I came out with an article on Test Batsmen a few days back and this article received an unprecedented number of responses. There were a number of useful suggestions and I am considering incorporation of a number of these suggestions. Since this process is going to take some time and I also have a number of T20/IPL analysis to be done, I have come out, in the interim, with a number of useful batsmen tables. The purpose of these are to provide at one place very useful information for comparison. There are no assumptions or parameters in these tables and what is presented are only facts. I have made only some explanatory comments on these tables.
The weighted bowling quality measure, which has been shown as the last column in all the tables, has been derived by using the career-to-date values as suggested by many readers. There is no major difference. However there is no doubt that this is the correct method. The real difference is with the lesser bowlers. For the great bowlers with 300+ wickets in their career, the CtD averages do not show any significant variation once a certain stage of career has been crossed.
1. Top run scorers for 8 periods
Aus Ponting R.T 2000s 8864 @ 59.89 36.00
Saf Kallis J.H 2000s 8428 @ 58.94 37.69
Aus Hayden M.L 2000s 8365 @ 52.94 36.23
Ind Dravid R 2000s 8125 @ 53.45 37.62
Ind Tendulkar S.R 2000s 6932 @ 52.92 37.74
Slk Jayawardene D.P.M.D 2000s 6581 @ 56.25 38.81
Win Lara B.C 2000s 6380 @ 54.07 34.43
Saf Smith G.C 2000s 6343 @ 50.34 38.70
Win Chanderpaul S 2000s 6342 @ 53.29 35.04
Ind Laxman V.V.S 2000s 6115 @ 49.72 36.61
Eng Stewart A.J 1990s 6409 @ 40.82 35.22
Aus Waugh M.E 1990s 6371 @ 41.64 36.73
Aus Taylor M.A 1990s 6306 @ 40.95 36.75
Eng Atherton M.A 1990s 6217 @ 38.38 34.31
Aus Waugh S.R 1990s 6213 @ 53.10 36.17
Ind Tendulkar S.R 1990s 5626 @ 58.00 37.97
Win Lara B.C 1990s 5573 @ 51.60 36.12
Slk de Silva P.A 1990s 4448 @ 46.82 37.85
Aus Slater M.J 1990s 4425 @ 45.15 37.02
Aus Boon D.C 1990s 4303 @ 45.29 38.61
Aus Border A.R 1980s 7418 @ 54.54 34.54
Eng Gower D.I 1980s 6196 @ 42.44 33.86
Pak Javed Miandad 1980s 5642 @ 54.78 37.02
Win Richards I.V.A 1980s 5209 @ 49.61 35.04
Win Greenidge C.G 1980s 5151 @ 45.99 35.43
Ind Gavaskar S.M 1970s 5647 @ 55.91 37.72
Ind Viswanath G.R 1970s 4611 @ 46.11 36.56
Aus Chappell G.S 1970s 4357 @ 53.79 36.08
Win Kallicharran A.I 1970s 3917 @ 49.58 34.70
Win Fredericks R.C 1970s 3809 @ 45.89 33.59
Eng Barrington K.F 1960s 6397 @ 59.79 37.82
Eng Cowdrey M.C 1960s 4788 @ 48.86 36.99
Aus Lawry W.M 1960s 4717 @ 49.65 37.27
Win Sobers G.St.A 1960s 4563 @ 60.04 36.48
Eng Dexter E.R 1960s 4232 @ 50.38 36.91
Eng Hutton L 40s-50s 5626 @ 54.62 39.27
Eng Compton D.C.S 40s-50s 5339 @ 49.90 40.02
Aus Harvey R.N 40s-50s 5107 @ 52.65 38.51
Win EdeC Weekes 40s-50s 4455 @ 58.62 37.83
Eng May P.B.H 40s-50s 4182 @ 49.20 36.49
Eng Hammond W.R WW1-WW2 6883 @ 61.46 42.87
Aus Bradman D.G WW1-WW2 5093 @ 97.94 34.50
Eng Sutcliffe H WW1-WW2 4555 @ 60.73 40.35
Eng Hendren E.H WW1-WW2 3525 @ 47.64 41.52
Eng Hobbs J.B WW1-WW2 2945 @ 56.63 39.95
Aus Hill C Pre-WW1 3412 @ 39.22 34.27
Aus Trumper V.T Pre-WW1 3163 @ 39.05 33.66
Eng Hobbs J.B Pre-WW1 2465 @ 57.33 32.64
Aus Gregory S.E Pre-WW1 2282 @ 24.54 31.60
Aus Armstrong W.W Pre-WW1 2247 @ 35.67 32.49
Note the current decade. All the ten batsmen, barring Laxman who chips in with 49.72 have averaged above 50. During the 1990s, only the three great batsmen, Steve Waugh, Tendulkar and Lara exceed 50. During the 1980s, only Border and the incomparable Miandad exceeded 50. Gavaskar, Greg Chappell and Boycott exceeded 50 during the 70's. During the early periods, Sobers, Hammond and Sutcliffe exceeded 60. Note the poor bowling faced by Hammond, however. All these are in the Top-10.
2. Top 20 aggregate scorers during a calendar year
Pak Mohammad Yousuf 2006 1788 @ 99.33 35.80
Win Richards I.V.A 1976 1710 @ 90.00 32.31
Ind Gavaskar S.M 1979 1696 @ 62.81 32.93
Saf Smith G.C 2008 1656 @ 72.00 41.33
Aus Ponting R.T 2005 1540 @ 66.96 34.66
Aus Ponting R.T 2003 1503 @ 100.20 41.81
Eng Vaughan M.P 2002 1481 @ 61.71 33.95
Aus Langer J.L 2004 1481 @ 54.85 36.76
Ind Sehwag V 2008 1462 @ 56.23 36.16
Ind Viswanath G.R 1979 1420 @ 56.80 32.90
Ind Tendulkar S.R 2002 1392 @ 55.68 40.08
Aus Hayden M.L 2001 1391 @ 63.23 36.46
Aus Simpson R.B 1964 1381 @ 60.04 43.94
Aus Hayden M.L 2005 1380 @ 53.08 34.23
Eng Amiss D.L 1974 1379 @ 68.95 36.32
Ind Dravid R 2002 1357 @ 59.00 39.39
Aus Martyn D.R 2004 1353 @ 56.38 35.23
Win Lara B.C 2003 1344 @ 74.67 34.35
Eng Pietersen K.P 2006 1343 @ 53.72 32.81
Aus Ponting R.T 2006 1333 @ 88.87 33.00
...
...
Win Sobers G.St.A 1958 1193 @ 132.56 35.24
Note the fantastic aggregates of the under-rated Mohammd Yousuf, Richards, Gavaskar, Smith and Ponting (twice). Also the 100+ averages of Sobers and Ponting.
Eng Hobbs J.B Australia 3636 @ 54.27 36.95
Eng Gower D.I Australia 3269 @ 44.78 35.69
Eng Boycott G Australia 2945 @ 47.50 37.25
Win Lara B.C Australia 2856 @ 51.00 31.39
Eng Hammond W.R Australia 2852 @ 51.85 39.95
Ind Gavaskar S.M West Indies 2749 @ 65.45 39.59
Eng Boycott G West Indies 2205 @ 45.94 34.79
Eng Gooch G.A West Indies 2197 @ 44.84 28.55
Aus Waugh S.R West Indies 2192 @ 49.82 34.92
Saf Kallis J.H West Indies 2073 @ 74.04 39.81
Win Lloyd C.H India 2344 @ 58.60 37.37
Pak Javed Miandad India 2228 @ 67.52 39.45
Win Richards I.V.A India 1927 @ 50.71 37.42
Win Sobers G.St.A India 1920 @ 83.48 37.07
Aus Hayden M.L India 1888 @ 59.00 37.76
Aus Bradman D.G England 5028 @ 89.79 34.28
Aus Border A.R England 3548 @ 56.32 36.08
Win Sobers G.St.A England 3214 @ 60.64 34.52
Aus Waugh S.R England 3200 @ 58.18 38.13
Win Lara B.C England 2983 @ 62.15 39.14
Pak Inzamam-ul-Haq Sri Lanka 1559 @ 59.96 39.03
Ind Tendulkar S.R Sri Lanka 1408 @ 56.32 38.56
Ind Azharuddin M Sri Lanka 1215 @ 55.23 40.34
Nzl Fleming S.P Sri Lanka 1166 @ 58.30 39.14
Win Lara B.C Sri Lanka 1125 @ 86.54 33.76
Eng Compton D.C.S South Africa 2205 @ 53.78 43.98
Eng Hammond W.R South Africa 2188 @ 62.51 44.29
Aus Ponting R.T South Africa 2030 @ 56.39 31.34
Win Lara B.C South Africa 1715 @ 49.00 31.96
Aus Harvey R.N South Africa 1625 @ 81.25 43.52
Ind Gavaskar S.M Pakistan 2089 @ 56.46 35.01
Aus Border A.R Pakistan 1666 @ 59.50 34.36
Aus Chappell G.S Pakistan 1581 @ 63.24 37.65
Slk Jayasuriya S.T Pakistan 1490 @ 51.38 35.42
Slk de Silva P.A Pakistan 1475 @ 42.14 35.52
Pak Javed Miandad New Zealand 1919 @ 79.96 42.31
Aus Border A.R New Zealand 1500 @ 51.72 35.72
Ind Tendulkar S.R New Zealand 1406 @ 52.07 34.82
Saf Kallis J.H New Zealand 1356 @ 67.80 40.58
Ind Dravid R New Zealand 1318 @ 59.91 33.63
Slk Atapattu M.S Zimbabwe 1145 @ 95.42 48.01
Ind Dravid R Zimbabwe 979 @ 97.90 42.11
Ind Tendulkar S.R Zimbabwe 918 @ 76.50 42.62
Nzl Astle N.J Zimbabwe 813 @ 50.81 40.86
Pak Inzamam-ul-Haq Zimbabwe 772 @ 42.89 42.03
Slk Sangakkara K.C Bangladesh 876 @ 73.00 46.75
Slk Jayawardene D.P.M.D Bangladesh 863 @ 66.38 47.63
Slk Dilshan T.M Bangladesh 771 @ 77.10 43.68
Saf Smith G.C Bangladesh 743 @ 82.56 51.82
Slk Samaraweera T.T Bangladesh 600 @ 66.67 45.56
Sehwag and Taylor against Pakistan, Kallis against West Indies, Javed Miandad against New Zealand, Sobers and Zaheer Abbas against India, Lara against Sri Lanka, Harvey and Jayawardene against South Africa and of course Bradman against England are the stand-out performers.
Ind Tendulkar S.R 7165 @ 54.28 37.18
Ind Dravid R 6430 @ 56.90 37.50
Win Lara B.C 5736 @ 47.80 34.12
Aus Border A.R 5431 @ 56.57 36.32
Win Richards I.V.A 5404 @ 50.50 34.98
Pak Inzamam-ul-Haq 5226 @ 47.94 37.92
Aus Waugh S.R 5217 @ 55.50 36.68
Ind Gavaskar S.M 5055 @ 52.11 36.94
Saf Kallis J.H 4713 @ 52.37 36.46
Aus Ponting R.T 4680 @ 50.87 35.89
Win Lloyd C.H 4634 @ 46.81 34.32
Pak Javed Miandad 4351 @ 45.80 34.31
Win Greenidge C.G 4349 @ 42.22 35.22
Eng Hammond W.R 4245 @ 66.33 43.22
Nzl Fleming S.P 4225 @ 45.92 36.68
Eng Cowdrey M.C 4087 @ 44.91 35.90
Win Chanderpaul S 4082 @ 42.52 34.10
Ind Ganguly S.C 4032 @ 41.57 37.68
Aus Waugh M.E 4010 @ 40.51 37.17
Win Sobers G.St.A 3957 @ 50.73 33.67
...
