April 13, 2016

How do you compare players across eras?

You can't rate apples against oranges, we're often told. True, but there still are ways of discussing players of one age in the context of those from another

When you compare Ajinkya Rahane's batting to David Gower's, you learn something new about both batsmen © BCCI

In my real job, I occasionally review scholarly papers for academic journals. Last week, the Journal of Quite Interesting Value Theory sent me a strange piece to look over, one that didn't really fit any of the norms of professional philosophy. The writer - peer reviews are anonymous, so I don't know who it was - was arguing, it seemed, for the value of human over natural beauty. The main argument appeared to be that a particular person was more beautiful than even paradigm cases of great natural beauty, and in several ways.

Anyway, I told the journal to reject it. Even ignoring the curious style, with its archaic language and the complete absence of footnotes or a literature review, the basic premise was misguided. How can you compare a person to a summer's day? What did it really mean, when you got right down to it, to say a person was "more lovely and more temperate"? Wasn't it egregious anthropomorphising to claim that "sometimes too hot the eye of heaven shines"? As I pointed out, I think incontrovertibly, heaven doesn't have an eye.

Finally, with a somewhat daring flourish in which I continue to take a certain amount of quiet pride, I said the piece was "comparing apples with oranges".

Which, of course, is what some people think I did in my piece on Virat Kohli and Sachin Tendulkar. I'm not interested in defending or renouncing that piece. But I am interested in exploring one objection people made to it, namely that the whole project of comparing the two is invalid because they played in different eras.

There certainly appears to be something to this. For example, in 35 Tests, Archie MacLaren scored 1931 runs at an average of 33.87, with five hundreds and eight fifties. In 40 Tests, Yuvraj Singh has scored 1900 runs at 33.92, with three hundreds and 11 fifties. The numbers are similar, but it would be absurd to use them to say that the two players are about as good as each other. On the one hand, MacLaren played on worse wickets and with much worse equipment. On the other, Yuvraj played in an era with much better standards of fielding, greater focus on tactics and analysis, and higher standards of fitness and training, which one would expect to increase the impact of the average quality of the bowling he faced. How on earth can we make any kind of meaningful comparison between the two?

"Better than" is only one type of comparative judgement, and when it comes to cricket, by far the least interesting. Let's say you manage to somehow establish that MacLaren was better than Yuvraj. So what?

One solution might be to compare within eras, and then use differences with peers to make cross-era comparisons. So, for example, you see how much more Don Bradman averaged than everyone else in his era, and compare this with how much someone would need to average now to have the same difference, and voila, you have an apparently knock-down argument that demonstrates Bradman was better than any other batsman who has ever played.

But there are problems with this too. For example, Stephen Jay Gould has argued (in the context of baseball) that the reason Bradman-type careers don't happen nowadays is not because there's no one with Bradman-type talent but because the average level of cricketers around the potential Bradmans keeps going up. Whether or not this hypothesis holds, the underlying point is that it shows another issue for cross-era comparisons, namely that if you're going to compare by using distance from the mean, you have to be sure that the level of the mean is constant over time. Further, in a sport like cricket, the average level has an effect on the performance of a given cricketer, so if bowlers are on average better than they were in the 1930s, this will have an effect on the performance of batsmen.

There's a much longer debate to be had, of course, and one can achieve a lot more with numbers than I can even understand (Anantha Narayanan regularly does interesting pieces of this kind on this site). But still, these sorts of problems can motivate the view I mentioned, that we should simply stop making comparisons across eras. This is legitimate to believe, and you can certainly defend that view. But it's worth pointing out one of its consequences: it means you can't say that Don Bradman was a better batsman than Chris Martin. That's simply entailed by saying you can't make comparative judgements across eras - it doesn't only mean that you can't make difficult comparisons, it also means you can't make the ones that seem obvious.

"Better than", however, is only one type of comparative judgement, and when it comes to cricket, by far the least interesting. Let's say you manage to somehow establish that MacLaren was better than Yuvraj. So what? What have you learned beyond that?

The more interesting sort of comparison, at least to me, is a kind of "reminds me of" relation. This is comparison as simile or metaphor. As Aristotle pointed out a while ago, we learn from a good metaphor, because when you say X is like Y, you learn something new about both X and Y. By regarding an unseen similarity, you learn a new feature of both X and Y.

