A most interesting piece of stat: Calling Varun

[Guest BossBhai]

-- Removed on request of the user --

[Guest BossBhai]

-- Removed on request of the user --

[Sachinism]

excluding drawn tests rules out some good match saving innings

[The Outsider]

Without somehow normalizing for the strengths of the bowling attacks and batting line ups featuring in the matches, the above stats are inherently pretty weak. For example, Tendulkar's 114 and 5 runs at Perth form a trifling 10.5% of the match aggregate while his 165 at Chennai next year is 15% of the match aggregate. Do we all not know which was the better and more challenging innings? BTW, to eliminate the boring draws from the rest, you can eliminate the ones which had less than 4 innings played out in the match (declaration would count as a completed innings, I suppose). Eliminating all the draws is injustice to innings like Gavaskar's 221 at the Oval.

[teacup]

Without somehow normalizing for the strengths of the bowling attacks and batting line ups featuring in the matches' date=' the above stats are inherently pretty weak. [b']For example, Tendulkar's 114 and 5 runs at Perth form a trifling 10.5% of the match aggregate while his 165 at Chennai next year is 15% of the match aggregate. Do we all not know which was the better and more challenging innings? BTW, to eliminate the boring draws from the rest, you can eliminate the ones which had less than 4 innings played out in the match (declaration would count as a completed innings, I suppose). Eliminating all the draws is injustice to innings like Gavaskar's 221 at the Oval.

Yep. A major inning on a flat bed with the inning declared at 5 or 6 wickets down will skew the ranking.

[patriot]

Yep. A major inning on a flat bed with the inning declared at 5 or 6 wickets down will skew the ranking.

Over an entire career - that is not likely to make any more than a very marginal difference.

[patriot]

So Bossbhai what are we looking at. Rerunning the query: 1) Without Bang and Zim 2) But including drawn tests ( except those with only 3 innings)

[patriot]

I'd exclude Bangladesh and Zimbabwe and re-run the query. These two sides really skew the figures. BTW Patriot...see the magnanimity of the man? Tune itni gaaliyan dee, phirbhi usne nehin mukraya. Boss...:adore:

I am feeling guilty..but not a lot. My anguish was against BB's views ( which i think are very imbalanced as far as SRT goes), not BB the person. I'd dare not utter a foul word in real life. against someone who started watching cricket way before I was even born. Sorry BB, dil pe mat lena.:icflove: IN future, I know what to do, to avoid any bad blood.

[zen]

Nice work by Bossbhai!

[teacup]

Over an entire career - that is not likely to make any more than a very marginal difference.

Not in Lara's case. His top 3 scores - 400, 375, 277 - were are all boring draws. That amounts to more than a 1000 runs or nearly 10% of the total runs he made in his career. And you are pretty clever aren't you - considering 3 inning draws against 4 knowing that it will eliminate most of the huge scores (including the above 3 scores) from the statistics?

[Dhondy]

Not in Lara's case. His top 3 scores - 400' date= 375, 277 - were are all boring draws. That amounts to more than a 1000 runs or nearly 10% of the total runs he made in his career. And you are pretty clever aren't you - considering 3 inning draws against 4 knowing that it will eliminate most of the huge scores (including the above 3 scores) from the statistics?

Maybe they were boring draws because of what Lara did- i.e. put the match beyond England to avoid even the slightest chance of defeat? By the same token, I agree with our resident Nobel Laureate...all draws must be considered, irrespective whether they are 2, 3 or 4 innings. There's no way of saying whether that innings precipitated that draw or the draw-predisposed milieu conceived the iniinings. The final table then, Boss.

[patriot]

Not in Lara's case. His top 3 scores - 400' date=' 375, 277 - were are all boring draws. That amounts to more than a 1000 runs or nearly 10% of the total runs he made in his career. And you are pretty clever aren't you - considering 3 inning draws against 4 knowing that it will eliminate most of the huge scores (including the above 3 scores) from the statistics?[/quote'] The same can be said of Sachin. His top 4 scores are : 248 ( against Bangers) , 241 ( a boring Sydney draw on a 700 + pitch) , 217 ( another boring draw against NZ on a flat home deck) and 201 ( Yet another flat pitch draw @ home, this time against lowly Zim).

