A minor point, but it appears that in the commentary to the 1st table above you are confusing Bert Sutcliffe of New Zealand with Herbert Sutcliffe of England.
[[
Deon
I stand, and the article stands, corrected.
Thanks
Ananth:
]]

Since 'Maximum ratio reached' table is all about comparing great batsman won't it would be lot better if only peer avg. of 1-6 or 1-7 is considered (same as your 1st table)?
[[
Arjun
Good point. Although my reading has been that the final ratios whether they are based on 1-11, 1-7, 1-6~7 are somewhat proportionately comparable with very little variations. However let me do that and post the results.
Ananth:
]]

I liked that you posted the 'highest peer rating reached' table. This is because now users don't need to post Tendulkar's prime era or Dravid's prime era and yell about their golden periods.. I would really have liked you to also do the same in the best batsmen/best bowler anaylysis. You could have say the highest rating that a batsmen got(as per our analysis). For eg. Lara has a career score of 50.26 points and Sachin 49.24. I'm pretty sure Sachin would have touched 50 at some point of time in his career(say May 2002).(The thing that Lara might have touched 51 is another thing). This way, you could have stopped a few comments of people who are staunch Sachin's fans who keep telling he is the greatest and keep asking you to rate him upto May 2002. Dravid himself would have touched 48-49 points..

Sorry for being a little off-topic though.
[[
Yash
As you have done consistently recently, the point you have made is very valid. Only problem is that it is a programming and calculating nightmare. One day I will do it because it would, to a certain extent, reflect the erstwhile P-W Ratings.
Thanks.
Ananth:
]]

Ananth,

Regarding your earlier comment:
“One day I will prove analytically that Kumble/KapilDev have contributed as much or nearly as much to Indian cricket as Tendulkar. “

If you use a run/wicket ratio of 20 then Kumble has contributed 14,886 “runs” to Indian cricket, Kapil Dev 13,928 – and Tendulkar 13,653.

If you look at these numbers on a “per match” basis it is even more in favour of Kumble and Kapil Dev.

But this obvously doesn't paint the whole picture.
[[
Deon
You are asking for trouble.
Anyhow when I do the analysis it will incorporate a lot more. Contribution to wins, share of team workload, all-round skills, possibly captaincy in addition to all the bread and butter numbers. Let me also add that I would do this not to denigrate Tendulkar in any way, he is amongst the best ever, but to give the others their due.
Ananth:
]]

Ananth
I couldn’t find Hussey on the list?
[[
Abhi
Hussey is 33rd in the main file.
Ananth:
]]

Yash
I don’t know if you are referring to me. But if you are, you have perhaps misunderstood my point. To take a rough parallel with tennis, suppose Federer is No.1 for 5 years and Nadal for 1 year... Don’t you feel that this reveals something? Perhaps not if players are from different eras…but if both are playing at the same time? Similarly if Tendulkar has been “better” for a greater majority of a peer’s career, shouldn’t this factor be given some weight?

Also the fact that he was better than the others at a clearly much more difficult time for batting?

This is all I’m saying. Through the 90s Tendulkar was out on top. As is quite clear Lara comes out ahead on the overall stats because of the 2003-07 period which adds considerable ballast to his stats. Similarly with Ponting. Richards comes lower on the list. But again the “peer comparison” tells us that when Richards was good…he was really good! Ananth noted that Richards suffers because towards the end of his career he had a few poor years. Similarly Tendulkar had a few poor years, ironically at the exact period when the others were piling on the runs. This is noted by how Richards and Tendulkar suffer in the peer ration “difference” list.

Ananth, xolile
Firstly, where modern cricket is concerned ODIs are to be given their due. In fact, till a few years back the Indian public was probably more obsessed with ODIs.
Secondly, if you seriously think that you can “analytically” prove something like you claim…then you are not only completely overestimating the limits of statistics but also displaying more than a little amount of naiveté.
[[
Kris
One thing is certain.
What I said I will do is a LOT MORE credible and objective than the "billion people" argument. Think of the absurdity of the billion peoples argument, you will know.
Who said I will not take into account ODI.
Ananth:
]]