Just off the top of my head, for example, Ajinkya Rahane reminds me of David Gower. I make no great claims for the educational value of this simile, but I think even at first glance it's more interesting than "Rahane is better than Gower". There are some obvious similarities I see between them, namely the elegance of their strokeplay, the almost peaceful way in which they strike the ball (neither of them hits it, even when it goes for six). They bat like Nehru under the influence of Gandhi's non-violence.

And there's something a little less obvious, perhaps. No one has ever accused Rahane of lazy elegance. Grit, a capacity for hard work, determination, a certain quiet, undemonstrative toughness - it's obvious Rahane has these qualities. The comparison between him and Gower may lead us to wonder, then, whether people sell Gower short when they talk about his foppishness; whether, in fact, Gower was also, in a certain sense, steely like Rahane seems to be.

What you think about the details of the comparison isn't important. My point isn't that I'm right about Gower and Rahane. Rather, it's that when we compare across eras, it may be more fruitful to talk in open-ended ways, to ask questions that invite qualitative discussion, to use simile and metaphor, rather than to ask the tired old question of whether one player is better than another.

Pranay Sanklecha is Assistant Professor of Philosophy at the University of Graz. @PranaySanklecha

Comments have now been closed for this article

  • Billy on April 20, 2016, 13:25 GMT

    @jay57870, it works both ways. 1930's cricket was markedly different with no professional training, formal coaching, long and tiring schedules of travelling by ship, & uncertain variables (pitches, conditions, weather, injuries, fatigue, etc). The bar was just different back then. You had a full time job and you had to be away from your family for long periods with no technology. After all, Don played 70+% of his Tests against England in unfamiliar conditions/grounds because I'm sure conditions change every 4 years. Also, they only played a few test matches a year. So the longevity in his results is there in both first class and Test cricket. Yes, he was the greatest of his era. And yes he was and is the GOAT (greatest of all time)!! You see how I changed your own biased hypotheses and came up with reasonable arguments to counter everything you said. People try, no one succeeds in the end.

  • Jay on April 20, 2016, 12:24 GMT

    The great Don's 99.94 is unmatchable. But it's an "average". The fixation on "one-size-fits-all" metric is overdone. As Don himself admitted: "No such comparison is possible in cricket. Averages can be a guide...but are not conclusive because pitches & conditions have changed". He felt it's "harder & more complex than in most other sports". Therefore: To hypothesise - that Bradman would have averaged no less now - is a futile exercise. Who knows? To assume he'd have even maintained his magical 99.94 is statistical nonsense. No statistician worth his salt can extrapolate results, given the uncertainties of the real world. Today's 24x7 global cricket is markedly different with 3 formats, tight schedules, intense competition & uncertain variables (pitches, conditions, weather, injuries, fatigue, etc). The bar's much higher. After all, Don played 70+% of his Tests against England in familiar conditions/grounds. Yes, he was the greatest of his era. But he's no GOAT (greatest of all time)!!

  • Clint on April 20, 2016, 5:08 GMT

    Australian academic Charles Davis wrote a fabulous book that uses statistical analysis to an academic level of rigor (based on deviation from mean) looking at comparisons. The results are clearly that Bradman is not just the greatest cricketer of all time, but almost definitely the greatest sportsman. People like Sobers (and more recently Kailis) tend to equate to Pele, Nicklaus etc in other sports in terms of how far ahead they are from their peers; Bradman is off the charts. Anyone literate in econometric theory /statistics would/should argue that there is no way he would average any less now. Incomparable.

  • GV on April 20, 2016, 4:50 GMT

    Also, regarding Bradman, I think he would have been tops in any era. In the bodyline series, he averaged the series highest, at 56, cmopared to McCabe at 42, Woodfull at 33 etc. So absolutes came off, but relatives were consistent. We cannot say that we believe Bradman saw himself in Tendulkar and at the same time say that cricket has changed so much since Bradman's time that comparisons are impossible. If it had changed so much, then why did Bradman see a resemblance in batting technique? Improvements keep happening but over decades, the rate of improvement is not the same. I think there is a massive difference between the actions of Bedser, Lindwall, Tyson, Trueman etc. and moving to the 70s, those of Thompson, Holding, Lillee, Imran, Roberts. The latter group was significantly faster on average, and with beautiful streamlined actions. Since then however, not much has changed and bowlers are pretty much in the same speed range. IMHO.

