8 53.3 (14.9) 38.4 2.Aus 45 Warne S.K Lee B 24 18 75.0 (37.0) 38.0 3.Aus 25 MacGill S.C.G Lee B 8 6 75.0 (37.0) 38.0 4.Aus 45 McGrath G.D Lee B 19 14 73.7 (37.0) 36.7 5.Aus 30 Gillespie J.N Lee B 18 13 72.2 (37.0) 35.2 6.Aus 53 Warne S.K Gillespie J.N 37 24 64.9 (37.0) 27.9 7.Aus 38 Miller K.R Johnston W.A 17 11 64.7 (37.0) 27.7 8.Ind 37 Harbhajan Singh Zaheer Khan 22 9 40.9 (14.9) 26.0 9.Aus 37 Lindwall R.R Johnston W.A 16 10 62.5 (37.0) 25.5 10.Pak 40 Waqar Younis Mushtaq Ahmed 28 14 50.0 (25.3) 24.7 ... ... 67.Ind 52 Kumble A Srinath J 23 1 4.3 (14.9) -10.6 68.Win 26 Garner J Croft C.E.H 15 3 20.0 (31.2) -11.2 69.Eng 35 Statham J.B Trueman F.S 12 2 16.7 (30.5) -13.8

Kumble and Pathan have been part of a great Indian away run in West Indies, Pakistan, Australia, Bangladesh and Zimbabwe. A 50+% result is way ahead of India's overall dismal away performance. Lee stars in the next four, with four different bowlers. Harbhajan/Zaheer have also done quite well away, though not as spectacularly as Kumble/Pathan. Incidentally this is Pathan's entire Test career and he just about managed to reach the 100-wicket mark.

At the other end Kumble/Srinath were part of the Indian teams of the 1990s when the Indians could win nothing while travelling. Statham and Trueman were part of the average 1960s England. The surprise is Garner and Croft who together won only 3 out of 15 matches away. This was due to the not very successful tours of West Indies to England, Pakistan and New Zealand.

To view the complete table, please click here.

7. % of Team wickets captured:

This is based on Pavan's request for analyzing the % of team wickets captured. I have limited to just analyzing the % of team wickets captured without bringing the other subjective factors.

Analysis of Test bowlers operating in tandem - % of Team Wickets captured

No Cty TstsWkts Player 1 Wkts Player 2 Wkts TeamWkts & %

1.Slk 92 875 Muralitharan M 572 Vaas WPUJC 303 1738 50.3 2.Aus 104 1001 Warne S.K 513 McGrath G.D 488 2016 49.7 3.Aus 27 263 McGrath G.D 121 MacGill S.C.G 142 531 49.5 4.Ind 53 498 Kumble A 281 Harbhajan Singh 217 1020 48.8 5.Aus 45 421 Warne S.K 246 Lee B 175 862 48.8 6.Pak 61 559 Wasim Akram 282 Waqar Younis 277 1167 47.9 7.Aus 25 227 MacGill S.C.G 119 Lee B 108 477 47.6 8.Aus 35 307 McDermott C.J 141 Warne S.K 166 660 46.5 9.Win 36 322 Garner J 152 Marshall M.D 170 700 46.0 10.Ind 29 249 Kumble A 149 Pathan I.K 100 545 45.7 ... ... 65.Saf 93 547 Pollock S.M 377 Kallis J.H 170 1792 30.5 66.Saf 83 488 Kallis J.H 170 Ntini M 318 1609 30.3 67.Eng 49 280 Underwood D.L 161 Greig A.W 119 941 29.8 68.Aus 43 241 Miller K.R 135 Johnson I.W 106 834 28.9 69.Ind 79 394 Kapil Dev N 243 Shastri R.J 151 1464 26.9

The top two Spin/Pace pairs have captured either side of 50% of the team wickets. It is quite surprising to see this with Australia's strong bowling line up. MacGill features in couple of top pairs. Kumble and Harbhajan have again captured nearly 49% as also Akram and Younis.

At the other end surprisingly the two South African pairs have captured only around 30%. It must be remembered that Pollock might appear in two pairs. Kapil Dev and Shastri prop up the list.

8. Win analysis - Revised:

This is based on Avi/Alex/Peter's suggestions to modify the Wins achieved analysis. They have in turn put in their suggestions based on my request.