I know this is just from top of your head, but the number has to be much higher; for 2000 onwards, it easily has to be something like 27-29. Further bowlers are limited by the maximum number of wickets(20) while batsmen are not limited by 400 runs(20*20)
Also Kumble would have scored many more runs if he batted at no 4 instead of 8/9 simply by having more and better batting partners present.Somehow,somewhere this also needs to be taken in consideration.
And people/crowd/audiences may like batsmen better and believe they are more important than bowlers.But you just need to go ask international captains who will they choose between a top bowler/batsmen and I am sure they will pick up bowler. I am dead sure Ricky ponting would have picked McGrath/Warney in Ashes even if that meant him sitting out for whole series. At least I would have.
If you use a run/wicket ratio of 20 then Kumble has contributed 14,886 “runs” to Indian cricket, Kapil Dev 13,928 – and Tendulkar 13,653
[[
Love
Let us avoid these simplistic calculations. People will pounce at such comparisons, with justification, and go to town.
It is essential to consider lot more factors, across different forms of cricket before coming to any conclusion. I will repeat what I told earlier. My purpose was only to say that the others who have performed equally well for over 20 years seem to be short-changed. It is never my intention to pull down Tendulkar. He is too great for that. However let us give the others also credit because they have also carried a bat into the ground at 15 for 1 in a cauldron of 60000 people or bowled over after over all round the world.

Ananth:
]]

Ananth, if you compare the batsman average with the combined bowling average in the test matches he played, that could be a good indicator of how good a batsman was because it will take into account the quality of bowlers, pitches, weather, etc.
[[
Prashant
If you have read the posts on top batsman analysis you will see that I have already done what you have suggested and more. For each match I take the weighted bowling average. Thus matches in which Imran Khan played as a batsman, his bowling average will not be included.
Ananth:
]]

A request to the readers:
Can you please stick to the subject covered in the article. Probably it was my mistake to allow an extraneous topic to be raised. The above comments have been published because they were received before this request of mine.
Jeff (Grimshaw) has put in hours of work and sent me a weighted peer ratio for the top players. My thanks to him. I will incorporate his findings in this article as soon as possible.
Ananth

I have a good idea(I'm sorry if someone would have already suggested this)

Instead of doing peer calculations considering only team mate's calculations(this unnecessarily inflates someone like Flower's ratio and deflates Ponting's ratio) or using a the entire period's calculations which can be unfair due to nature of the pitch, conditions etc. Why don't you just add up the player's own team PLUS opposition's numbers and calculate the peer average?
In this case:
1) A player like Flower would enjoy a good advantage of his own poor teammates but it would be compensated by the opposition's high scores because of his team's poor bowling.

2) Similarly, in case of Ponting, the disadvantage of having good batsmen like Hayden, Gilly etc. would be offset by low opposition averages due to team's good bowling strength.

3) Only perhaps a player like Sachin/Dravid would suffer because India's batting strength is very good but bowling, pretty average when compared to its own batting.

Makes sense?

Hi Abhi,

I wasn't pointing to you or to anyone in particular. In fact I agree with you that Sachin's period from 1989-2002 was awesome and players like Ponting look inferior if you see that period of Sachin. Same with Richards. But then, why didn't Sachin and Richards retire in 2002 and 1987 respectively. They took the gamble(I'm not telling that its been bad for their own teams) of playing the additional years. Had they scored heavily in the latter periods, you'd have included those in the calculations. But now that they haven't lived upto their own expectations(still an avg of 49 post 2002 is pretty good for an ordinary batsman), you cannot just discount the calculations.

Having said all that, I must confess that I'm one of the biggest Sachin and Lara fan. Off-topic: But probably Kambli was better than Sachin or atleast as good, but ALAS!!!!

Hi Ananth,
Sorry for bothering you again...this is a repitition of my question in your previous article.
Is there any effect of the timeless tests in surtcliff and hobbs's peer ratio/record? I'm just curious; but did not find enough record in the net.
I know this is not in the interest of the majority readers here..can you tell us something about those time less tests spanned between post world war1 and 1930s?
[[
Romel
I am indeed sorry for not taking any action on your comment. Have put in a reminder for myself. Will revert soon.
Ananth:
]]

