THE CORDON HOME

BLOGS ARCHIVES
SELECT BLOG
August 18, 2010

Test cricket

The richest Test teams ever

Anantha Narayanan

ICC World XI: top team based on runs scored  © AFP
Enlarge

Ah! I fooled you, didn't I. You must have thought that I have sent an article meant for Fortune or Forbes magazine by mistake to Cricinfo. First let us put that matter to rest. The richest team must be the Indian team which collectively must be earning more than the rest of the teams together. Not that I care two cents about what the players earn. By "richest team" I mean, cricketing riches, in other words, the sum total of matches, runs and wickets which the team members take on to the field. It is also a lighter piece, coming in the wake of some serious and heavy analysis which have been done by me recently.

When a match is being telecast, the broadcasters talk about the experience behind a team in terms of matches played. However the real measures in this aspect are the number of runs scored and wickets captured by the concerned players.

I had done a sub-set of this article for another blog. I used the career figures. As I finished the article I realized that the career-to-date figures are the more appropriate figures to be used. So I applied the career-to-date figures, expanded the scope to matches also and have come out with a more comprehensive article here.

Since it usually happens that these analyses develop further based on user comments, I have done a simple accumulation of career-to-date figures of all 11 players. It certainly will give us some insight into the richness of teams in terms of the aggregates. I could look at the following options in later articles. I would also like the readers to come in with their own takes on these numbers and how these could be interpreted.

- Adding only the runs of the first 7 batsmen
- Add only for career-to-date runs greater than a certain value
- Add the wickets for the best 5 bowlers
- Do a comparison of the two team aggregates and analyse the most matched or ill-matched pairings.

I am aware that these are quantitative measures and not performance oriented. However there is no substitute for experience as the Indian batsmen showed at Dambulla against a New Zealand attack, which can at best be termed good and effective. It must be understood that the aggregates normally keep on increasing for a team but take a dip when senior player(s) retire.

The other thing I have done is not to show the top-10 accumulations. That would be quite silly on my part since the same team is likely to occupy the top-10 positions in terms of aggregates. Instead I have taken the top 8 teams + ICC XI and found the best for each of these teams and then ordered the tables accordingly. The other key feature is that I have taken the first innings figures. In other words, the test-beginning values. The second innings values would be higher, but this is a very minor matter.

Let us look at the tables. I have also provided the relevant details for the players who played in the particular innings for the top 3 teams. The support files provide the complete data.

Top teams based on number of Tests played

2008 1887 India           861 vs Aus
Gambhir G            18
Sehwag V             61
Dravid R            126
Tendulkar S.R       151
Laxman V.V.S         97
Ganguly S.C         110
Dhoni M.S            30
Harbhajan Singh      70
Zaheer Khan          57
Kumble A            131
Sharma I             10
2005 1768 ICC XI          818 vs Aus
Smith G.C            40
Sehwag V             37
Dravid R             92
Lara B.C            118
Kallis J.H           94
Inzamam-ul-Haq      102
Flintoff A           53
Boucher M.V          85
Vettori D.L          65
Harmison S.J         36
Muralitharan M       96
2006 1819 Australia       811 vs Eng
Langer J.L          102
Hayden M.L           86
Ponting R.T         107
Martyn D.R           67
Hussey M.E.K         13
Clarke M.J           24
Gilchrist A.C        87
Warne S.K           142
Lee B                56
Clark S.R             6
McGrath G.D         121

2008 1860 South Africa 707 vs Win 1991 1170 West Indies 701 vs Aus 2007 1845 Sri Lanka 614 vs Aus 2001 1532 England 542 vs Slk 1999 1443 Pakistan 504 vs Ind 2006 1822 New Zealand 415 vs Slk


All 11 players are shown.

The 2008 Indian team against Australia carried on to the field the collective tally of 861 matches. Kumble and Ganguly, with their 100+ Tests were the main contributors for this huge totals. Both were coming to the end of their careers.

Not so surprisingly the ICC XI comes in next with 818 collective Test appearances. Let me mention at this point that the ICC-Aus Test is official as determined by ICC. At no time would I ignore the match, as a few readers have suggested earlier. Flintoff's wickets were against a quality Australian batting line-up and Hayden's runs were scored against a quality ICC bowling attack.

Australia's 2006 team comes in next, just behind the ICC aggregate.

To download the complete all-time list, please right-click here and save the file.