[patriot]

Maybe they were boring draws because of what Lara did- i.e. put the match beyond England to avoid even the slightest chance of defeat? By the same token, I agree with our resident Nobel Laureate...all draws must be considered, irrespective whether they are 2, 3 or 4 innings. There's no way of saying whether that innings precipitated that draw or the draw-predisposed milieu conceived the iniinings. The final table then, Boss.

Over 2 Boss: Final table excluding the 2 minnows BUT including all matches ( irrespective of result).

[The Outsider]

You guys are committing the same folly that Wisden did in picking their top 100 test innings - too much weight to the sheer volume of runs scored in match, which led to innings like the following by Jayawardene making their top 100 : http://www.cricinfo.com/ci/engine/match/63831.html This exercise is pointless without contextualizing with the strength of bowling attacks and batting line ups involved. But I am going on like a broken record when apparently most of you seem to be happy with it staring right through the glaring flaw in the analysis, so I'll rest.

[Dhondy]

Without somehow normalizing for the strengths of the bowling attacks and batting line ups featuring in the matches' date=' the above stats are inherently pretty weak. For example, Tendulkar's 114 and 5 runs at Perth form a trifling 10.5% of the match aggregate while his 165 at Chennai next year is 15% of the match aggregate. Do we all not know which was the better and more challenging innings?[/quote'] Normalisation is occuring because of the sheer sample size involved. And we are not judging the quality of individual innings. The question is over the length of a career, who's the guy who's outscored his peers against all-comers? The concept has construct validity. Not a single

[patriot]

You guys are committing the same folly that Wisden did in picking their top 100 test innings - too much weight to the sheer volume of runs scored in match, which led to innings like the following by Jayawardene making their top 100 : http://www.cricinfo.com/ci/engine/match/63831.html This exercise is pointless without contextualizing with the strength of bowling attacks and batting line ups involved. But I am going on like a broken record when apparently most of you seem to be happy with it staring right through the glaring flaw in the analysis, so I'll rest.

I see where you are getting at - but that innings of 240+ by Jayawardene, was 50% of his side's total , when his side was facing a massive 520ish total by the opposition. Credit in such big scores is duely given to what % of your team's total you make up. IN any case, that is ranked 91. Now that Wisden ranking was done in 2001/02. I dare say, Sehwag's knock of 201 in Galle, will figure amongst the top 15 of all time ,should Wisden release it's latest set of ratings , because of he did almost 70 % of his own side's score on a treacherous pitch, Sehwag's individual score was more than the oppositions score in a whole completed innings and it resulted in victory for his side.

[Dhondy]

You guys are committing the same folly that Wisden did in picking their top 100 test innings - too much weight to the sheer volume of runs scored in match, which led to innings like the following by Jayawardene making their top 100 : http://www.cricinfo.com/ci/engine/match/63831.html This exercise is pointless without contextualizing with the strength of bowling attacks and batting line ups involved. But I am going on like a broken record when apparently most of you seem to be happy with it staring right through the glaring flaw in the analysis, so I'll rest.

Stop being so fussy. You know very well that there can't be a perfect analysis that takes into account all variables, without making it absurdly abstact, a la Ananth Yawnfest. This is just one way of looking at merit lists. Strength of bowling attacks has been covered here. http://www.indiancricketfans.com/showthread.php?t=22987

[Mr. Wicket]

Stop being so fussy. You know very well that there can't be a perfect analysis that takes into account all variables' date=' without making it absurdly abstact, a la Ananth Yawnfest.[/quote'] But wouldn't an econometric regression be more effective here in accounting for far more deterministic/independent variables? If someone really had the time/effort to build one, it'd be possible to add in variables for just about every bowler (good or bad) and see the impact that the presence of said bowler had on the batsman's average (or whatever other y-variable you set).

[The Outsider]

Stop being so fussy. You know very well that there can't be a perfect analysis that takes into account all variables, without making it absurdly abstact, a la Ananth Yawnfest. This is just one way of looking at merit lists. Strength of bowling attacks has been covered here. http://www.indiancricketfans.com/showthread.php?t=22987

No, I'm not suggesting a 20 parameter covariance calculation. Just trying to think of a simple, intuitive way to make the above results more meaningful by picking and dropping some important parameters. Something along these lines : http://www.indiancricketfans.com/showthread.php?t=14687

[Guest BossBhai]

-- Removed on request of the user --