Before Tendulkar last player to have 'Maximum Peer ratio' of over 2.00 is Javed Miandad and that was in 80s. Since than it has become exremely difficult to outscore your peers by alomst 100 %. Incidently it is also a period that saw rise and rise of ODIs. Another factor could be abundance of good batsmen in last 30 years. This will make virtulally impossible to reach Maximun peer ratio of 2.00 in future. So well done to Tendulkar.
Actulally i am waiting for Comparision of Maximum ratio with only proper batsmen.i.e 1-6.
[[
Arjun
If you see the Peer Ratio of Pietersen, that is 39+. In other words, he has to reach 78+, after 50 tests, to reach 2.00. So, as you say, Tendulkar might very well be the last of the great batsmen to reach 2.00. That top-10 certainly is an exclusive club.
Look what happened to Hussey.
Ananth:
]]

Ananth,
I am not a real huge cricket fan as such. A normal fan. But for many years now whenever Tendulkar has been batting I have always been glued to the match. When he was out my interest in the match diminished considerably. This is also true for most people I know. Is this unfair to the other team stalwarts? Definitely. But what to do? That’s just the way it is.
[[
Purohit
I have no problem with such comments since you are only talking about yourself.
Ananth:
]]

Ananth-resending my previous comment. I think I mixed up the names.

@arjun
Yes. I don’t think it is the quality of batsmen. I feel it is mostly the nature of pitches in the 2000s. Several pitches have been extremely disheartening for bowlers and so very dead that sometimes it felt that practically any reasonably competent batsman could simply help themselves to a hundred. Also, there don’t seem to be the same quality of bowling attacks in the 2000s. Only the few truly outstanding bowlers. And it is these bowlers who have been successful. I mean if the “peer average” is 40, it is incredible!! This used to be considered the benchmark for a “good” Test batsman in the 90s! (50+ were the “great” batsmen)

Arjun,
The “max peer” ratio backs up your subjective arguments quite well. Bravo.

Ananth, is it possible to see what highest peer ratio MEK Hussey has achieved (42 tests, avg 52.65 at the end of the Oval Test). I recognise that bringing the cut-off to 42 just for this calculation is not fair and also it will take a lot of working. So can we arrive at an estimate based on the assumption that he will average 45 for 8 further tests? I understand this isn't most accurate (as other's performance also needs to be accounted for), but can you do it please?
[[
Ravi
No problems at all. As easy as falling off a log. Remember when James wanted Faulkner's figures. I just do a special run. Takes a no more than 10 minutes.
Only thing is that I still need a minimum number of tests which I have taken as 20. At the end of Hussey's 20th test (Test # 1862 (2008)) vs India, he averaged 84.80, which when compared with the rather high Peer average of 39.12, we get the outstanding maximum Peer ratio of 2.17.
Now he is on earth with figures of 52.65/40.02/1.29. Note how the Peer average has crossed 40. Amazing.
Ananth:
]]

Ananth : Not a comment to post.
Otherwise I don't see another place to write it.
w.r.t a wicket's run value. From the basic batsman peer average table, we have seen that mostly this is hovering around the 30 run mark consistently. Can this be the 'prize' of a wicket in terms of runs? [I'm still amazed with the fact that 30 repeating for all eras and that too on very extreme numbers like Tendulkar's peer runs such as 7 lakh odd]
[[
Unni
30 seems to be a bit too high. I would expect a number between 20 and 25. Will do a small analysis on this one of these days.
Ananth:
]]

Ananth-
As I glanced thru the tables particularly the one on Maximum Peer Ratio, it seemed plausible that most batsmen (if not almost all) in this list would attain the Max PR somewhere during the middle stretch of their Tests played. That seemed to hold true. The exception is the one and only Sir Don who attained the Max PR at the beginning of his last Test. I thought that was incredible and, needless to say, continues to show why Sir Don is head and shoulders above the rest.
[[
Youvi
It also happens that Bradman played only 52 tests keeping his maximum-minimum window to 2 tests. However it is almost certain that he would have maintained a similar average over a longer number of tests, not necessarily after he was 40 but between 1939 and 1945. He averaged 108.5 and 97.14 in the two series during 1938 and 1946-47 respectively at either end of this 5-year gap.
Ananth:
]]

Jeff Grimshaw had done some extensive work on weighting the Peer averages by the quality of bowling teams. This was done to take care of situations such as Ponting not facing the Australian bowlers. He had sent me his article and I have now incorporated this in the main article after editing for presentation. Many thanks to Jeff.
Ananth