Top teams based on number of career-to-date runs scored

2005 1768 ICC XI        49141 vs Aus
Smith G.C            3441
Sehwag V             3181
Dravid R             7871
Lara B.C            10818
Kallis J.H           7337
Inzamam-ul-Haq       7620
Flintoff A           2641
Boucher M.V          3007
Vettori D.L          1855
Harmison S.J          347
Muralitharan M       1023
2010 1964 India         47232 vs Slk
Gambhir G            2798
Sehwag V             6691
Dravid R            11395
Tendulkar S.R       13447
Laxman V.V.S         7136
Yuvraj Singh         1582
Dhoni M.S            2428
Harbhajan Singh      1585
2006 1819 Australia     40682 vs Eng
Langer J.L           7575
Hayden M.L           7385
Ponting R.T          9044
Martyn D.R           4390
Hussey M.E.K         1225
Clarke M.J           1179
Gilchrist A.C        5124
Warne S.K            2975
Lee B                1076
McGrath G.D           639

2010 1962 South Africa 35736 vs Win 1991 1170 West Indies 35090 vs Aus 2010 1964 Sri Lanka 27925 vs Ind 2001 1532 England 27585 vs Slk 2007 1830 Pakistan 25303 vs Saf 2004 1721 New Zealand 20312 vs Aus


Only career-to-date runs exceeding 250 are shown.

The ICC XI tops in batting aggregate. No surprise considering that they had Lara, Kallis, Inzamam and Kallis. If Tendulkar had come in for one of the later three batsmen, the tally would have been still higher. The high run values of Vettori, Flintoff and Boucher have helped the ICC XI>

India's team for the first Test against Sri Lanka clocks in next. The problem has been the low run values for the last three batsmen, Ishant, Ojha and Mithun contributing 180 runs. Even this total drops off drastically in the next two Tests.

Some distance behind in third place is the Australian team of 2006, clocking in at 40682 runs. After this match slowly the top Australian batsmen started going off.

To download the complete all-time list, please right-click here and save the file.

Top teams based on number of career-to-date wickets captured

2006 1780 Australia      1574 vs Saf
Warne S.K           657
Lee B               188
MacGill S.C.G       178
McGrath G.D         539
2007 1851 Sri Lanka      1290 vs Eng
Jayasuriya S.T       97
Vaas WPUJC          320
Fernando C.R.D       80
Malinga L.S          85
Muralitharan M      704
2005 1768 ICC XI         1246 vs Aus
Kallis J.H          183
Flintoff A          143
Vettori D.L         207
Harmison S.J        138
Muralitharan M      563

2008 1887 India 1209 vs Aus 2008 1860 South Africa 1177 vs Win 2000 1497 Pakistan 985 vs Win 2000 1506 West Indies 972 vs Eng 1982 0920 England 847 vs Ind 1989 1116 New Zealand 684 vs Pak


Only career-to-date wicket values of above 10 are shown.

Another Australian team of 2006 has the next wicket aggregate, 1574 wickets. This team's attack was led by Warne and McGrath and well supported by MacGill and Lee.

With Muralitharan contributing over half, Vaas nearly a quarter, the Sri Lankan attack of 2007 is in second place with 1290 wickets, way, way behind the Australian attack, which looks like it will not be surpassed for a very long times.

The ICC XI attack, not necessarily the best one at that time, is in third position. I have forgotten the match. However since Kumble had 464 wickets at that stage, a replacement of Vettori by Kumble would have moved them way up.

To download the complete all-time list, please right-click here and save the file.

It can clearly be seen that the experience in terms of matches, runs and wickets, are the cornerstone for success. The top teams shown above have all been very successful, ignoring the hotch-potch ICC XI. I will later do a correlation between the experience factor and the results, especially when there is a significant dip in the numbers, as happened with Australia last year.

Sriram, my in-house Editor has correctly mentioned that I have left the article short and it does not have any analytical conclusions. He is perfectly right. There are two reasons. One is that I wanted to get the reader comments/suggestions and come with a meaningful concluding article. The other is that I have to say something important here.

A regular reader has stated that he will not be visiting this site and stated three reasons. The first is a coloured-glass narrow-minded view on my being India-centric which needs no further look in. However there are two other reasons mentioned which need a response and that too, to the readers, since it concerns them also.

He has said that he will not participate in discussions in this blog because

1. I am not a mathematician/statistician and "only" an IT person. and
2. I have not played cricket.

Both perfectly true. Impeccable statements of 100% veracity. So let me say something on these.