Analysis of Test bowlers operating in tandem - Wins achieved (Alternate)

No Cty Bowler 1 Bowler 2 Pair T/W/% Ctry T/W/% Diff

1.Win Gibbs L.R & Hall W.W 37 19 51.4 97 32 33.0 18.4 2.Eng Trueman F.S & Lock G.A.R 27 15 55.6 148 59 39.9 15.7 3.Win Marshall M.D & Ambrose C.E.L 29 17 58.6 180 78 43.3 15.3 4.Aus McGrath G.D & Lee B 45 36 80.0 166 108 65.1 14.9 5.Aus Warne S.K & Lee B 45 35 77.8 186 117 62.9 14.9 6.Eng Caddick A.R & Gough D 30 13 43.3 118 34 28.8 14.5 7.Aus MacGill S.C.G & Lee B 25 21 84.0 119 83 69.7 14.3 8.Eng Willis R.G.D & Botham I.T 60 25 41.7 223 61 27.4 14.3 9.Ind Harbhajan Singh & Zaheer Khan 37 18 48.6 104 36 34.6 14.0 10.Win Marshall M.D & Walsh C.A 42 23 54.8 190 79 41.6 13.2 ... ... 60.Pak Waqar Younis & Saqlain Mushtaq 36 14 38.9 119 50 42.0 -3.1 61.Eng Statham J.B & Trueman F.S 35 13 37.1 136 56 41.2 -4.0 62.Eng Snow J.A & Underwood D.L 29 8 27.6 164 53 32.3 -4.7 63.Ind Kapil Dev N & Shastri R.J 79 10 12.7 132 24 18.2 -5.5 64.Pak Imran Khan & Wasim Akram 34 9 26.5 221 71 32.1 -5.7 65.Eng Underwood D.L & Willis R.G.D 37 10 27.0 173 57 32.9 -5.9 66.Saf Ntini M & Nel A 34 15 44.1 116 59 50.9 -6.7 67.Ind Kumble A & Srinath J 52 13 25.0 156 52 33.3 -8.3 68.Win Garner J & Croft C.E.H 26 9 34.6 79 37 46.8 -12.2 69.Aus Lindwall R.R & Benaud R 30 11 36.7 98 50 51.0 -14.4

This time the country win % is achieved in a different manner. Instead of taking a 100+ years span, I have taken the span as between the first test of the bowler in the pair who made his debut first and the last test played by either of the bowlers in the combination. I have taken this finite span rather than an year earlier and later as suggested by Alex since I wanted to avoid any subjective decisions. Effectively we now compare with the span of the careers of the two bowlers and the comparisons are far more meaningful.

There are significant changes at the top. The Australians effectively move down and are replaced by less fancied pairs. Also the differential is far less pronounced now. At the other end two pairs, Kumble/Srikkanth and Lindwall/Benaud remain and the West Indian pair of Garner/Croft get in. The reason has been explained elsewhere.

Let me conclude by saying that this analysis is not to determine the best bowling pair. I have just presented various aspects of bowlers bowling together and pointed out to some great and some not-so-great performances. Different bowlers have achieved greatness in different measures operating together and the article brings these to light.

Finally five pairs stand out. The Spin/Pace pairs of Warne/McGrath and Murali/Vaas. Then the two great fast bowling combinations of Walsh/Ambrose and Wasim/Waqar. Finally the spin duo of Kumble/Harbhajan. Out of these only two pairs are still operating and might go on to improve their results.

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

RSS Feeds: Anantha Narayanan

Keywords: Trivia

© ESPN Sports Media Ltd.

Posted by karthik on (October 4, 2008, 17:27 GMT)

A pair which i think was highly lethal or could have become highly lethal were wasim akram and waqar . Also we need to think about the great formation of Garner , Marshall , Holding and Croft . It was a pity they all played at the same time .

Posted by Charindra on (October 1, 2008, 14:58 GMT)

Warne/Mcgrath is possibly the greatest bowling pair ever, but the fact that Vaas benefitted by bowling in tandem with Murali shows that Murali had to do most of the work while Vaas just kept the lid on most of the times. But Warne had Mcgrath at the other end to create the pressure and that helped him take wickets. Murali/Vaas combo would have been the greatest pair ever if only Vaas was a bit more potent because at the other and he had the greatest spinner of all time.

Posted by Michael Brown on (September 28, 2008, 11:38 GMT)

I would be interested to see an analysis of 3 bowlers because Kallis, pollock and Ntini would surely dominate that.

Posted by PasserBy on (September 22, 2008, 22:58 GMT)

Hi Ananth,

Good analysis. I have one question. In table 1 I see that Warne has 513 wickets while McGrath has 488. It seems from the numbers that the 513 wickets captured by Warne are in those Tests where both he and McGrath bowled and the same goes for McGrath. But it doesn't automatically follow from that Warne took 513 wickets while McGrath was bowling from the other end. Of the 513 wickets Warne took some may have been taken with another bowler bowling in tandem with Warne.

I think this analysis would be improved if for pairs under consideration only those wickets and overs for a bowler should count when the other bowler in the pair bowls the immediate preceding or following over.

