Men, women and chess skill: The whole truth (1)

German grandmaster and chess24 author Ilja Zaragatski, who has a degree in Economics and Sociology as well as considerable expertise in (evolutionary) psychology, takes a scientific but light-hearted look at a delicate but intriguing issue: why are most strong chess players male? If that's something you've always wondered about you’ll definitely enjoy this little journey!

This article is translated from the German original.

Part 1: The story so far

The topic at hand is old, but nonetheless fascinating and heavily-debated. Everyone has an opinion on it, but no-one really seems to know. For starters, then, a quick reality check: men are simply better chess players than women, and there’s no getting away from that. Or is there?!

No, there isn't. The numbers are clear on this one: on the current rating list (March 2014) only Judit Polgar, universally praised as an exceptional talent, has been able to establish herself in the world Top 100. China’s Ju Wenjun, for example, is the number ten in the female world rankings but can only make it up to 721 when both sexes are taken into account. FIDE, the World Chess Federation, currently lists 1440 grandmasters, of whom only 31 are members of the gentle sex.

GM Ilja Zaragatski at the board | photo: Georgios Souleidis

Sure, when it comes to physical sports like football or athletics it's understandable. Men have a physical advantage - they’re stronger, tougher, more athletic and have greater endurance than women. In order to move felt-padded wooden pieces across a chequered board, however, you need rudimentary muscle power at best. So why should women not be as capable as (or - when playing against men and the weather outside is really warm - even more capable than) men!?

In the past people have resorted to many different explanations to resolve this enigma: historical, social, cultural and physiological. A brief overview for interested readers is offered by the self-proclaimed chess queen herself, Alexandra Kosteniuk, on her blog. The prolific chess blogger and wife of GM Erwin l'Ami, Alina l'Ami (herself a graduate in Psychology) also has some valid reflections on the same topic.

In many cases, however – and if you think about it, it’s hardly surprising – authors appear not entirely unbiased. They have a tendency to advocate their own ideologies or follow their own agendas, even when that mainly consists in supporting or justifying their personal world views. While men like to refer to the biological factor in order to support or legitimise their apparent superiority, feminists invoke gender stereotypes and a misogynistic culture and upbringing. Other lay theories range from "women are hormonally unbalanced, emotionally unstable...

...and have to deal with the time of the month" to "women allow themselves to be distracted by other things more quickly and want to have a family", "women fall in love, women have children, breastfeed and then sleep badly. Women are less analytic, they get into time trouble and they lose the thread" (Elisabeth Pähtz) and "...the imperfections of the female psyche. No woman can endure a prolonged fight. They're fighting against the habits of centuries and centuries, since the world began." (Garry Kasparov)

The difference between the sexes isn’t confined to their respective playing strengths. Interestingly, it’s also been found that men and women let their way of playing be influenced by the mere sex of their opponent. Men tend to choose riskier strategies when facing women rather than men, while women appear simply to play worse when sitting across from a member of the opposite sex. 

In 2009 a study by Bosnian Professor of Psychology and FIDE Master Merim Bilalic was published that promised to solve the puzzle and cure the headache once and for all. Bilalic teaches at the University of Tübingen and sometimes still competes in the regional chess league. He and his colleagues found that as much as 96% of the difference in playing ability can be accounted for by the mere higher number of men playing chess.

The authors concluded that as soon as the participation rate of men and women (the proportion of female chess players) was statistically accounted for there would hardly be any room for biological, environmental, cultural or other explanations. Male superiority at the highest levels in chess could therefore be explained in purely statistical terms: since many more men play chess there’s a much higher range of chess skill among them, with the result that more individuals are able to make it to the top. If true, this simple statistical fact has been overlooked by experts just as much as by lay people and the players themselves.

Christopher Chabris | photo: Wikipedia

A similar conclusion was reached in 2006 by the American psychologist Christopher Chabris (rated 2245) and the American statistician Mark Glickman (inventor of the Glicko system, an alternative to the well-known Elo rating system). The authors also found that the higher number of men at the highest levels of chess can be explained by the higher numbers of boys entering chess at the lowest levels. In their analysis of database games they established that girls and boys improved and dropped out of chess at similar rates, while boys entered chess competitions in higher numbers and at a higher level. On the other hand, girls' initial ratings were no lower than those of boys in environments where at least half of all beginners were female.

Cool. So everything's clear, the debate is settled and we can all go home, right?! Not quite! Unfortunately/luckily things aren't quite so trivial.

For example, Bilalic et al.'s claims have been challenged and criticised as premature on the grounds that they were based on an inappropriate statistical method. According to epidemiologist Michael Knapp from Bonn only 41 to 71% (mean value 66.9%) of the actual rating difference between men and women can be explained by their differing participation rates. Moreover, while Bilalic and his colleagues base their analysis on a data sample of the Top 100 German men and women, in the year of the article's publication FIDE Executive Officer David Jarrett says women accounted for approximately 10% of their estimated one million members, but only 7.6% of the 100,456 members with a FIDE rating, and just 2% of the world's Top 1000 players. Crunching these numbers you instead arrive at an effect in the region of Knapp's proposed 66.9%, still leaving a third of the effect open to alternative explanations.

In fact, in their original paper Bilalic et al. already concede that the participation rates used as a benchmark in their study might actually underestimate the real proportion of female chess players. So a process of self-selection based on innate biological differences in intellectual abilities could conceivably have been at work, meaning that the effects of this self-selection would already be observable in the rating lists they looked at. Then women might therefore be somewhat inferior to men in those intellectual abilities that are required to play chess successfully. Then this inherited competitive disadvantage would lead to women dropping out of chess at higher rates compared to the more prolific men. Then according to this whole train of thought the lower number of female chess players would be a consequence of their higher dropout rates, which in turn would be caused by their innate lack of the cognitive skills necessary for success in chess. (Of course not talking individually but at a population level - statements like this would always apply only on average, while single cases might very well differ...) The differing participation rates would provide an intermediate-level explanation for the discrepancy at the top (as a so-called intermittent variable), while the differing participation rates themselves would be a product of innate differences in the mental predisposition of the sexes (explanatory variable).

However, it still remains unclear which intellectual faculties where men and women supposedly differ would correlate to chess talent.

Everybody either confused or asleep? Excellent!

"No, I get bright pink!" Women's chess | image: toonpool.com

In any case, we can pretty safely assume that the enormous male over-representation in chess leads to most successful chess players being men. We can all live with that - it's politically correct and it sounds reasonable as well. Whether almost all of the observable variance can be accounted for by pure statistics can be left undecided for the moment. Apparently, as so often in life, a variety of factors interact and end up leading to varying success for male and female chess players. Statistics, however, are obviously a major component of the equation!

And yet... this explanation doesn't satisfy me. Curious guy that I am, I really like to ask why - and I like to keep on doing so until it plainly makes no more sense. Here, however, it still seems to make a great deal of sense. So: Why in the world do individuals of the species homo sapiens with two X chromosomes end up in chess clubs much less often, stick around there for much less time and subsequently reach the highest levels much more rarely than their male colleagues? Why is it so much less desirable for women to seriously engage in chess, shoot for a rating above 3000 and ultimately knock Magnus off his throne?

You'll find the definite, irrefutable and seminal answer - which will change everything... in the second part of our little saga!

Keep reading: Men, women and chess skill: The whole truth (2)