I am proud that I am an Engineer/IT Person/Analyst. Foremost, I love the game, in all its areas, the play, the players, the prose, the analysis and the debates. Without this omnipresent love for the game, it does not matter who one is, one cannot do anything which will carry a very high degree of conviction leading to acceptance. I may not understand Chi-squared methodology or Gini adjustment or the philosophy behind stochastic processes. But I know when a 153 is better than a 400 or when a 2-wicket haul is superior to a 5-wicket haul or under what match conditions would a 50 made at no.7 be far superior to a 150 at no.1 and so on.

When I look at Cricinfo's wonderful analytical brain-bank, what do I see, an MBA, an Engineer, an MS from Kansas and so on. There might certainly be a statistician/mathematician or two there also. However they would be there, not just for their academic qualifications, but for their love of cricket and the ability to weave excellent articles around the dull and dreary numbers.

Let me take the recent theoretical study undertaken by two academicians which is currently doing the rounds. I will not make a single comment on the merits or demerits of this analysis. I may do that at a later date or someone from Cricinfo might do that. I downloaded the article and spent an hour trying to understand the same. I am sorry, I failed. Cricketing statements are interspersed with obscure (for the common man) statistical statements. The numbers do not make sense immediately since these are derived based on complex statistical processes. In other words, this article, possibly great in its own sphere, is not meant for the common man, but for other academics. The reader is expected to accept the findings even if he/she does not understand the basis.

To download the above referenced article, please right-click here and save the file.

I do not work that way. Every one of my articles has to be read and understood by all the readers, none excepted. If they do not understand something, it is my duty to explain. If they point out an error, it is my duty to correct the same. If they suggest something better, it is my duty to incorporate the same or explain why I have not done so. That is the way I have worked for the past two years and will continue to do so. If my articles do not sound technical and complicated enough for the segments of the academic readers, not all, let me add, so be it. I am proud of what I do and more proud of my own rapport with the readers, despite the many arguments I have had with them. And if ever I deviate from these self-imposed principles, send an immediate electronic brickbat.

Also if a reader suggest something unusual, as Soundararajan from Stanford who has suggested the factor, h-index. I have studied the same and am amazed at the simplicity and effectiveness of the same. I am in touch with Soundar to work out how it can be done effectively. At the same time, another reader, Murali, suggested a Gaussian distribution analysis on the top bowler tables. Since I do not understand the methodology completely I have requested him to do the work himself and offered to publish the results. The bottom line is that this is not a scientific journal but a blog, open to all.

Now for the second shortcoming I have. That I have not played any cricket. Eminently true. My highest score is 18 not-out, coming in at no.11, in a school match, played with tennis balls. Although I must add that this score was out of 50 for 9, and we won the match. Coming in at 28 for 9, I closed my eyes, swished and swished, and was incredibly lucky. So I have played no cricket.

Does it make me ineligible to write on or analyse the game. Even though I cannot last 6 balls against Sehwag bowling blind-folded and left-handed, that does not prevent me from understanding the value of his sub-100 innings at Chennai or Dambulla. If this is correct, most of the writers and analysts would have to stop doing what they are doing. Pauline Kael or Roger Ebert or James Agee did not act in films to write on films. Nirmal Shekhar, arguably the best Indian tennis correspondent, probably has not played tennis and so on.

Looking at the other side, a number of past players make good commentators, far fewer players can write and only very few are good analysts. I have worked with quite a few past cricketers and barring two leg-spinners, one very successful and the other, not-so-successful, the others could not understand even the rudiments of cricket analysis.

I apologise if I have gone on. However it needed to be said.

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

© ESPN Sports Media Ltd.

Posted by vipin on (September 3, 2010, 19:06 GMT)

well, good article ananth...I m a sports fan in general n like you, have not played much of sports. i m a medical student btw, but i wanted to say dat wouldn't sum total of avg in batting n strike rates n avg in bowling be better than total career runs or wicket till date......

Posted by physical therapist on (August 22, 2010, 19:18 GMT)

Terrific work! This is the type of information that should be shared around the web. Shame on the search engines for not positioning this post higher!

Posted by VIvek S on (August 21, 2010, 16:37 GMT)

@CS comment Aug20th, We cannot fully agree that pure cricketing knowledge trumps the statistical analysis. I have a question here, why is Sachin called the best ever and unparalleled by most of the Indians?? Its mostly due to the tons of tons he is about to get. and partly due the mastery of every aspect of batting. There were people who played the game equally magnificiently as him, but his stats will never be matched. as per my knowledge and eyes, i feel Lara and Laxman and Martyn batted more stylishly than him, but i would look foolish when i compare their stats. One more example to highlight knowledge over stats, take the example of Strauss and KP. They have identicial numbers in tests, But our knowledge says England is all bout KP, but stats would reveal that England failed more when Strauss falied, rather than when KP failed..Now which one to beleive??