...
Aus Bradman D.G 2674 @ 102.85 35.18
The list is headed by Tendulkar and followed by two great contemporary batsmen. Tendulkar has been equally effective home and away while Dravid has been better away. Lara has significantly under-performed away from home, but against very good bowling attacks. Hammond has a very high away average, tempered by the fact that a lot of these runs have been scored against weak bowling attackes. For the record, Bradman has the highest away batting average. Note Steve Waugh's excellent away average.
Eng Hobbs J.B Aus 3636 @ 54.27 36.95
Eng Hobbs J.B Saf 1562 @ 60.08 33.52
Eng Hobbs J.B Win 212 @ 106.00 53.90
Career Total 5410 @ 56.95 36.62
Aus Bradman D.G Eng 5028 @ 89.79 34.28
Aus Bradman D.G Ind 715 @ 178.75 45.15
Aus Bradman D.G Saf 806 @ 201.50 36.54
Aus Bradman D.G Win 447 @ 74.50 45.89
Career Total 6996 @ 99.94 36.39
Win Richards I.V.A Aus 2266 @ 44.43 33.79
Win Richards I.V.A Eng 2869 @ 62.37 35.40
Win Richards I.V.A Ind 1927 @ 50.71 37.42
Win Richards I.V.A Nzl 387 @ 43.00 30.80
Win Richards I.V.A Pak 1091 @ 41.96 32.44
Career Total 8540 @ 50.24 34.84
Ind Tendulkar S.R Aus 2748 @ 56.08 34.45
Ind Tendulkar S.R Bng 556 @ 139.00 44.19
Ind Tendulkar S.R Eng 2150 @ 61.43 40.07
Ind Tendulkar S.R Nzl 1406 @ 52.07 34.82
Ind Tendulkar S.R Pak 1057 @ 42.28 36.08
Ind Tendulkar S.R Saf 1202 @ 35.35 35.16
Ind Tendulkar S.R Slk 1408 @ 56.32 38.56
Ind Tendulkar S.R Win 1328 @ 57.74 41.18
Ind Tendulkar S.R Zim 918 @ 76.50 42.62
Career Total 12773 @ 54.59 37.80
Win Lara B.C Aus 2856 @ 51.00 31.39
Win Lara B.C Bng 173 @ 86.50 45.84
Win Lara B.C Eng 2983 @ 62.15 39.14
Win Lara B.C Ind 1002 @ 34.55 35.36
Win Lara B.C Nzl 704 @ 41.41 39.19
Win Lara B.C Pak 1173 @ 53.32 34.95
Win Lara B.C Saf 1715 @ 49.00 31.96
Win Lara B.C Slk 1125 @ 86.54 33.76
Win Lara B.C Zim 222 @ 55.50 44.09
Career Total 11953 @ 52.89 35.22
Hobbs and Bradman have scored most of their runs against the good attacks of Australia/England and South Africa. Almost all of Richards' runs have come against good attacks. The major difference between Tendulkar and Lara is that Tendulkar has scored nearly 1500 runs against the weak attacks of Bangladesh and Zimbabwe. Lara has scored only 400 runs against these attacks. This explains the overall poorer bowling quality faced by Tendulkar.
Note how Richards finished his career. The last 5 years have been well below the high standards he had set. The other batsmen have finished well, Tendulkar playing very well and counting. He is going to comfortably maintain the career average of around 55+ to the end.
So that the readers can download these six tables and study at leisure I have uploaded the zip file.. Please click here to download the zip file.
Finally I have given below an idea of the changes expected to be done when I re-do the Top Test Batsmen analysis.
1. Use CtD bowling values.
2. Use new Wted Bow Index, using Average and Strike rate, based on Arjun Hemnani's suggestion.
3. Remove Scoring Rate measure.
4. Reduce % Team share weight to 5%.
5. Work out Consistency index for 5% based on 10-Test slices for each batsman.
6. Improve the pitch Index calculations.
7. Determine match performance figures based on number of innings played.
8. Prepare two independent sets of tables: Pre-1940 and Post-1945.
9. Finally R-Factor (you have to wait for the article to know about this).
As a certain Vulcan with pointy ears would say 'Fascinating' analysis. I was quite amazed at Sachin's away record 54 avg but as a scrolled down i saw that he played more away games against Zim and Bang than either Lara or Ponting. In fact Tendulkar scored 796 away runs @ 79 away to Zim and Bang. Didnt hurt his away avg did it? All perfectly 'Logical'!
Posted by: Ashik Uzzaman at May 22, 2009 11:22 PM
Interesting follow up of your earlier article. I wonder if there is any easy way to know how much Tendulkar scored against Australia. CricInfo Recrods/Statistics should be able to help me on that, right? Ananth, please keep adding more tables like this if you have to spend more time before you come up with the second part/revised version of Great Test Batsmen. [[
Ashik
Tendulkar has done quite well against Australia and this information is there in the article.
Ananth:
]]
Posted by: Marcus at May 23, 2009 12:58 AM
It's interesting to note that Gavaskar seemingly rarely faced the West Indies at their best, judging by the Ctd measure of 39.59. I saw something to that effect years ago in one of the "Numbers Game" columns. [[
Marcus,
It is true. Just look at the 1971 West indian bowlers, average at best. But what must be appreciated that he faced a lot of these bowling with virtually no support.
Ananth:
]]
Posted by: dguru at May 23, 2009 1:19 AM
Ananth,
Not sure if this comment is most appropriate for this or the previous post.
Would it be possible for you to just use these and similar statistical facts and create a sort of 'mean field' from which you can the evaluate one player at a time relative to the field value? Does that make any sense to you? That my be a first order approximation of sorts to a version of greatest viz, player with the strongest ensemble stats.
Posted by: graeme at May 23, 2009 1:35 AM
How about a country graduation mark??? Its subjective in one sense but black and white in another. Until a team win an overseas series any statistics made against that country before that series win are not included, that will weed out a lot of the weak attacks that pump up averages, so NZs first few decades and Zim and Bangladesh basically don't make the cut, as well the begining years of the other test nations. I think you will see some big differences. [[
Greame,
To a great extent the Weighted Bowling quality measure takes care of that. The early Bangladeshi teams have this index hovering at around the 70 mark
Ananth:
]]
Posted by: Raman at May 23, 2009 1:58 AM
Ananth, excellent analysis. Gooch's figures are astonishing. Just confirms that he faced some of the best bowling of all times and still scored tons of runs. Gavaskar in 79, Richards in 76 are remarkable. Tendulkar has been very good against the Australians (we all know that) and good bowling attacks of NZL(??) [[
Gooch's 154 against West Indies, out of 252, during 1991, against a truly frightening attack (Ambrose/Patterson/Walsh/Marshall), is possibly the best innings played by any opener.
Ananth:
]]
Posted by: Neeraj at May 23, 2009 3:43 AM
@Ananth
It is commendable that you have incorporated reader suggestions. Most of these, however, are not real fundamental changes but “nitty grittys” which will impact the previous list minimally. But if you scroll through the comments of your previous post it is evident that the pet peeve (by a huge no. of people) is a lop sided weightage given to match performances. This factor brings in a host of variables including performances of other team players and so dilutes the analysis when comparing individual batsmen to each other directly. [[
Neeraj
The weight given to match performances is 50% as against career achievements. No one has questioned this equal allocation. Since the match performances have not benefited SRT, one's favourite batsman, they are questioned with no rational basis. If, by chance, these figures had been better for SRT, all these complaints would disappear.
The changes are not cosmetic and will have a great impact. I would appreciate your looking at these again.
Ananth:
]]
@eddy
Firstly, Tendulkar had many more N.O than Lara. So, if you factor them in he actually gets more runs in less innings. Several statisticians, inc. Charles Davis are of the view that if a batsman was allowed to complete N.O inn. he would obviously not only have more runs but ALSO a higher avg.
Also, considering that Tendulkar did better against Aus.than Lara (both home and away .and also incidentally in ODIs while at it), it would have been ideal he actually had a choice of playing Aus. instead of Zim.etc.- That would have only increased his Legend. Personally, I wish he would have and that the BCCI had arranged the India schedule keeping in mind future “best batsman” lists. Since ppl like you obviously cannot recognise genius even if it stares you in the face. [[
Eddy, and others, could make a similar statement to your last one, to you re Lara. It is unfortunate that people think that they shuld push up their favourite batsman by pulling down the other. Why would you not appreciate the fact that they are two great batsmen, peers without parallel. One comes ahead of the other in some lists, they would switch places in another. Before you rush to make another set of statements, think of the very high mutual regard and respect these two great batsmen have for each other.
Ananth:
]]
Posted by: Narinder Sharma at May 23, 2009 3:50 AM
Again when the writer writes very clearly that Lara has faced more formidable attack than Lara it just goes on to show his bias again. I can never understand why he is so biased and if he is so then why compare???just go ahead and say that i like such and such player so he is the greatest of all times. Pakistan attack headed by Gul can never be more formidable than playing against Zim in Zim facing their seam bowlers under tough conditions. [[
Narender
Does any one force you to read such a biased, off-the-cuff, non-objective, (as some one said) non-patriotic article.
Ananth:
]]
Posted by: GregB at May 23, 2009 4:41 AM
It still amazes me when you read Bradman's stats. I am no doubt bias being an Aussie, but to see him score his highest tally in 1948 (his last year) at a truly remarkable average is astonishing. Is it possible to predict how many runs Bradman, Hammond and Hobbs would have scored if he played some of the weaker and novice nations, and played as many tests as they do today? I really enjoy reading these tables.
Posted by: Prashant at May 23, 2009 4:47 AM
Eddy,
Even if you completely take away (never mind factor in bowling quality etc.but completely take away (which of course is incorrect) Zim/Bang away runs.) you have:
Tendulkar: 6369 @ 52.2
Lara : 5515 @ 47.5
STILL no contest.
Which, of course, is perfectly “Logical”. Since we know that Tendulkar is the overall superior batsman. [[
Prashant
If you take away Zimbabwe/Bangladesh runs, the following are the career figures.
SRT: 11299/218=51.83
BCL: 11558/220=52.54
If I agree with you that "Tendulkar is the overall superior batsman" will such mails stop. I will stipulate that.
Ananth:
]]
Posted by: Sandeep Majumdar at May 23, 2009 5:23 AM
Hi Ananth,
Excellent Analysis.
As an extension to your above analysis on Great Test Batsmen.
Can you also do a research based on the relative ranking of a Batsman's scores in his team. That is; whether he was the highest scorer, 2nd highest and so on.
Based on the above; you can have the results like who was the batsman with the greatest percentage of being the Top scorer in his team OR
Which Batsman had the maximum percentage of their careers being in the Top 3 Scores in their team.
This analysis can be extended to include both the teams playing. That is among the 2 teams playing which batsman consistently top scored from either side, etc.
I am sure; even these analysis will point to the same geniuses like Lara, The Great Sir Don, Sachin, Gavaskar, Ponting, etc.
Posted by: pam at May 23, 2009 6:37 AM
"The major difference between Tendulkar and Lara is that Tendulkar has scored nearly 1500 runs against the weak attacks of Bangladesh and Zimbabwe. Lara has scored only 400 runs against these attacks. This explains the overall poorer bowling quality faced by Tendulkar."