This would be trivial (but not necessarily easy) to do for matches with detailed over-by-over records (which could cover all recent matches). Older matches, especially pre WW-II may be more difficult.

Thanks.

P.S. Need a bigger character count limit for comments :) [[ The analysis you are asking for needs ball-by-data which is available only for the past 10 years. Even there the data is proprietary in nature. I did a lot of work with Wisden Online in capturing the ball-by-ball data. Even then I may not have access to the same. ]]

Posted by Haliastur Indus on (September 21, 2008, 7:54 GMT)

Another excellent research work & article by Ananth. Thanks for feeding the fanatics like us, who love to look at cricket (& its statistics) in a whole new perspective all the time.

Now to the article: You claim right at the beginning that "Muralitharan and Vaas captured the bulk of wickets secured by Sri Lanka while Warne and McGrath shared the spoils with, on an average, two bowlers". {This has been an argument thrown by so many people in user forums, to support their argument, without relevant stats}. But, when you include item 7 (as per Pavan's request), we see that there's not much truth in that claim. In fact, Murali/Vaas & McGrath/Warne share almost (differed by a mere 0.6%) a similar amount of opposition wickets. Proof that some "statistics" are illusions created by commentators/fans alike. [[ I think your point is correct. However I made a general statement that Murali/Vaas had very little support (and consequently sharing of spoils) while Warne/Murali had Lee/Gillespie/Kaspro/MacGill/Clark at the other side. However it would be a fascinating study to do a deeper analysis on some of these specific situations. ]]

Posted by atif hussain on (September 20, 2008, 22:47 GMT)

this is a good effort by cricinfo team,but i must say that you have to look at the team as a whole too,thier batting strenth thier fielding because it matters a lot,for example in akram/younis's case we all know that how "good" pakistan's fielding is.warne/glen has had a great fielding side as thier team's batting.i still remember pakistani fielders who use to drop sitters in w's bowling and at the other hand glen have more of his wickets in slips.i can bet that glen would have 100 or more less wickets if australia'a fielding were as bad as pakistan's.

Posted by hastagiri on (September 15, 2008, 11:19 GMT)

i had 2 suggestions here: 1) in each of the victories of the team, you can get a wickets/bowler average (with a minimum overs% for the bowler to be considered) and check how the pair has differed from the overall average. this would give an idea of dominance of this pair over the others in the pack during their time and otherwise.

2)difference in averages of the pair between won tests, drawn tests and lost tests would give an idea of the effect the pair had on the overall result of their teams' success

Posted by Alex M on (September 15, 2008, 7:10 GMT)

Interesting piece.

I agree with comments by Avi Singh and Peter Banks.

I would add one minor tweak to Avi Singh's suggestion ("perhaps use the team win% for the total span of the 2 players' careers")

I say the team win % stat should cover a period beginning 1 year before the first player's debut. And (for retired players) the time span should end 1 year after the 2nd retiree's last match.

i.e. Player A played jun 1990 to aug 2000 Player B played aug 1991 to aug 2001

then, calculate Team's win percentage from jun 1989 (not 1990) to aug 2002 (not 2001).

This data would be harder to gather, but should be a better indicator of the players' impact on their team, by considering team conditions just before/after their careers. It would also increase the sample size in cases where the players had few interruptions to career due to injury, etc. [[ As I have already mentioned in my reply to Avi, quite tough. But I will see what can be done. If not immediately certainly worth a later post. ]]

Posted by Ananth on (September 15, 2008, 4:06 GMT)

% of Team wickets captured (New table added) This is based on Pavan's request for analyzing the % of team wickets captured. I have limited to just analyzing the % of team wickets captured without bringing the other subjective factors. Ananth

Posted by heath on (September 15, 2008, 0:53 GMT)

Jonathan Ellis, was thinking the same thing, about overlap, with the point that akram and imran are right down the bottom of wickets per test, despite both being great bowlers. It is because akram emerged when imran would have been 35ish, and bowling less, in fact, Imran would have bowled less, no doubt, because he had akram in his team.

Comments have now been closed for this article

ABOUT THE AUTHOR

Anantha Narayanan
Anantha spent the first half of his four-decade working career with corporates like IBM, Shaw Wallace, NCR, Sime Darby and the Spinneys group in IT-related positions. In the second half, he has worked on cricket simulation, ratings, data mining, analysis and writing, amongst other things. He was the creator of the Wisden 100 lists, released in 2001. He has written for ESPNcricinfo and CastrolCricket, and worked extensively with Maruti Motors, Idea Cellular and Castrol on their performance ratings-related systems. He is an armchair connoisseur of most sports. His other passion is tennis, and he thinks Roger Federer is the greatest sportsman to have walked on earth.

All articles by this writer