Tuesday 25 November 2008

The meaning of 77, or The Don recedes

99.94 is the one stat that, to cricketers, does not need to be explained or contextualised. You don't need to be told what it means or who it belongs to. 99.94 stands as the landmark number in the sport.

It's widely known, too, that 99.94 makes Bradman not just the best cricketer of all time, but the best sportsman. Statistically, no-one else in any sport has dominated as Bradman did. 99.94 made him almost 40 per cent better than anyone else who has ever played cricket, a margin that Pele, Nicklaus, Jordan or anyone else cannot approach in their disciplines.

Jeff Thomson once said, 'I didn't believe that anyone could be twice as good as Greg Chappell,' and then he saw the Don messing around in the nets as a 60-something, no pads, dispatching it everywhere.

I came across a lesser-known study that attempted to compare Bradman with players from different eras. Using a 'coefficient of variation' of batting averages, it calculated that a modern player would need to average 77 to match Bradman. 

The study was made at the end of the 90s, before the current era of the bat, so the number may have ticked up a notch, but modern gods Ponting (57), Yousuf and Kallis (55) and Tendulkar (54) fall a long way short. Mike Hussey is the closest at 64, but it feels like he's not begun properly yet. Gilchrist was in the 60s too at one point. None of Pollock, Headley and Sutcliffe, who all finished with career averages above 60, made more than 5000 test runs; Pollock and Headley made less than Hussey has now.

Still, 77. Not even on the horizon. And now it may never be. What I was driving at in this post was that the measures that describe success in cricket feel as though they are about to change. An average, as long as it's acceptable, already means less in limited overs cricket than strike rate. Just as Twenty20 has accelerated 50 over matches, so it will accelerate Tests.

Geoff Boycott, a surprisingly progressive commentator, has already suggested four-day Tests, played as day-nighters, would be a more viable commercial proposition. It will surely happen.

When it does, the meaning of stats will change, subtly at first, and then irrevocably. 99.94 will prove harder to understand.

In the study that showed Bradman was statistically better than anyone in any other sport, the third-most dominant athlete was baseball's Ty Cobb. He played a version of the game that is unrecognisable to the baseball fans of today. They attach more meaning to power-hitting records like those held by McGwire and Bonds than they do to Cobb's base stealing and RBIs.

How we adjust 99.94 in a new era will become important in keeping the game connected to its history. The Don is receding, but not in meaning.


Damith S. said...

A superb post old batsman !

Yes the meaning of those older stats have less and less meaning to the modern generation.

I hold Bradman in a god like status but some ppl are now starting to question whether bradman was actually that good , cuz he played less opposition and most of runs came against eng etc.

i think how we approach this will in the end be the determining factor in whether we still hold bradman and the past greats in the same plane.

ps- i think you have added me wrongly to your blogroll :)

it should be

you have added cricketanytime by mistake i think?

David Barry said...

When it does, the meaning of stats will change, subtly at first, and then irrevocably. 99.94 will prove harder to understand.

I disagree when it comes to Test cricket. As long as innings aren't limited, fundamentally the average will remain by far the best simple measure of batsmen and bowlers. Day-night Tests would reduce averages because in most places it's harder to bat under lights (and this would, in my view, change the game much more than is desirable), but there have been low-scoring eras in the past. Not just in the early days, but also in the 1950's.

I'm not sure that average means less than strike rate in limited-overs cricket for batsmen. You need at least some minimum value in both to be successful.

In T20 cricket economy rate is much more important for bowlers than average. In 50-over cricket I'm not sure, since I haven't done a detailed study on it. My suspicion is that the bowling average is still the most important thing.

The Old Batsman said...

Thanks Rob, I'll email you.

sorry Damith, have sorted the flyslip listing out. Technology is not my strongpoint...

Hi David, good point. I do hope the primacy of average remains, although I guess ultimately what we're trying to measure is effectiveness. As long as there's always some way of comparing that, then Bradman will be unapproachable. I've always got a kick out of the fact that cricket has the most dominant sportsman, although Woods might win 40 per cent more majors than Nicklaus... maybe!

David Barry said...

I have seen some golf ranking thingy which on face value looked like Woods is roughly as dominant as Bradman was, albeit not in terms of wins in Majors. Wayne Gretzky in ice hockey is another one who could statistically challenge Bradman as the greatest sportsman ever. I had a look at Gretzky vs other NHL'ers earlier in the year, and found that he was roughly as many standard deviations above the mean as Bradman was. But Gretzky played in a high-scoring era, and I forgot to transfer my NHL data when I moved back to Brisbane, so I can't do era adjustments and get a better estimate on it.

The Old Batsman said...

There was a great stat on Woods that at one point last year he was further ahead in the world rankings of the guy at number two, than the guy at number two was ahead of the world number 1000. My rudimentary grasp of stats reckons he'd need to win 26 majors to be as far ahead as Bradman, so he could do it.

Would be really interested to see the Gretsky stuff, too. I love these kind of stats, even though I have no idea how to work them out...

Damith S. said...

David, dont mind seeing the stats on Grekzky either.

Put them up when you have a chance.

And Cheers OTB.

FifthBeatle said...

Who is/was the second most dominant sportsman as compared to their peers? (I'm assuming you have 1. Bradman, 3. Ty Cobb)

price per head reviews said...

oh I did not have any idea about the meaning of 77 or the Don recedes, but I do know after I read your blog and I thank you for the information