- How the above prove that Tendulkar faced poor bowling quality, when it seems Lara wasn't good against those attacks? [[
Zimbabwe and Bangladesh have poor bowling attacks is an obvious factt. When Tendulkar has scored 10% of his runs against these weak attacks his overall quality of bowling faced goes down slightly.
Why do you say that Lara wasn't good against these attacks when he has scored 400 in very few innings.
Ananth:
]]
Posted by: pam at May 23, 2009 6:50 AM
"Tendulkar has been equally effective home and away while Dravid has been better away. Lara has significantly under-performed away from home, but against very good bowling attacks."
- Does that mean Tendulkar and Dravid faced very poor bowling?
How do you determine weak bowling attacks?
I would like have pitches thrown into the calculation( i know it would be hard ). How do the batsmen fare in unfamiliar pitches like Indian in fast/bouncy pitches, aussies in swinging/turning pitches ignoring the flat beds!
Posted by: Prashant at May 23, 2009 6:59 AM
Ananth,
1) Firstly , I was referring to AWAY runs.
Prashanth:
Sorry I overlooked that, but anyway the career figures matter more.
Ananth
2) You seem to be clearly statistically wrong!! what I get is:
Tendulkar 11299/51.8
Lara 11588/51.6.
Even if you completely take away some test runs (which is ridiculous). I suggest that eddy then simply “takes away” the dead runs of just 2 inns.375 and 400* and checks the runs/avgs.
So one wonders where you are getting your figures from. I’m using cricinfo. So, it seems your entire analysis is based on some faulty figures!
Prashanth:
The following figures are exactly as per Cricinfo AND my Database.
SRT 261 27 12773 54.58
BCL 232 6 11953 52.88
Now do tthe following calculations. All numbers are picked up from the tables.
SRT: 12773-556-918=11299, OK.
234-4-12 = 218.
11299/218=51.83.
BCL: 11953-173-222=11558. OK
226-2-4=220.
11558/220=52.53.
I have gone to this extent since you have questioned my figures. Your other points are deleted since these have no relevance to the current article.
Ananth
Posted by: John Clark at May 23, 2009 7:49 AM
Ananth - thanks for the details - interesting. I am inclined to think that the quality of the pitch needs to be accounted for similtaneously with the quality of the opposition. For example, a hundred against the best bowling attack on an easy wicket approximately equivalent to a hundred against a weak attack on a difficult wicket.
Weird how some people get so attached to the results of Tendulkar vs Lara. Jack Hobbs has a significantly better record than either, particularly before WW1.
Regards.
Posted by: knucks at May 23, 2009 7:57 AM
i agree that lara has been with a much weaker team, hes practically been holding the team for 5-7 years by himself.
all this lara vs tendulkar debate is silly. why isnt there ponting? he ahs the higher average of the 3?
its because he is with the australian team, however being just a batsman is not as hard as being a batsman/captain, which both lara and tendulkar experienced.
besides the indian contingent striving to make this a "tendulkar is the best batsman ever" analysis it is completely unbiased and enjoyed reading it. i love watching every batsman, thats what being a true cricketing fan is all about.
Posted by: Pankaj at May 23, 2009 8:44 AM
Ananth.
I too am a “Tendulkar fan”. But, this “match performance” criteria lop sidedness is not just restricted to Tendulkar vis a vis Lara, but to all the other batsmen on the list except for Bradman. With the current methodology it is almost impossible for anyone to “catch up”to Lara. All else staying the same i.e match performances, bowling qual., pitches ,runs,longevity etc etc ,simple calculations would reveal that both Tendulkar and Ponting would have to currently average about 81 and Richards would have had to finish with a career average of 101, in order to even catch up with Lara on the total “points”. How do you explain an anomaly like that?
Also,Prashant was clearly referring to “away” runs. Even cutting out bang and zim , Tendulkar has done much better away.
Posted by: David at May 23, 2009 10:01 AM
I was interested to see the CtD bowling figures for Bradman, especially since I was almost starting to believe those on your other post who were insisting that if he had batted today, he would have been average at best, and that SRT is therefore the greatest batsman of all time. According to the CtD figures, the bowling he faced was more difficult than SRT and only a little easier than Lara and Richards.
Posted by: eddy at May 23, 2009 10:40 AM
Dear Ananth, you may or may not post this mail but
if this is just between me and you let me say this. I was the 10th published responder to the original 'Greatest test bastman' blog. I said i agreed with 99% of your findings and i wrote 'Having said all of this it is clear that Tendulkar is the greatest overall(odi& test)'. That is a direct quote. As i have said in another more recent post i tend to get slaughtered when i state the case for Lara being the greater 'Test' player. It know appears that even you have tired of the Tendulkar lovers barrage as you have humourously agreed to agreed with 'Prashant'.
Anyway the whole exercise wasnt about Tendulkar V Lara! Maybe that's a job for another day?
Best wishes Eddy
Posted by: knight at May 23, 2009 10:53 AM
Tendulkar plays for not outs. Lara makes more runs per test match. And in cricket most of the time more runs you scored is better for the team.
Tendulkar may have scored runs against Australia but comprared to Lara the Australian Attack he faced have been comparitively weaker. Take a case of 1998 when Australia tourned India, when Australian side only had one good bowler in Shane warne. It would also be interesting to know what's the record of both when Glenn McGrath was playing.
Posted by: Nick at May 23, 2009 12:09 PM
Interesting that i always here that as Ponting has a shocking record against India (what is his overall avg against them anyway?) that he is below the class of Tendulkar...but lokk at Tendulkar's avg versus SA? 35! way below Pontings versus SA (56). I agree that Tendulkar is the better batsman but statistics show that every great batsman has at least one team he struggles against compared to the other teams. COMPARED to the other teams...not compared to other batsman.
Posted by: Zain at May 23, 2009 12:37 PM
Anath,
great addition to some continually great analyses. It's a shame you still get so much hassle for presenting dispassionate, statistical overviews!
Please keep up the good work.
Posted by: Anand at May 23, 2009 2:21 PM
Ananth: The split up presented here may probably put a stop to people making comments like "How is xyz agead of Richards, How come so and so is not there in the top 10" etc. Great analysis as always from you. I am also very much disappointed that people are still out to prove who is better than who? I believe we are all fortunate to see artists like Sachin, Lara, Ponting, Dravid, Sangakara, Jayasurya all in the same era. Same goes with the likes of Gavaskar, Richards, Gooch, Gower and co. A simple fact of the matter is that every era produces some legends and as cricket lovers, we must just sit back and enjoy watching them. Infact as you have shown with your many analyses, changing the criteria gives different lists with different rankings for the same players-just going on to show that each one is special in a different way. I really hope this kind of x is better than y posts reduce if not stop completely. Your analyses are surely enabling things in that direction. Keep them coming.
Posted by: Anand at May 23, 2009 2:33 PM
I am just wondering about comments like "Imagine Bradman today with helmets and againt the likes of Zimbabwe and Bangladesh". One can extrapolate based on available data but coming to think of it, probably the lack of challenge against these teams might prevent such legends from giving their best (I am not predicting this, just guessing so). Take Lara for example, he has scores of 277, 212, the epic 153* against McGrath, Shane Warne and co but no such score against India. Surely it does not suggest that Srinath, Prasad and Kumble form a more formidable attack!! On the contrary compare the average of Chanderpaul against the two teams!! I agree with Nick's observation. Batsmen may have jinx against a particular team. Based on VVS Laxman's average against Aus he must have more than 10000 runs by now (if one were to just extrapolate against other teams) but doesnt quite work out that way does it? Bradman for me would be THE BEST TEST BATSMAN irrespective of any extrapolations.
Posted by: Voltaire at May 23, 2009 8:02 PM
Ananth:Exhaustive analysis...only buttresses the fact that even the most sophisticated analysis cannot convince all. I am a little miffed that Sunny comes as a lesser batsmen in most of your analyses which is patently untrue. And the claim that he didn't face the Windies at their best is also false....barring 1971 series the quality of fast bowling he faced is the best ever in history. 74-75-in India-only played one test i think. 75-76 in Windies...Roberts,Holding in full cry-2 centuries and plenty of runs. Discount 78-79 series.82-83 in Windies against Roberts,Holding,Garner,DevastatingMarshall..failed but for a century.83 home series against almost unplayable Macko,Holding,Daniel,Davis..scored plenty. His century in Delhi came off 94 balls and even more his 90 at Ahmd on a treacherous pitch has to be the best ever 90. Then that 236 against Chennai! BTW.he also continually faced and did fantastically well against other fast bowlers in their acme-Imran/Thommo/Botham-79-80/Willis.
Posted by: Chris at May 23, 2009 8:06 PM
An excellent appendum to what is a magisterial analysis. I can only commend your rigour and apparent indefatigabilty, and disdain for the tedious Tendulkar/Lara debate. I do wonder whether the analysis might be benefitted by inclusion of the averages of the top 6, or the win/draw/lose ratio, of respective batsmen's teams? I believe this might go some way to answering questions of character and other such qualitative factors within a statistical spectrum, and enrich the quality of results greatly.
Posted by: James at May 23, 2009 10:38 PM
Just on SRT's record against Aus: SRT has had the advantage of playing in some series against Aus where neither McGrath or Warne (or both) were present. Of his 29 Test v Aus only 8 of those featured both McGrath and Warne. 5 had Warne only and 1 featured McGrath only. So he had 15 of 29 Tests against Australia with neither bowler (17 really since Warne in 1991/92 was a shadow of his future self). Check out Sachin's averages in each category of game:
Against just McGrath (1 Test) - avg = 5.00, Against just Warne (5 Tests) - avg = 102.83, Against neither bowler (15 Tests) - avg = 58.76
Against both bowlers (8 Tests) - avg = 40.75
Warne never particularly troubled Tendulkar too much but McGrath got him quite a few times. They only faced each other in 9 Tests and Sachin averaged only 36.78 in those Tests - 6 of which were in India. McGrath got him 6 times in those 9 Tests.
Sachin undoubtedly is still great, but he faced a weakened Aussie attack more often than is realised.
Posted by: Mohanlal at May 23, 2009 10:46 PM
Ananth sir,i cannot but criticize your complete neglect of notouts.It must be noted that not all notouts occur intentionally.Also it is not Chrismartins' about whose notouts are fully ignored.it would be worth thinking as to how some of these notout scores would have evolved in batsmen's cases like SRT's.take his 4rth inns 119* vs ENG.He was not 2nd down in those times.Had there a session left young nothing to loose SRT would have scored atleast 31 of those rem: 57 runs there by scoring a 150 in winning 4rth inns cause abroad.Also take 148* vs Aus.Had he got some support he would have scored atleast 200 and with Merv getting out soon that would have resulted in a famous win too.So is his 155* at Chennai vs Aus.In that mood had he continued some more time would easily have crossed 200.like wise some more of his not out scores.my point is that notouts infact need to given their worths too.They cannot be fully neglected. [[
Mohanlal
Where did you get the impression that I have not included Not outs. Tendulkar's 12773 runs include the 155* amongst other not out innings. His average is 54.58 including all these Not out innings. Nowhere have I mentioned that Not out innings are excluded. This bogey has been raised by a few people. If I have excluded Not outs then Lara would lose his 400 innings.
What you say is true if I have any calculation involving only dismissed innings. I can assure you, NO.