Ananth,
How about a max. peer ratio along “Jeff’s lines too. and also as “arjun”has been asking for a 1-6 peer comparison both on your table and similarly for “jeffs” table?
[[
Kris
Lot of work. Will do when I can.
Ananth:
]]

Alright,
But is the 1-6 peer comparison also a lot of work? Because as “arjun” says that gives us a much better comparison of batsman vs. batsman, instead of bringing in allrounders and such.
[[
Kris
I also have a lot of other commitments. So it will be done in a few days.
Ananth:
]]

Regarding Kris' comment:
I still believe that comparing batsmen to 1-5 will give the best result over time. In the past No 6 has often been viewed as an all-rounder position, and players coming in at No 6 were therefore not specialst batsman in the stricktest sense of the word. Since WWI No's 1-5 average 38.61 whereas No 6 averages 32.75; moreover, this difference fluctuates wildly overtime as ideas and strategies change.
[[
Deon
There is no way we can stop at 5. The no.6 position has been manned by some of the greatest middle-order batsmen ever, to name a few, Steve Waugh, Sobers, Gilchrist, Lloyd, Laxman et al. For that matter the no.7 position itself is fluid. The Wisden-100 all-time greatest innings, 270 by Bradman was played at no.7. Hence what I have now done, which is to take the nos 1-6 and no.7 with batting average exceeding 30.00 seems to be the best solution.
Ananth:
]]

Ananth - if we were to use this set of analysis (peer analysis as well as the earlier one on best batsmen)as a decision tool, whom would you pick between Vishwanath, Laxman and Azhar for the number 5 slot in a hypothetical all time Indian XI? (I am assuming that Dravid and Sachin will bat at 3 and 4)thanks.
[[
Aditya
I will always answer these questions from my heart. Like the time when I selected 5 bowlers, not necessarily in the top-10. I will go with Laxman, if nothing else, for his 281 and 167 although these masterpieces were played at no.3 and no.1. And that is my prerogative. The numbers need not always support this decision.
Ananth:
]]

Hey Ananth,

Did you consider my idea above as well? Taking only the stats of both own team and opposition team would remove a lot of discrepanicies like pitch, ground, conditions etc. Please see my post above. This is a simple yet fairly accurate measure of a peer ratio of a batsman. Your new method 'weighting the Peer averages by the quality of bowling teams' maybe better but it must be a little complicated?

Jeff's calculation is really good though since it brings Lara and Tendulkar little over Ponting which seems correct.
[[
Yash
"Own team" has been done. I am immersed with various bits of work. But will do "both teams" one day. I have no idea how I will present it to the readers. I will ensure that you get a copy.
Ananth:
]]

I am requesting you to put a table of batsman who scored max runs against any team considering the bowling strength of that team considering some ratio in between to rate the batsman that how good he is against that team (home & away).
Similar analysis can be created for bowlers too.
[[
Amit
Your point is not clear. Can you please explain a bit more.
Ananth:
]]

I find the “maximum peer ratio” achieved table simply wonderful. It very clearly tells us why certain batsmen were considered the best at their peaks.
Overall, due to various ups and downs their careers may end up looking different. But the “maximum peer ratio” backs up with some statistics why the general public felt that certain batsmen at their peaks were in their times the best in the world.

Arjun Hemnani has asked for a Maximum/Minimum table based on the top-6/7 batsmen only. I have completed that work and have uploaded the tables to my site. It can be downloaded by clicking on the following links.

http://www.thirdslip.com/misc/peermax1.txt

http://www.thirdslip.com/misc/peermin1.txt

Ananth

Ananth,
I take your point on the bating line-up cutoff point. The main benefit of what I suggested is that it will slightly favour modern batsman over the oldies. We are talking about maybe 2% or 3% in terms of peer ratio, which is not significant.

Looking at Table 5 above, and also noticing that Vettori is again among the runs against Sri Lanka today, I was wondering whether you could publish the maximum peer ratio reached for any 80 innings (or 50 match) stretch during a batsman’s career? The results will be most interesting, as it will highlight players that have been dominant for a large part of their careers, but due to either a slow start (e.g. Vettori, Imran, Waugh) or a late career fade (e.g. Sobers, Tendulkar, Dravid), have not been able to sustain this dominance over their entire careers.