  • GV on April 20, 2016, 4:12 GMT

    Many nice comments. Some points. A reasonable numerical comparison can be made across eras as follows (it was done by Anantha Narayanan in these columns three years ago) 1) construct a purely numerical "degree of difficulty index" for each innings of every batsman's career based on average score in the match of all batsmen, recent history of bowling averages of the opposition bowlers, home or away, 1st or 2nd inn etc.; 2) classify all innings of all batsmen into 3-4 broad buckets (say easy, normal, tough) for this Index; 3) compute the batting averages within each group. In easy batting eras, higher proportion of innings will go into easy group, and in tough conditions eras, high proportion will slot into the tough group. But across eras, the numerical cutoffs will be similar. So what one ends up with is comparable. In the tough group for instance, batsmen like Mark Waugh, who usually shone under pressure, scored highly, though they did not have high stats in the easy groups.

  • Jay on April 20, 2016, 3:17 GMT

    Comparisons are inevitable - why, even between icons across different sports! Interestingly, TIME Magazine (USA) conducted a special report in 2012, where it compared Sachin Tendulkar's international cricket performance versus his nearest cricket rival Ricky Ponting (100-run innings). The gap of 29 centuries (by 41%) was the largest of 5 team sports in the study & concluded that Sachin led by a "margin wider than the gap between the two top scorers in other major sports"! They called it "Total Dominance"!! This comprehensive study also compared America's big team sports & their icons: football (Favre vs Marino, passing touchdowns, 21%), ice-hockey (Gretzky vs Howe, goals, 12%), basketball (Abdul-Jabbar vs Malone, points, 4%), baseball (Bonds vs Aaron, home runs, 1%). TIME proclaimed: "his ability to carry it for more than 22 years (then) while utterly dominating his sport makes a good case that Tendulkar is the world's greatest athlete"!! They declared him "The God of Big Things"!!!

  • Ashok on April 19, 2016, 19:26 GMT

    True, Apples vs. Oranges comparison is unfair. But in sports talent, skill & grace stand out irrespective of the Era. Stats. & averages are deceptive in rating these skills in Cricketer of different era. But their greatness is so far ahead of the rest that it stands out. In my opinion, Don Bradman undeniably was the greatest batsman ever & he was also a scoring machine. Gary Sobers was the greatest All rounder ever & majestically graceful Cricketer across any era. But when it comes to Bowling, it is far more complex to establish a legend of this sort because one has to pick the greatest in various types of spin & Pace, including a rightie or a lefty + their records on various qualities of pitches. The other difficulty is trying to judge the "Old Timers" due to absence of their videos. I never saw Bradman but his genius even Vs. bodyline bowling stood out. I saw Young & older Sobers in action - he was easily the greatest & the most graceful all rounder ever to grace the Cricket fields!

  • A on April 19, 2016, 9:02 GMT

    I think Jayasuriya changed entire complexion of ODIs. This resulted in change of mind across all formats. He should be hailed as modern cricket best.

  • Jay on April 19, 2016, 2:55 GMT

    One cannot measure a cricketer's greatness by stats or charts alone. There are intangibles. What sets Tendulkar apart from the rest of the best is his phenomenal Staying Power over 24 years. He is an Outlier! That's precisely the point: he has outlasted his rivals - not just with longevity or playing more matches - but by staying the long course and doing things out of the ordinary. He has pushed the limits of human performance. The physical endurance & mental toughness have enabled him to play through pain & injury; to bounce back from slumps & fatigue; to handle adversity & crises; and, yes, to face constant media scrutiny & expectations of a billion people. Add to it, his honesty, integrity & character. That's why TIME Magazine named Tendulkar among "The 100 Most Influential People in the World" in 2010. The citation: "Sports heroes such as Tendulkar, 37 (then), stand for national dignity in a way that perhaps only a postcolonial nation can understand. And feel grateful for"!!

  • grant on April 18, 2016, 9:00 GMT

    Its funny how some places tendulkar ahead of others just for his weight of runs and centuries. Well if you played 200 games and the rest played much less games than you, you should score more runs than them because you had more opportunity to score runs. In his career as a test batsman his average never crossed 59. It was 58 something. Kallis, sangakarra, dravid all had averages of 58 during their career. We want to compare tendulkar to all but in his own era there are batsmen just as good as he was and some even surpassed him. There have been great cricketers who changed the game and the 20/20 era changed the way test and odi cricket are played. The game of cricket are changing all the time and the skills of players are growing to keep up with the changing game. The Don played 52 games. He played in an ere when the rules favored the bowlers. let the old man be.