Ananth:
]]
Posted by: Salim Mian at May 23, 2009 11:06 PM
Apart from the soon to be worked on and
published Bowlers version of this analysis (i hope) how about
'The greatest match winner' analysis. Starting with the batsmen,
you could explore and find out what actually match winning
(for a single player) would entail.
Hundreds in both innings? Big hundred in first inning?,
fourth inning tons, Speed of scoring(rapid or steady building).
Of course a match winning knock insnt always a 100, but often is!
Off the top of my head i guess Bradman would figure highly but im sure
Lara, Tendulkar, Richards, Ponting, Sobers etc would all figure.
For instance the recent test match between End and WI at Durham.
Jimmy Anderson got the man-of-the-match with 9 wickets in the match,
but would Cook's 8 hour 160 be classed as a 'match winning knock'?
What's 'setting up the win' and 'match winning'?
The exercise would be much more than just seeing who was man-of-the-match.
Goochies 154* and Lara's 153* are obvious ones but it would be intriguing
to find out who really was the greatest match winner. Wouldnt this also
tell us who the best batsmen really was because cricket should be about
winning matches for your team and not just about personal records.
Just for the record Only 16 of Tendulkars 42 tons have come in an
Indian win (thats not saying he hasnt won matches scoring less than 100
or that his ton won the match!).
Two examples; tendulkar's 109 in the 4th test V Aus in 2008 was supported
by sehwag 66,Laxman 64, ganguly 85,Dhoni 56. In the second inning when
Tendulkar, dravid, ganguly, laxman all scored less than 13,
sehwag made 92! surely more match winning?
second example; in november 2007 tendulkar top-scored
in the fourth inning against Pak in a low scoring match with 56*
which took him 3 hours. match winning?
lara's record appears worse, 8 out of 34, but again thats not saying
he hasnt won matches scoring less than 100
or that his ton won the match!
23 of Bradmans 29 ton's came with Aus victories!!!!!
I think ive found the answer already, Bradman the greatest match winner and
greatest player ever.
But no2? no3? no4? etc......
regards Salim
Posted by: S.m at May 24, 2009 3:13 AM
Ananth, pankaj
Of all the many comments on both your posts “pankaj” makes the MOST astonishing one!!
Please confirm the veracity of his statement.
This would mean that even if Richards had displayed some superhuman batting ability and technique by managing to remain Not Out a staggering 96 TIMES out of his 182 innings (and so averaging the same as the Don) he would STILL not be “as good” a batsman as Lara?!
It is almost like saying that if Richards had batted TWICE as well as he actually did, Lara is still the “better” batsman. This is what your table/points seem to be implying.
Needless to say this is scarcely believable [[
S.m
That is Pankaj's comment. I don't have anything to do with that. That is to take a single measure and twist it completely to come to an outrageous conclusion. Tendulkar can catch up with Lara through many methods, score more runs, improve his record against South Africa, improve his average and play many match-winning innings. Why consider one factor, the average, and distort it.
Ananth:
]]
Posted by: SuryaP at May 24, 2009 3:33 AM
between Sachin and lara it is much difficult to tell that who is the best batsman. My observations are like this
1. We can't decide which team has got weak bowling attack, as one of you said Ban and Zim hasweak bowling attacks then why BCL couldn't score more runs against these teams. In fact India struggled somany years to win a test match and series in Zim.
2. Lara's 400 didn't help team to win the match. He just wanted to score more runs and regain his supremcy against a mediocre bowling attack ( of course pitch and weather made the bowling attack look mediocre )
and many more comaprision soon
Posted by: Robert at May 24, 2009 4:00 AM
I think something that can't ever really be measured is the quality of the pitch. It's no secret that the pitches in the sub-continent are more batsmen friendly so generally tend to offer more runs. It then becomes quite interesting when you note that someone like Dravid has actually performed better away. Going back through the years where lower scores were more common place (weaker batting line ups, better bowlers or pitches that fell apart?) it is even more amazing that Bradman averaged so well.
To sit and argue that one batsmen is so much better than the rest is fairly relative as someone like Hayden never faced the formidable Aussie attack... then again spare a thought for someone like Andy Flower who was a one man show against the best bowling attacks that Aus, SA and even SL have had to offer.
Posted by: Gul Khan at May 24, 2009 6:22 AM
Ananth : Thanks for the stats. Its a shame that whenever you do one of these tables, it always seems to become a tendulkar v lara debate instigated by a very pro-Tendulkar contingent. Why people just cant see that they are/were both great batsmen and we were priviliged to see such talents at the same time i do not know. Who would i rather watch, then its lara, not because i think he may or not be a better batsman, just i think he is more entertaining. Over the years who out of these two would i prefer to bat for my life, then probably tendulkar, however, when answering that question, i still have go back to lara against the australians and how he single handedly beat them at times. Lara also held together a weak team, where tendulkar made a good batting line-up into a great line-up. Despite both their attributes and achievemnets, they both pale into insignificance when compared to The Don. It is still astonishing what he achieved on bigger grounds, uncovered pitches and a lighter bat. [[
Gul
It is not often that I get a comment like this. You have reflected exactly what any discerning cricket follower feels. You have also not put down one batsman as many have done. You have also been gracious also not to bring any country feelings.
I have told many times that both are great batsmen and we should feel privileged to be contemporary to these two playing careers. Because I like Lara more, does it mean I cannot appreciate the grace, in public and private, the poise, the batting excellence non-pareil of Tendulkar. If I love the Bridgetown 153*, does it mean I cannot appreciate the almost equally great, unfortunately unsuccessful, 136 at Chennai.
It is unfortunate that the ultra-chauvinistic feelings have clouded the whole issue and are preventing people from commenting on the intrinsic faults/benefits of such a methodology. This forum, which should be an ideas-exchange forum has become an acrimonious mud-slinging one.
Your mail is an eye-opener and I hope people read and appreciate the intrinsic merit of the comments.
Once again many thanks.
Ananth:
]]
Posted by: Sathya at May 24, 2009 7:16 AM
[i]As a certain Vulcan with pointy ears would say 'Fascinating' analysis. I was quite amazed at Sachin's away record 54 avg but as a scrolled down i saw that he played more away games against Zim and Bang than either Lara or Ponting. In fact Tendulkar scored 796 away runs @ 79 away to Zim and Bang. Didnt hurt his away avg did it? All perfectly 'Logical'![/i]
Even if you remove Bang and Zim from Sachin's innings and runs part he averages 52 and Lara with avg of 86 against Bang is just averaging 47 away, Hope this is "logical" to you and nice collection of stats by the way,Don as we all know averages 99 but during those time, Eng was the strongest side, against that team he has an average of 89.79 which is really great.
Posted by: Mohanlal at May 24, 2009 11:09 AM
Ananth sir,this is in response to your reply to my previous post.What I was refering to was your reply to Prasanth's mail.While calculting avg:s of SRT & BCL excluding Zim & Ban you have completely excluded notouts.Then SRT's avg: became 51.83 & BCL's 52.53.How can it be justified to ignore all not outs in this particular case? [[
Mohanlal
You are again mistaken. I had only indicated the innings used to determine average. In other words,
Avge=Runs/(Inns-Notouts).
I had not shown the Inns and Notouts values separately but showed the (Inns-Notouts) value. You yourself can work that out.
Once again I will repeat that neither the Notout innings nor the runs scored have been excluded.
Ananth:
]]
Posted by: Ananth at May 25, 2009 2:35 AM
The following is an extract from a reader comment.
""
1)I’ve said often enough that if you simply take away a couple of Lara’s inn. i.e the dead runs of the 375 and 400 (similar to the way Ananth has simply taken away Tendulkars runs) Lara ends up with a test avg. of below 50.
""
I have never seen anything more ridiculous than this allegation. I have deliberately not published his comment directly since that means I have given validity to his outrageous comments and would have givem him publicity which is what being sought.
However this is to confirm that all Tendulkar's runs are included in the analysis, WITHOUT EXCEPTION, in every calculation. That is all. No further explanation need to be given.
If people do not like the article or cannot understand it, I suggest they turn to other columns. Surely Tendulkar does not need supporters like this reader and others like him. He is one of the all-time greats and does not need such fabricated support.
Also if people want their comments to be accepted, it is high time they do not resort to such blatantly wrong and wild statements.
Ananth:
Posted by: T.Achry at May 25, 2009 4:54 AM
Something I intuitively felt but now there are some stats to support it.
2000s : Tendulkar avg. 52.92 (B.q 37.7) vs. Lara 54.07 (Bq. 34.4)
In the 1990s Tendulkar was clearly the better batsman.
So tendulkars performance dropped overall by 9% in the 2000s. Laras conversely improved by about 5 %. In both cases with apparent better Bq.
This is a clear indication of the impact of injuries on Tendulkars performance.
Also the Lara avg. is a bit unclear because though the Bq. may have been better during the 400, it was on an absolute feather bed. And that one single inn. if factored in brings his avg. down from 54.07 to 51.1. So inspite of Tendulkars injury problems and not to mention playing some hundred ODIs more with thousands of more runs there…Lara basically comes out on top even in the 2000s based on one innings. [[
Why single out one or two innings, take away the same etc. After all Lara scored his big innings, albeit on a feather-bed, against England. Also then what do we do with Tendulkar's 248* against Bangladesh.
The point also is that the big innings do not necessarily contribute significantly to the Match Performance ratings. In fact Lara's 400* contributes well below his 153* or Tendulkar's 241* innings.
Ananth:
]]
Posted by: ranade at May 25, 2009 5:00 AM
Your last comment leads to some interesting observations:
1) If you discount Lara’s 2 highest scores and Tendulkar’s 2 highest scores :
Lara (375,400*; = 775 @ 775) and Tendulkar (241*, 248* = 489 @ n.a)….
Then Lara’s career reads as 11178 @ 49.68 and Tendulkar as 12284 @ 52.50.
So, would it be correct to say that besides Lara outscoring and averaging Tendulkar for just the 2 innings, Tendulkar has actually done better for the rest of their careers? [[
The point is that even without doing this theoretical taking-away exercise, Tendulkar's career figures are better than Lara's.
Ananth:
]]
This naturally leads to the second point:
2) Lara’s few big innings then distort true consistency. Most series nowadays are 3 test series. So if you use a 10 test slice that would mean over 3 series. So a few big scores may cover up a string of low scores and so would not be a true measure of consistency. Taking 3 test slices is more accurate. Over a period of 10 tests a class batsman is almost certain to find some form and put up a few big ones. [[
Your comment on 10-test slices is quite valid. However it should be noted that if we lower the number of tests to 3 and these tests split over years and long gaps, it will not make sense. That is the reason why I selected 10 tests since it would have normally spanned an year or so. There is also the question of the difficulty in handling too many test slices, in Tendulkar's case it would be over 50 such groups. But I get your point and may look at 5 tests.
Thanks for your point made in a helpful manner.
Ananth:
]]
Posted by: Sham at May 25, 2009 11:19 AM
Ananth,
“pankaj” though perhaps extreme has a point. If you take just the match performance points of Lara as compared to Tendulkar and Richards it works out to almost 22 % more .And then of course this parameter is given a 50% weightage.
Considering the “common sense” approach that Tendulkar has done as well or better than Lara against some of the stronger teams, but also giving that Lara has a handful of classics, to have Lara out by almost 22 % on this front doesn’t quite make sense.
Perhaps with the new means you intend to incorporate viz. diff. means of calculating bowing quality, consistency etc. this gap will reduce to a more sensible one.