Posted by Abhi on (August 21, 2010, 5:10 GMT)

Ananth, Further to my point about number of matches,runs,wickets etc acting as a "filter". For eg. all players who have played a 100 Tests,10000 + runs,500 wickets would automatically qualify as Greats or very good players.

That is - for the current generation.

Similarly, for previous generations a similar cut off would act like a filter.

Most players who cross these "thresholds" would probably be the cream of the crop. So any team comprising these players will in all probability do very well. [[ Abhi This was raw unfiltered table. Certainly further attempts would be period-sensitive. Ananth: ]]

Posted by Ramesh Kumar on (August 20, 2010, 12:44 GMT)

Ananth,

I think we should treat this as the team analysis. So while Bond can take wickets, will he and his bowlers have the strike rate to get the opposition out less than their batsmen's contribution? We need to take 3 to 4 NZL teams and compare their career figures with real performance in a series. How many times they would have performed together? If Indian top 6 have not performed together(in line with theier career figures) and don't have the bowlers to take the 20 wickets quickly, we know the result. If we don't have many teams with steady top 6, then may be we can look at 3 top order batsmen. Atleast we can look at specific teams-like 80sWI, Recent Aus teams, India with fab 4/5, Pak with 2Ws. They had steady core team members. [[ Ramesh Whatever you say requires the use of performance measures rather than the aggregate runs/wickets measures mentioned in this article. As such it falls outside the scope of this article. However It is a very interesting and intriguing idea. How often have team functions (batting or bowling) out-performed or under-performed. The logical follow-up will be "by how much". Ananth: ]]

Posted by Ramesh Kumar on (August 20, 2010, 11:03 GMT)

Ananth,

Is it possible to get the figures where a team's performance reflected the potential of the team? Supposing we take a series involving India. How many times the top 6 averaged equal or more to their career average in a series? similarly the front line bowlers. Probably we can limit the study to the top 50/100 groups based on your experience charts. We can then see the returns from experience. An extension to this is to see the value in real terms of runs and wickets. Assuming a bench mark figure of 400 in 1st innings and 300 in completed 2nd innings(we can agree on a figure), do we get the returns from Frontline batsmen(top 6) and front line bowlers(top 4)in absolute terms of runs and wickets across a series? This may throw up strong/weak spots of the team-like current Pak team-strong in bowling, weak in batting, India with strong batting,poor in bowling etc. Also possibly explain 80s WI and last decade Aus teams. [[ Ramesh That is one of the things I intended to do. Although I have to be careful that a bowler like Bond has to be treated correctly. Low in number of tests but high in performance. I am also intrigued by the difference analysis. Ananth: ]]

Posted by Ajai Banerji on (August 20, 2010, 10:46 GMT)

Thanks for the nice little puzzle you included. A few minutes with Statsguru revealed your friends to be Kumble and L.Sivaramkrishnan. [[ Two bowlers who were poles apart in every way. Short and name length. Height. Type of leg spin. What they achieved in Tests. As a leg spinner Sivaramakrishnan was one of the best, comparable to Qadir. He could not do justice to his talents due to many reasons, within and without. It is a sad tale of great talent not utilized and going waste. Kumble was an example of a player, not necessarily at the top of the pole, talent-wise, who achieved phenomenal results for himself and his team through fitness, bloody-mindedness, aggressive attitude and a never-say-die spirit. Bit both had one thing in common. They knew the numbers and the implications.. Ananth: ]]

Posted by prathyusha on (August 20, 2010, 10:16 GMT)

your article is excellent

Posted by Abhi on (August 20, 2010, 5:43 GMT)

CS Well said! and absolutely brilliant!

Now- Tell us something we don't already know.

Posted by Puneet on (August 20, 2010, 4:50 GMT)

Ananth,

I am sorry if I sound rude in this comment, but your statistics and article do not make any sense. You can play around with numbers and could produce a different result using a little different criteria.

I am an engineer and very well know all statistical techniques. I have played cricket for Rajasthan under 13 team and great fan of cricket. [[ Puneet You are in this space by choice. No one has forced you to come here. You are welcome to leave and peruse other articles which satisfy your academic and cricketing requirements. Ananth: ]]

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