The benefit that this will have is again that the moderns will be pushed up the list. At the moment we have the Top 11 in table 2 above all coming from the pre-1970 period.

To add some colour to my point above:
Vettori averaged 16.49 during his first 67 innings (after being selected as an 18 year old finger spinner). Since then he has averaged 41.33 in 74 innings. And he is still improving; in his last 10 innings he averaged 56.12 runs. Imran Khan went through a similar early development phase. So have Steve Waugh.
[[
Deon
You have given me an excellent idea of doing a comparison of a batsman/bowler's performance in two halves of his career. Then the comparison is with himself and might make a lot of sense. It also will show the trends which are present in careers.
Ananth:
]]

@xolile
As usual you (and arjun) provide the best and most discerning comments. What are you guys- a mixture of cricket pros and stats whizzes?!
But again the Dan Vettori innings adds more weight to my argument of Murali vs. the top lefties. For some reason he can’t seem to break through several top lefties. No such problem with the righties. One of those mysteries.
This is perhaps similar to the puzzle as to why the leftie tennis players get more “swerve” on their serves to the ad court than righties get on the deuce court. No one can really figure it out.

Last 30 years, Tendulkar's Max peer ratio(1-6) of 1.61 is best of all. This inspite of playing in ODI's era. Just to get an idea of what today's Batsmen need to do to achieve that hieght in this Batting Era(Peer avg. of Top-6 has crossed 40.0).

In 50 Tests(or 80-90 Innings), Score close to 5000 test runs at an average of more than 65.0. Seems almost impossible with increase in 20-20 mts.
If someone decides to stick to only Test Cricket and not play ODI's and 20-20, he can reach the ratios of pre ODI era as Top-8 batsmen(above 1.63. They all basically played Test Cricket.
[[
Arjun
The highest current Peer Ratio for qualifying batsmen at the end of the Slk-Nzl match is Samaraweera with 40.30 which will require an average, again in over 50 tests, of 65.
Ananth:
]]

Ananth,
Maybe another point you can take into consideration while doing your combined Test and ODI analysis is that it is very difficult to reach such hieghts if you play more than one form of cricket. Not even Ponting has reached same greatness in oneday cricket.
[[
Arjun
One reason I am not rushing into that. I am going to do couple of light-weight ones before attempting that mammoth task. Also remember that I have done the Peer ratio work in great depth for Tests but not ODIs.
Ananth:
]]

when you calculate the peak peer ratio achieved by a batsman, why do we always start from the begining of his career. Some players are slow starters but have blossomed into all time greats.

e.g. In case of Steve Waugh, he started out as an allrounder and had a batting average of late30-early40s for quite a few tests. But after about 50 tests, he really took off and ended with average above 50.

Can you do a maximum peer ratio backwards in a career as well.

one trivia on Steve Waugh - all of his 15 MoM awards have been match winning efforts.

Hi Raghav Bihani,

Cricinfo says Steve Waugh got only 14 Man of the match awards in Tests, not 15.

Just to add “the cherry on top” of my previous arguments there is a blog by “S.Rajesh” titled “The decade of the batsmen” right here on cricinfo itself.
I had essentially argued that several batsmen filled their boots in the mid 2000s and that this period was much easier for batting. Essentially a run scored in the 90s is worth more than the mid 2000s...So several mathematical attempts (Bqty, pqty etc don’t really capture the picture accurately in the transition period) .Also, Tendulkar being injured almost continually from 2003-07 and particularly after the 2003 WC where he was in great form…and aged 30…so we have probably missed out the BEST of Tendulkar (incredible as that may seem) from age 30-34!!
So, all the “best batsmen” analysis heavily “favour” the 2000s batsmen. i.e those who have started around this period or those who scored heavily in the period- as opposed to the predominantly 80/90s batsmen or the batsmen like Tendulkar who missed out on most of the fun due to injuries. So, the “best batsman” title in this scenario is a highly dubious one.
S.Rajesh’s blog clearly validates most of my points.

I stand corrected. But all 14 have been match winning efforts. None of the men above him have managed that. Though the person tied with him, Ambrose, has a similar record of 14 awards in 14 wins.