Most people, including the pundits would have a hairline between Tendulkar and Lara mostly depending on individual preferences. But with this analysis the gap becomes a yawning chasm between both Lara/Tendulkar and Lara/Richards. So, the title of your blog “It figures” doesn’t quite apply to this post :) [[
Sham
A mail like this deserves consideration and an answer since it seeks to elicit a clarification in a proper manner.
I agree that the difference of 22% is probably not justified by the common sense approach. Partly, one reason could as well be that I had determined the total of Match performance points by summing the individual innings points but divided by the number of matches (with no ulterior motives I can assure you, but to get a good unscaled number). In the revised calculations I am going to divide by the number of innings. This will redress the balance. I am sure the other tweaks will go some way in reducing the gap.
It is unfortunate that people have gone to town on Lara having played in a weaker team and so has benefited. If so, think of this that Gavaskar, having played in a weak team for quite some time has benefited. It is an accepted norm that if a player delivers with lower support than another, this fact has to be recognized.
I have always understood that the discerning readers were not comfortable with the gap between Lara and Tendulkar and for that matter between Lara and Richards and I hope between Tendulkar and Richards. It is when the mud is slung that I see red. Good that you have taken the correct approach.
Thanks.
Ananth:
]]
Posted by: Marcus at May 25, 2009 2:28 PM
Just on Ranade's point on consistency, I just looked up the innings/50 ratio of both Lara and Tendulkar- maybe not a very sophisticated method of determining consistency, but a pretty handy one, I think. Tendulkar's ratio is 2.76, while Lara's is 2.83, so while Tendulkar is slightly more consistent based on this measure, the consistency argument doesn't quite clinch it for me. [[
Marcus
I am going to work on collection of tests as a group and determine the consistency across these groups. Obviously not penalize above-average performances but penalize below-average performances. Let us see how it comes out.
Ananth:
]]
Posted by: Ritwik at May 25, 2009 6:44 PM
Ananth,
Must say your work on the top test batsmen was the best of your work till date. I have one fundamental suggestion though - have you considered using multiplicative weights instead of additive ones? For one, due to the effect of the geometric mean trimming out extremities, such analyses are more robust to tweaks in weights allocated to different criteria. Second, given that the overall batting average and the overall bowling average are very close to each other (by definition), such an analysis will also give you a means to compare batsmen and bowlers across generations - something that can be useful in removing the double counting inherent in the 'today's bowlers are weaker but today's batsmen are also stronger' type bowling/batting quality adjustments made in different analyses.
I have tried something similar with the overall career stats of top batsmen and bowlers (no match stats data crunching possible with cricinfo records and a simple excel spreadsheet). [[
Ritwik
A few hundred comments back someone had asked why I had not used the Wisden-Hallmark Innings Ratings work I had done a few years earlier. I had replied saying that in the older analysis I had used additive methods for the Match performance analysis and because of the inherent drawbacks I have now done the revised Match performance calculations using multiplicative methods. So what you have suggested has already been incorporated.
For the career figures multiplicative methodology will throw everything out of gear. The base measure (say, Runs scored) will have so much importance that there will be no chance for someone with, say , 5000 runs even in the top-12. So here, additive methods are more appropriate.
Ananth:
]]
Posted by: KJH at May 26, 2009 3:33 AM
Nice one Ananth, what stands out for me is the bowling attacks! I can't wait for the bowlers edition.
Also, (AGAIN!) how good was Bradman! One of the common detractors used against him last time was the supposed weak bowling attacks he faced. So it was with great joy that I saw Bradman avg. 90 as highest scorer against an Eng bowling attack of 34.28, bettered (though not in avg.) only 6 times against all opposition; particularly Gooch, who faced an attack that must have been truly frightening @28! I now have a much increased respect for ol' Gooch & the WI attack of old. What does stand out though is SRT's absence from certain areas. Also interesting to see S.Waugh & Ponting faced (marginally) tougher attacks and scored more runs through the 90'-00's. So much for Aus dominance.
We should just appreciate how lucky we've been to see so many talented batsmen in modern times; Kallis, Dravid, Ponting, SRT, Lara etc., and realise that Tendulkar is just one among many (albeit better than most)
Posted by: Zeeshan Ahmed Siddiqui at May 26, 2009 6:53 AM
Tendulkar is technically more corretct than Lara.
I think comparison between Lara and Tendulkar is that both are greatest batsmen of this century. If Tendulkar is no. 1 then Lara's achievements will not effect by it or vice versa. We should focus on that how they have done for their countries. In one interview, Tendulkar said main thing is that what we have done. For him no. 1 and no. 2, does not matter.
Ponting is another category and then Dravid and Haq. These are top five batsmen of mine and then Hayden, Kallis, Yusuf, Sehwag and Sangakara or Jayawardene.
In Sehwag's case, he is rising day by day. Suppose, if we multiply his test career by 2 then it will become, more than 11,500 runs with 30 centuries with four triple and also ten double hundreds and also 22 scores 150 or more. It is not necessary that he will perform in same manner but he has a chance to score one or two more triple in his career again.
Also Gambhir is performing and soon he will get his name in top 10
Posted by: Pankaj at May 26, 2009 8:07 AM
I’m surprised at the apparent agitation about my comment. Actually it is common practice while testing a Theory to push the variables in its relevant formulae to its extremes. If the result isn’t coherent, i.e. if using an extreme value for a variable with its attached Multiplication Factor/weightage ends up with an impossible or absurd result it is a clear indication of a flawed Model/Theory. [[
Pankaj
There is no problem with what you did. It is only when the idea is used to achieve a limited objective that there are problems.
Let me also say this. If you extend your logic to not Tendulkar/Richards against Lara but rather Lara/Tendulkar against Bradman. Can we say that Lara (or Tendulkar) should have scored another 10000 runs in 25 innings to overtake Bradman. Bradman is that far ahead that any such suggestion becomes ridiculous.
Finally wait for Part-2 and then let us look at the whole thing.
Ananth:
]]
Posted by: E.j at May 26, 2009 8:20 AM
Ananth:
Even a cursory glance at your tables clearly shows us that something extraordinary has happened in the 2000s which has never before occurred in the history of cricket.
Batsmen have scored runs as if they were being distributed for free!
Even compared to the 1990s the change is drastic; With many many top class players around, only the “3 great players” (as you say) have averaged above 50 in the 90s. A Test batting avg. of 50 used to be the benchmark and reserve of the “Great” Batsmen. Now, apparently, it is par for the course.
If you look at the “Bowling quality” there is apparently not much diff. in the 90s figures and the 2000s figures. But something has undoubtedly happened…
Something dramatic. Which has made batting a relative walk in the park.
Based on the tables it is almost like cricket was a different game in the 2000s.
The stats don’t really show what happened. For eg. For those of us who have watched cricket throughout the 90s KNOW that the bowling was definitely much superior as compared to the 2000s (even though the bowling quality figures surprising show up the same). Also there is NO way the current lot of batsmen can be called “better” than the top class 90s lot…
So, I would like to ask this question of you and ALL of your readers:
“What in the world has happened which simply doesn’t show up in the stats?!” [[
My take, E.j, is that the pace of scoring has certainly been a contributing factor. If a batsman lasts 200 balls nowadays he is likely to be 100+ as compared to a few years back when he would have been 80+ and during the 60s when he would have been 60+.
Also the chances of one getting out when one adapts a more attacking vein is probably lower than the quantum of runs scored.
Maybe also the umpiring which tends to benefit the batsmen slightly more.
Ananth:
]]
Posted by: eddy at May 26, 2009 5:38 PM
hi Ananth, what about an allrounder analysis?
Sobers, Khan, Botham, Dev, Kallis, Hadlee, Rice, even Wasim, Flintoff, Gayle, Pollock, etc. As Salim Mian stated earlier it would be very interesting to find which of these players were the best 'Match winners', NOT just who had the best averages.
Maybe after your bowling analysis we will find out?
regards Eddy [[
Eddy
A very comprehensive all-rounder analysis was done a few months back. If you go down the blog pages you can locate it.
Ananth:
]]
Posted by: Gyan at May 27, 2009 6:10 AM
@ Ananth
You are absolutely right :There is no doubt whatsoever that runs scored in a weaker batting unit or runs which ultimately result in a team win should be valued more than runs scored under lesser pressure ,dead rubbers or losses.
The Real question is whether, in the final analysis, this tells us who is the better batsman relative to another batsman in another team or whether it indicates which batsman has done better relative to the other batsmen in his own team.
Tricky issue.
@Ej
Your observation is absolutely correct. Ananth says the scoring rate has gone up …but the reason that has gone up in the first place is that batting is infinitely easier. The Bowling quality index is misleading…pitches are flatter, fast bowlers have been defanged (bouncer rule), helmets (One may argue that helmets were around in the 90s as well, but the top batsmen then were groomed by prehelmet coaches for techniques in facing the short ball as kids) etc.
A batsman like Hayden for eg. plays some 90% of his shots off the front foot…Also, amazingly, 6 of the top 20 are basically products of the 2000s.
@T.achry
You are right about Tendulkar. At one stage most people thought the injuries had finished him...”End ulkar”. But he has hung on. Infact he has been relatively injury free since the beginning of 2008. If you take the years before 2008 his record will look much worse…in every respect…runs, avg., hundreds, consistency etc.
Posted by: Arjun at May 27, 2009 6:49 AM
Hi Ananth,
I think "weighted bowler quality measure" depends on the proportion of quick bowling (fast bowlers, meduim fast bowlers, meduim bowlers) and slow bowling ( leg spin, off spin, left-arm spin) faced by the batsman during their career. Quick bowlers tend to average less than slow bowlers. Typical average of quick bowlers is between 21.00 and 28.00, while typical average of slow bowlers is between 25.00 to 35.00. Most of asian batsman face high proportion of spin bowlers than fast bowlers, so their "weighted bowler quality measure" will be worse than non-asian batsmen.
eg. Ratio of lara. fast bowlers faced during career - 75 %, Spin bowlers faced 25 %.
Similarly, Tendulkar. fast bowlers faced - 50 %, Spin bowlers faced- 50 %.
Lara- W.BowQ- 35.21
Tendulkar - W.BowQ- 37.79.
Can you do a separate "weighted bowler quality measure" of fast bowlers and spinners for each batsmen ?
thanks,
Arjun. [[
Arjun
As I have mentioned quite a few times I don't have ball-by-ball data and so it will be impossible to separate spin and pace for the batsmen whether in individual matches or careers.
My feeling is that once I implement your idea of sqrt of bowler average x bowler strike rate as the "quality" measure and then weight that with the runs scored, these anomalies will iron out.
For instance Lara's 688 against Sri lanka will be valued slightly lower since Muralitharan's strike rate is only 54.5 against sub-50 for top fast bowlers. We have to assume that the batsmen would have faced the bowlers in the same porportion of balls bowled by the bowlers in the innings.
Ananth:
]]
Posted by: Venkatesh at May 27, 2009 8:56 PM
Ananth: A very useful analysis - I think most Indians would take umbrage if anything negative is written about Tendulkar but we need to recognize that Lara had fewer not outs - he flourished and perished with the same sword the reason he had some innings among the best ever and fell more often to mere mortals - he may be the better batsman. We do need to look at Sobers, Ponting and Barrington - wonder what their detailed analysis would reveal - among contemporary players, Ponting has a better average and strike rate, could pace an innings incredibly fast - Indians need to cut down the anit-Australian sentiment and acknowledge that he is indeed a great player, perhaps even better than Tendulkar.
Posted by: raghunath v.j at May 28, 2009 2:54 AM
why this obsession with winning the match alone.What if you are in a weak side and bat consistently and score runs against the best teams of that time.is that any less skillful or valiant?
in the period 53 to 55,take Clyde Walcott against England and Australia at home-see the no. of centuries and the average-against Australia,in the second and fifth tests,both defeats,he scored a century in each innings-twice in the series!!
nobody,not Bradman,Lara or Tendulkar has done that even for a winning side.
what would this feat be worth as per Anant's scoring,I wonder!! [[
Winning the match takes a weight of 5-10% of the Match Performance ratings, in other words between 3 and 6% of the final valuation. This is certainly not an over-weightage.
The only thing I would say is that if there are two very similar innings and one takes a team to a win as against the other not suceeding, the first one should get additional credit.
Ananth:
]]
Posted by: Kalverts at May 28, 2009 6:26 AM
Incontrovertible and irrefutable evidence that if not for injuries Tendulkar would have been head and shoulders above any other modern day batsmen.
Kalverts
I am posting a summary since otherwise your mail is too long. You have also ignored batsmen like Ponting who averaged a very high 70+ during the period you have considered. However I have posted your key argument.
Ananth
Posted by: Prashant J at May 28, 2009 11:29 AM
Ananth, it seems we can't see beyond Tendulkar, Lara n Bradman when the whole analysis mentions a lot of other great batsmen of all times. I was really expecting some interesting comments about other batsmen like Sutcliffe, Hammond, Hobbs and comparing them with Tendulkar, Lara n all. I am also baised towards Tendulkar and Lara was the one I enjoyed batting most and one of my unfulfilled dream has been to see Lara score big against India.
Ponting might have a high average but I would regard Hayden a better and more complete batsman. Why do we compare Lara n Tendulkar with Ponting but not Hayden ?
I would suggest Ananth should devote a separate statistical analysis for Lara vs Tendulkar debate and end it all there ;-)
As both of them are contemporaries it would be easier to jugge them on similar parameters .
-> Against best fast bowlers, best spinnners of their time
-> Home n Away
-> Quality of opposition
-> Match situation
-> Their own team's strength
etc etc
Posted by: Yogesh at May 28, 2009 4:45 PM
I have some reservation against your measure for quality of bowling attack. They seem to indicate that Tendulkar faced a better bowling attack in 1990's than 2000's. Most people believe it to be the other way round. Is this a wrong belief ?? But what disturbs me is the following stat : The quality of bowling attack faced by Tendulkar in 2008 is 37+ while in 2009 is 33.11. Infact, the 2009 attack of O'Brien, Mills, Franklin, Vettori & Ryder seems to have the second best weighted average in Tendulkar's entire career. In 2008 Tendulkar faced Australia(Lee, Clark, Johnson), South Africa (Steyn, Ntini, Morkel, Harris albeit on a chennai belter), Sri lanka (Vaas, M & M), England (Anderson,Flintoff,Harmison,Swann, Panesar). I cannot see how the Kiwi attack is rated better than these ??
Posted by: Yogesh at May 29, 2009 12:11 AM
Though this brings in more readership to your articles, statistical analysis of great batsmen/bowlers seem to be less illuminating than others. They just seem to give another ordering of the top 10 batsmen and some more fodder for certain fans to claim superiority. On comparison, your articles on T20 bowling or IPL seem to invite less manic or chauvnistic responses.
Posted by: Indian at May 29, 2009 3:48 AM
@Kalverts
based on the same argument.. if Lawrence Rowe did not suffer from allergy to grass and if he did not have mysterious eye problems, he would have probably ruled the cricketing world.
Posted by: ted at May 29, 2009 11:17 AM
not to annoy the lara and tendulkar fans the big problem with there ability is that they where never what you would call successful captains.thus players like border lloyd waugh vaughan ponting even a ranatunga to adegree stood up for there country when it mattered.bl@st stats might be good but give me one of those other captains anyday.that makes bradman even better considering he was captain too. [[
Ted
Captaincy is one thing, batsmanship is another. No point in confusing both. I myself have done different analysis for both taking care not to overlap one facet with the other. For people like Tendulkar, Lara, Botham et al, they were the greatest of their eras and they found it difficult to handle lesser gifted players possibly evaluating the others by their own lofty standards. A Brearley, Benaud or Border were able to do that.
Ananth:
]]
Posted by: Ritwik at May 29, 2009 12:08 PM
Echoing Yogesh, are you sure there are no calculation mistakes? 33.5 for the 2008 NZ attack and 36+ for the 2008 attack seem incompatible. [[
The above sentence is incomplete. I don't understand what is being said.
Ananth:
]]
Also, I don't understand how people are remarking that batting average has gone up because of pitches, over rules etc. If that was the case, bowling quality should have gone down concurrently, by definition. Since it has not (by an equal measure), your argument about faster cricket being better for batting seems true, as a first cut. We might even conclude taht today's batsmen are more skilled than what we give them credit for, unless one was to argue that the quality of both bowling AND batting has gone down and so there wouldn't be much change in the overall averages.
Posted by: Ritwik at May 29, 2009 2:19 PM
I am sorry there were some typos, and I presumed you had read Yogesh's comment. I was talking about the bowling quality index for Sachin, 33 in 2009, when he has played only against NZ, as opposed to 37 odd in other years, including 2008 when he played against seemingly stronger teams. [[
Ritwik
I have checked the figures. They are correct. Franklin and O'Brien have CTD averages of 32+ and Vettori/Mills/Martin have 33+. So the average works out to exactly the figure of 33+ for tendulkar.
Ananth:
]]
Posted by: Ananth at May 29, 2009 4:30 PM
Mr.Kalverts
This mail is specially for you. You have bombarded me with many long mails
with figures which I cannot post because I am not sure whether these are
correct as evidenced by the following obviously erroneous table.
Debut-2003
Ten : 169 i; 8811 @ 57.6 ;31 100s
Lara :157 i; 7572 @ 49.5 ; 8 ?
Pont : 99 i; 4246 @ 48.8 ; 4 ?
I do not have the time to check the correctness of all these numbers. Where
you sent a text mail I posted the same. I would appreciate if you avoid
sending any more numbers-based comments. Make your point in short
paragraphs. If these make sense these will be posted. Alternately keep your
ammunition for the follow-up article.
Thanks you.
Ananth
Posted by: Yogesh at May 29, 2009 4:38 PM
Your reply to V.J. Raghunath : " The only thing I would say is that if there are two very similar innings and one takes a team to a win as against the other not suceeding, the first one should get additional credit."
Why ? When two men make similar scores and one team wins, it says more about the team than the player. In effect, you are crediting a player for being in a good team. In 1990s, out of roughly 75 matches Sachin played, he was on winning side only in 17 matches (no Zim) and in that only 4 centuries. In 2000s with nearly same matches (no Zim & Ban), the stats are 25 wins and 8 centuries.
This is just because the team has been able to build upon his centuries. Tendulkar once made 114* on a pacy perth and ended up losing, but a year back he made just 71 on a less pacier track and still India won. The difference in the two matches was that the rest of the team did very well.
Posted by: eddy at May 29, 2009 8:40 PM
My dear Ananth, i really get the sense that you are at your wits end with some of these Tendulkar/Lara posts. However you also appear to have a sense of humour and some of your comments made me laugh out loud. By the way a tracked down the allrounder analysis you mentioned....very very good again and much as most people would have guessed.
Sobers the greatest and most complete player to play the game of cricket. Whereas Bradman specialised in batting to a level no other sportsman matched in their respected sport.
regards again, Eddy
Posted by: Kalverts at May 30, 2009 3:32 AM
OK Ananth ,as short as I can make it. Some typos there. Add a “1”to the 100s.
And will DEF. save some ammo for the follow up!!
@Indian
My arg.is diff. from rowe and well supported. In the less than 4 yrs from 2003-07 Lara scored some 40% of career runs and 50% of his 100s. Ponting some 50% of both. ALL the batsmen gorged on easy runs having the BEST yrs of their careers. Tendulkar, due to injuries,had the worst (10% or car. Runs and 100s). This was the BIGGEST and easiest time for batting in HISTORY.
the ONLY reason Lara is above SRT on that list (& Pont even close) is because of this 4 yr period when Lara made merry in a “weak” team , so even getting “bonus” points. NOT as is being passed off as Lara’s great inn. against Aus or handling of Murli or some such.
Till 2003 SRT did far better than Lara/Ponting overall .He also did better than Lara against ALL main Test nations. The stats/data are easy enough to check.
Posted by: Ananth at May 30, 2009 4:45 AM
[[
Mr.Kalvert has drawn his own observations which are not borne by numbers. I have given below the extract from my Test summary article which came out a few months back. There is no marked gorging of runs during the 2000-20003 period, during which Tendulkar is injured (???). There is a clear increase in the Left-hander's averages (Hayden/Lara/Sangakkara/Gilchrist/Gambhir recently). However there is no increase in the averages of right-handed batsmen.
Quote
1. Batting analysis 1 (Batting average - Left & Right)
Period R-Avg L-Avg T-Avg
Pre-WW1 22.73 25.33 23.06
WW1-WW2 31.68 29.73 31.40
40s-50s 28.44 30.98 28.81
1960s 29.72 35.37 30.82
1970s 30.29 32.86 30.79
1980s 29.65 33.61 30.44
1990s 28.21 33.32 29.45
2000s 28.99 37.89 31.68
All Tests 28.77 34.19 29.92
First the period changes. After a relatively difficult first period, the other seven periods have seen very little variations in Batting Average. The current decade has seen the best batting average of all times. This is almost 5% above the all-test average.
However the real shock comes when we see the Right-hand Let-hand figures. Barring a single period (the in-between Wars period - no doubt caused by Bradman & co), the Left handers have consistently averaged between 10 and 30% more than the Right handers. Across all tests there is a 15% variance. Look at the current decade. The Left handers have averaged nearly 30% more than the Right handers. I have no explanations. The readers will certainly have a few.
Unquote
This thread is closed and please do not send any more comments.
Posted by: Nitin Bindal at May 30, 2009 8:07 AM
Hi Ananth,I have got to commend you on the insightful breakdown. I also feel sorry for the people who feel that they must poke holes in the theory behind it.I think SRT is the best batsman ever as I have grown watching him bat,just like most of the people who have commented on this blog. Hence we feel a bias towards favouring him.One part of the problem in the comparison between BCL and SRT is he fact that Sach played against a weaker WI bowling attack (by your own stats) than Lara did while playing India.If we could have a like for like comparison of Sach facing the indian bowlers and lara facing the WI attack,we probably would have a better understanding of how good both players are(and not which one of the two is better)They both are great.I wont be surprised if Sanga & Yousuf are the best batsmen with Punter, Smith & deVilliers @ this moment in time,but the memories that BCL and SRT have provided will always over shaddow all the others
Nitin
Posted by: Salim Mian at May 30, 2009 9:16 PM
Apart from Tendulkar fans/lovers constant denial of your findings and their dismay at Lara being second to the DON, i keep reading feeble excuses of what might have been. 'If it wasnt for Tendulkars injuries' or 'Before tendulkar's injuries'. My god are we really suppose to make up what might or might not have been? If George Best had have been english they would have won the world cup. If there hadnt been WW2 Bradman would have....? If If If If.
Tendulkar fans also appear to have a problem with 375 and 400*. They speak of 'featherbeds' and 'deadruns' and 'if your take away 400*....'
A test run is a test run, whether it's 400* in Antigua against Flintoff, Jones, Harmison, Hoggard (the same attack that won the Ashes 15 months later) or 248* against Baisya, Mortaza,Rahman, Rafique and Rana.
Good night and thank you. [[
Salim
I am publishing this letter since it makes a valid point.
No player scores all his test runs against good attacks nor all the runs against weak attacks. The attacks vary in quality and this fact has been incorporated in the weighted bowling quality. All arguments lose credibility if only one side is presented. It is unfortunate that some of the readers feel the need to put down one (great) player to pull up another (equally great) player. People must learn to appreciate greatness from everywhere.
Ananth:
]]
Posted by: Pankaj at May 31, 2009 2:55 AM
kalvert has made a point which rings true.
a basic perusal of the stats will reveal that SRT was well ahead of all contemporaries till the beginning of 2003- this, as the stats reveal, is plain fact.
So, the others caught up with him in the period 2003-2007.
The hows and whys can be debated - but this basic fact cannot. [[
What you say is probably true. However I have analyzed the careers in their entirety not specific parts. While the peaks and troughs of each batsman can be looked at and analyzed, these are part of anyone's career. If Tendulkar was not injured or if Dravid's form did not slump or if Ponting did not go through his horror time in India or Lara's sojourn to South Africa was more succesful et al are questions which can be raised and debated endlessly but cannot be taken out of the whole equation.
Thanks for making your very valid point in a good and acceptable manner.
Ananth:
]]
Posted by: Ramanathan at May 31, 2009 9:20 AM
Dear Ananth. I Wanted to check whether in the analysis can we also consider the following:
1. runs scored against left arm and right arm bowlers against the batsmen (in my opinion left arm seamers are difficult to play for right hand batsmen and vice versa)
2. number of lifes batsmen got during an innings.
I know that i am asking toomuch.. but just thought if this will give any added weightage to any batsmen's performance.
Regards
Ramanathan [[
Ramanathan
First of all I do not agree that right hand batsmen against left arm seamers have a problem. Even if that is true what about left hand batsneb against left hand seamers or for that matter against right hand seamers. I think it is essential to work on the basis that if one is a good batsman he has to face up to all types of bowling and score. The top three batsmen of the past 20 years, Lara/Tendulkar and Dravid had no weakness against any type of bowling.
There is no known information on the lives got by batsman during an innings. Also what about earlier years.
Ananth:
]]
Posted by: Zeeshan Ahmed Siddiqui at June 2, 2009 7:00 AM
I think you are right that top three since 1990 are Tendulkar / Lara and no. 3 Dravid. They had no weakness against any type of bowling.
Ponting always has a problem with off spinners. This is the lacking in part of him. He proved him as a leading batsman but I think he never touch the class of Lara and Tendulkar although we can easily count him in top five batsmen among his contemporaries.
What about Inzamam as he can easily tackle fast bowlers and spinners. He used to play lazy and lost his wickets so many times. Please let me know if he faced any problem to any sort of bowling except sometimes, he is sluggish. Yousuf always has a problem with Murli although we can count him in top ten among his contemporaries. [[
Zeeshan,
I would say the only weakness with Inzamam was within himself, his judgement of a run. Notwithstanding this flaw, he is good enough to be rated in the Top-5 batsmen group of the 1990-2000 period and to be bracketed with Miandad as Pakistan's best.
Ananth:
]]
Beside these facts, I think the most fit batsman in test and one day is Tendulkar with 40 or more centuries in both versions with 12,000 or more runs in both forms of cricket.
My own opinion is that Tendulkar is equivalent to or very near to legend Sir. Bradman. [[
I have no problems with that assessment. I myself have said that and my future study should prove it. The only thing I get tired of reminding readers is that the current study is a Test-only one.
Ananth:
]]
Posted by: Shiraz at June 4, 2009 2:05 PM
Ananth,
Here’s a suggestion which is more “match specific”…rather than using vague generalisations over all matches alike.
Awarding a Batsman in a weak batting lineup extra points for coming early or batting with the tail is fine. However, that is likely to be a chronic condition in a weak batting team.
A Batsman in a Strong batting lineup should then be awarded even MORE points for similar situations. Since, if he has to come in early or bat with tail etc it means that the bowling/pitch is probably MORE potent than will show up in general numbers such as Bqty (ctd), pitch qty etc.on that particular/unique occasion. [[
Shiraz
Your comment is a facetious one. It is similar to saying that the lot of the rich is more difficult since they have difficulty in choosing which option to go for as against the lot of the poor who have no choice.
To come in to bat at 10 for 2 is always a difficult situation. Let me also say that whether Lara came in at 10 for 2 with non-performers to follow or Tendulkar came in at 10 for 2 with 3 very good batsmen to follow ARE EXACTLY IDENTICAL SITUATIONS in my analysis.
Also whether Lara moved from 100 for 6 to 300 all out or Tendulkar moved from 100 for 6 to 300 all out ARE EXACTLY IDENTICAL SITUATIONS in my analysis. The quality of the late order batsmen would be the only difference.
There has been lot of disinformation in these areas. I would appreciate your considering the above statements carefully.
Ananth:
]]
Posted by: Shiraz at June 5, 2009 4:42 AM
Ananth,
Again, on reading my previous comment I AGain don’t know if I am being clear!!
So I will try again with an eg:
1)Richards( In a strong batting line up)… 10/2 (so 3 good bats out)…then 100/7 (another 5 good batters out)…So, definitely bowling/pitch quality deadlier than may be suggested by blanket Bqty/pitch qty.?
2)Lara 10/2(weak batting line up)………...(2 poor bats out)…then 100/7 (5 other poor bats out)…this may not be because of deadly bowling/pitch…but simply because of basically inept batsmen.
So ,though case 2 may be given additional points as per your system…case 1 should be awarded MORE points ,although on the surface it appears to be the same scorelines in both cases.
I hope this eg.is clearer! [[
Shiraz
What you say is a round-about route. It requires an analysis of the batsmen dismissed, their quality and then deciding that the bowling was good or not.
It is not necessary. If Richards' 10 for 2 was off Lillee/Thomson it will be recognized so. If Lara's 10 for 2 was off Fraser/Caddick it would be recognized so. Or vive versa. I determine the actual bowling quality instead of a round-about method.
Also 10 for 2 on a flat bed will be recognized. Example, Lara's 375 and 400 are not valued high at all. In fact Gooch's 154 is valued much higher than these two knocks.
I get your point and let us close this.
Thanks.
Ananth:
]]
Posted by: Rk at June 6, 2009 3:04 AM
Ananth,
“Shiraz” may be on to something.
Your system of “match performance” for batsmen you is based on the underlying assumption that this performance fluctuates on a match to match basis (and rightly so).
This fluctuation will also occur to bowlers performances over matches.
So, with what “Shiraz” is proposing we can ascertain the bowlers performance for the “particular/unique” match in question, and also factor that in (in addition), instead of simply(only) using blanket bowling/pitch qtys/avg.s .which are not very indicative of a bowlers then form/performance for the particular match in question.
This would be a nightmare to analyse, I’m sure!!!
(Though it is undoubtedly more accurate and reveals the truer picture)
Posted by: Dr.Raks at June 6, 2009 3:30 AM
I feel your concept about pre 1940 & post 1945 batsmen is misguided.
The periods in consideration should denote differences in some radical method of playing the game itself. (say graphite and wood rackets in tennis)
In cricket the single most drastic change has been the protective helmet and bouncer rule.
This has enabled batsmen to take risks and adopt techniques which would have earlier been either suicidal or inconceivable.
The only reason a Hayden (or sehwag etc) is above say a Boycott is because of the helmet.
That’s all.
The pre helmet era batsmen first had to save their very lives…then score runs. “Seeing off the new ball” has now become an arcane concept. As has “back and across”…Most of the batsmen simply plonk a foot either way and take a mighty swipe. One of the main causes of damage earlier used to be a top edge on to the face. Now, the fear of hooking is almost gone. Even a batsman as prolific as Lara has been hit innumerable times on the head and has never mastered ducking techniques. Most of the modern batsmen are either “swayers” or “happy hookers”.
To most neutral observers there is no way batsmen such as Lara/sehwag/Hayden etc would have survived long in the prehelmet era.
So placing these batsmen in an “All time” list is grossly inaccurate.
They are products of their times. This must not be forgotten.
Also, as several ppl have mentioned, the fact that almost half of your top 20 are there mainly because of a 4 year period (2003-07) clearly signifies the advantages batsmen had at that time which batsmen of an earlier era did not.
There is no way we will ever again see a Richards(helmetless) get out of the way of 5 consecutive Lillee bouncers and get bowled on the 6th ball.
So, if we are making lists of periods…a more accurate one would be batsmen who played without helmets and no bouncer rule... and batsmen who did.
Posted by: T.pt at June 6, 2009 11:04 AM
@Shiraz
You are right, but ironically NOT for the exact reason you mention.
Of course, a single fixed number cannot capture a bowler’s performance across all matches…Oddly enough tho,what ACTUALLY happens is that even an avg bowler “lifts” when bowling to a great batsman.
This, of course, cannot be captured with mathematical/statistical formulae, or any such apparently logical process.
This phenomenon of playing “above yourself” on certain occasions is common across all sports. As is said: “Its easier to become No.1 than it is to STAY there”…because then everyone is gunning for you.
So, the great batsmen would actually be facing a “Bowling Quality” (or whatever you want to mathematically call it) which is BETTER than can be depicted statistically.
If you see some of the truly great balls and even great catches are often off a great batsman… even the fielders “lift “themselves!
You can well imagine how a Viv/T’kar/Lara scalp would look on a bowler’s victim list. Something to “tell their grandchildren about”…
Posted by: Nik at June 7, 2009 9:24 AM
@Dr.raks
Certain flaws in your argument:
1) Even Richards has been hit on the head. So have the two most technically accomplished batsmen of the current generation: Tendulkar and Dravid. So, it is not that they have NEVER been hit-just that they have been hit few and far between, as compared to the mechanical regularity with which the other batsmen got hit-even with a limited bouncer rule in operation.
2) Lara’s problems against raw pace were not so much because he was getting hit on the head. Donald’s complete control over Lara was mainly because of hostile short pitched stuff to the throat/ribs-not so much the head. This used to cause Lara to awkwardly fend off, try to weave, so upsetting his rhythm and balance. His inability/unwillingness to duck (as you correctly mention) is what caused the problems against genuine pace. If he could have combined ducking/weaving depending on the ball ,the going would have been much easier.
3) Your parallel of tennis with wood and graphite rackets is incorrect. In tennis BOTH the players have the same equipment, so it gets neutralized. In cricket the fast bowler has been rendered comparatively impotent with ALL advantages going to the batsman. His main weapons of intimidation and fear, along with the ability to keep the batsman guessing and off balance have been cut off at the knees. One wonders whether they should abandon the bouncer rule now that batsmen have helmets.
4) Actually, the periods you mention (pre and post helmet) are quite complicated and difficult to define. This is because even though Bradman and co. also faced hostile fast bowling (inc. bodyline)-it was really used as a weapon of mass intimidation only later by the WI fast batteries and then Lillee/Thommo/Donald, Pak bowlers.( Perhaps the reason the England dominated ICC decided on the bouncer rules!!)
So, though I agree with your general argument there are a lot of finer points involved
Posted by: hblove at June 9, 2009 7:17 AM
Hello Ananth,
Very interesting analyses in this blog of yours. I peruse them frequently and enjoy the different nuances that I feel I glean each time, no doubt also helped by the often perceptive and insightful comments of other readers!
Here's a request, if you have the time and the inclination, of course!
Could you please do an analyses between modern batsmen and the golden oldies including factors such as the increased incidence of lbws over time, especially from 1990 onwards. I do feel that there is a lot of criticism of modern batsmen and how easy they have run scoring and while I do feel that there is some truth in the matter, I do feel that it is taken for granted without some of the subtleties being taken into consideration. Now, obviously some subtleties such as lighter bats and longer boundaries etc. might be slightly harder to quantify without ball-by-ball data; however, the rising incidence of lbws over time would be easy to incorporate into the analysis. (contd.) [[
I have done an analysis recently in which I separated the test years into 8 periods and derived various tables. You could look at that, if you have not already done so.
Ananth:
]]
Posted by: hblove at June 9, 2009 7:24 AM
(contd)...I do believe that there was an analysis done in the 'guest' blogs recently by two gentlemen which showed the increased incidence in lbws, especially since the 1990 ie when the neutral umps started coming onto the scene first, I'd imagine. I suspect that the increased willingness of an umpire to give a modern batsman out lbw (if statistically significant) will and does easily outweigh all the perceived advantages in better bats, better protection, covered pitches etc. that they certainly have over their forebears since it would essentially be almost like another mode of dismissal (or a significant fraction thereof!) that has suddenly entered the bowlers' armoury, in a manner of speaking!
Sorry for this rambling email but I thought I'd just write and bring this up with you.
Thanks again for all your interesting analyses!
Posted by: Salim Main at June 9, 2009 9:17 AM
@ DR.Raks.....to say Sehwag, Lara, Hayden (Tendulkar you dont mention?) would not survive without helmets is as meaningless as any 'IF' question posted before in these 400 odd comments. I have already spoken about such nonsense 7 comments before yours. Do you see any good hookers today? Nope! Back in the day you had Viv, Greendige, Fredricks even Botham went after Lillie in 'Bothams Ashes'...all without helmets. You would think that helmets would encourage hooking but in fact it's a part of the game that has died. If anything helmets have made the current players lazy. There is no fear, so no need to learn a wonderful explosive stroke. Would Viv have been a worse batsmen 'IF' he had worn a helmet???
Posted by: Sheet at June 12, 2009 6:31 AM
@nik
Your point is very good. And I don’t think many people realize it. It was only with the WI fast batteries etc that fast bowling became really intimidatory. So, as you say, approx the 70s ,80s were perhaps most deadly.
Surely, earlier if they wanted they could have put on helmets? Motorcyclists etc. used to wear helmets even during WW1. So the idea and technology was very much there.The fact that they didn’t wear helmets during cricket then was perhaps that it simply wasn’t so dangerous and so not required!!
So , the bouncer rule ,helmets etc only came into being much later when it was clearly realized that the batsman was in grave physical danger!
So, as you say…then perhaps the most difficult time to face fast bowling was probably during the 70s,80s.
Posted by: Nik at June 12, 2009 1:19 PM
@sheet
Yes. Actually a brief look at the decade wise avgs. as given by Ananth tells you a lot more than the overall analysis does.
In the 2000s just about everyone avg.s 50.
90s: The 3 true greats: Tendulkar, Lara, S.Waugh.
80s: Just 2 Bats.
70s: 2 again
60s: 3 again, and so with the 50s.
WW1-2: almost all the top scorers avg 50
Pre WW: 1.
Also, the sheer quantum of runs scored in the 2000s is unprecedented. More than 10 batsmen have already scored more than 6000 runs.
This tells us that pre WW was probably the most difficult (due to pitches?)
Then the 70s and 80s. The 2000s have been the best time in cricket history for batsmen- this shows up in the type of analysis as done here.
As mentioned, this sort of decade wise perusal tells us more than a standard blanket analysis covering all 140 yrs of test cricket. It is highly illogical and incorrect to state that most modern batsmen are better than those of prior generations based only runs scored and avgs.
@salim, dr.rak
I would have to lean towards dr.rak. as per the above you can clearly see that batting was undoubtedly easier recently. It is difficult to accept that suddenly all the batsmen are superior to the 70/80s ones. If you see, the standard of bowling, esp. fast cannot be said to have been inferior to that in the 70/80s. so, the batsmen now are facing similar fast bowling but with better protection, better bats, modified techniques etc. i.e. the balance has indeed shifted dramatically in favour of the batsmen. As far as Richards is concerned, he would have certainly scored more heavily if in the 2000s...esp. considering that he was actually a bit ahead of his times as far as strike rate and attacking cricket was concerned.
So, to a great extent, the scales definitely tip towards the modern batsmen, esp. in the 2000s as far as batting is concerned. And it is not a very fair comparison to the 70/80s batsmen to say that just about all the batsmen who scored heavily in the 2000s are superior to the earlier ones. This just isn’t true.
Posted by: eddy at June 15, 2009 1:02 PM
@ NIK.... only two batsmen with 50+ in the 80's. i'd say there were three.
Border played two years in the 70's and 4 in the 90's but 10 in the 80's he avg 50+. Surley more of an 80's batsman????
Their is a similar spilt with Miandad and Richards 50+
Posted by: Nik at June 17, 2009 6:07 AM
@Eddy,
Yes you are right. So tried to check the recent player’s stats in another manner as below:
And I confess my initial hypothesis doesn’t seem to be quite true! Am a bit flummoxed, which just goes to show with stats…
However …
Some folks in here may be coming on a bit strong. However, they have a good argument.
Batting is now undoubtedly easier, especially in the 2000s.
This may not show up in overall general tables for ALL batsmen which Ananth has posted earlier in reply to some comments, but clearly for a good batsman ,if fit, the 2000s were the place to be.
The stats show that 38 batsmen in history avg. 50+ to date (MIN.20 matches), albeit for their entire careers.
Splitting into 3 periods (B4 1979;70/80s;90s and 2000s- so some players will have their careers split in by this period, as you correctly mention, here are the avgs. (Min 20 matches)
15 Mar 1877 to 31 Dec 1969(93 yrs) :14 batsmen avg.50+
01 Jan 1970 to 31 Dec 1989(20 yrs) : 7(Mia,Chap,Bor,Ric,Dean,Gav,Boy)
01 Jan 1990 to 31 Dec 1999(10 yrs) : 4(Ten,Stev,Lar,Gooc)
01 Jan 2000 to Date (
(As you mention some player’s careers would overlap a bit, so NOT totally accurate and a bit of a rough split)
But still…..
All quite revealing:
So, actually from the 60s to the end 90s things seem more or less equally tough for batsmen….hmm..i.e,only the true greats avg.50+.
BUT(and this is a big “but”)..If the players who avg.50+ and made their debuts after the mid 90s are counted (and if they now retire immediately)…they would account for MORE than half of all players in history who avg. above 50(min 20 matches)!!
So, I guess if given a choice most batsmen would have loved to bat in the 2000s, especially looking at the stats-the mid 2000s(2003-07) as several ppl have mentioned.
So, I guess it is the combination of all things put together (a perfect storm):
Dead pitches, protection, modified techniques, bouncer rule, lack of truly great attacks etc etc which made the mid 2000s a paradise for batsmen and a seemingly unique period in cricket history.
Posted by: Susobhan at June 25, 2009 7:58 AM
Another well researched and deep analysis by Ananth Narayan. Statistics are useful only when it is deeply viewed. Anath Narayan does it again. Tables show the true performance by the batsmen.
Lara faced the most vicious attacks of 28.49, 28.28 and 29.93('99-01)and avereged 59, 29 and 64.Even against Aussies Lara faced an attack of 31.39 and avereaged 51, No batsman is as succesful as him against the ferocious attack. So I personally think he is the best post WW2 test batsman.
Posted by: Ramesh Kumar at June 26, 2009 6:56 AM
Mr. Ananth,
Very good analysis. One observation. Some of the players have a long career. While they score more runs as they play more matches, it will be impossible to have high averages for a very long period. Will it make sense to factor some greater weightage for a player's longetivity?
regards,
Ramesh Kumar
Posted by: diggygong at June 30, 2009 9:25 AM
We have all these debates ( some substantive some not so substantive)because there isn't a 'standard criteria' for judging who is or is not was or was not the best batsman of all time. Therefore friends it can be concluded that "the best batsman" title is a relative one so 'relatively' speaking "the Prince", "the Jewel of the Caribbean", "the best stroke player to play the game 'thus far' ( watch out for Darren Bravo ),"the 'Krypto-Litharan"etc etc is best batsman to play the game!!!
Posted by: Woril Turner at June 30, 2009 5:11 PM
I know your analysis is based on statistics, but the only way you can use those statistics to compare batsmen across different era, is to use performances under the same conditions, i.e. pitch conditions, pacemen of the same speed and variation, spinners of same quality etc. Otherwise this is seriously flawed, regardless of what the statistics say, because you are not comparing apples with apples. Note how may bats men average over 50 since the 90s and look at what has happened in cricket since the 80s and you realise why. There is the # of bouncer rule per over, there are no good quality pacemen around, the seam on the cricket ball has gotten thinner, the pitches are dead and so all sort of jokers whose average would not have been more than 20+ if they had the pace quartet of the WI or Australia to face in the 70s and 80s are now 50+. Just look how some of these guys fared against Ambrose and Walsh when these guys were in the twilight of their career and you see the truth.
Posted by: cliff at July 20, 2009 4:18 PM
Superb analysis, I was pleasantly surprised to see Alec Stewart heading the 1990's, which considering he was also keeping wicket for a lot of games in this period is a remarkable performance
Y Anantha Narayanan has over 35 years of IT background. Over the past 15 years, he has been concentrating on Cricket analysis and software development. He has been involved with StumpVision, Wisden, Hallmark Software and his own site www.thirdslip.com during this period.
David Barry was cricket-starved when teaching English in France, and
study of cricket stats was his only way to stay sane. He is now back
in Brisbane, Australia, and working towards a PhD in Physics. He once
played for the worst team in the G-division of Muscat's cricket
league.
After doing an MBA in marketing and working in an advertising agency, S Rajesh decided that his skills might be put to better use by number-crunching on cricket. He hasn’t regretted that decision in the last six years, and edits the Numbers Game column on cricinfo.com every Friday.
Andrew Samson had his moments with bat and ball, once scoring 43 and taking 3 for 14 with his legbreaks, but he was much better at arithmetic, which explains why he is where he is today. Andrew has been keeping cricket stats since the days when it used to be done with pen and paper, and has been involved in scoring/stats for Radio and TV since 1987. He has been Cricket South Africa's official statistician since1994.
A former scientist and occasional TV quiz champion, Charles Davis now works full time at sports statistics in Melbourne.
His only real contribution to the Test record books came at age 4, when he formed part of the record 90,800 crowd
who saw West Indies at the MCG in 1961. He has two books to his credit, and claims to be the only cricket statistician
ever who has been quoted in the New York Times and in Australian Federal Parliament on the same day. Not to be
confused with the West Indian batsman Charlie Davis, especially in terms of ability.
Having just taken early retirement as a Mathematics teacher in Hobart, Ric
Finlay now fully devotes his time to recording cricket, both past and
present, for the popular CSW cricket database, along with his colleague
David Fitzgerald (www.tastats.com.au). His interest in the game is
inversely proportional to his ability as a player, but he did once score a
century after being dropped at 3 and running out three of his team-mates.
His first memory of international cricket is the 1962-63 MCC tour of
Australia, described as one of the most boring ever. Totally fascinated, he
was instantly hooked, and has never looked back. Author of three books on
cricket of a historical nature, he has provided statistics and scored for
radio and television cricket coverage since 1983.
RSS Feed
