We are all aware of the age-old debate as to why there are so few top female chess players. Nature versus nurture takes on a fundamental role in this argument, with scientific studies supporting both sides of the issue. While I can’t see how anyone would believe there are innate or biological differences in the potential chess abilities of men and women, there are many among us in our (albeit largely male-dominated) sport who think differently.
But now there’s a different facet to gender biases in chess. Two scientists from Stockholm have just published
It has been well documented that males are generally riskier in strategic game behaviour than females, which is why male financial traders do better in ‘bull’ markets (when prices are going up) than in ‘bear’ markets – their risks pay off. So it’s no surprise that the authors confirmed that chess-playing men generally play riskier openings than women (although, have you ever seen Natalia Pogonina play the Dragon?!).
But the real finding is that men choose to play riskier, more aggressive openings if they are paired with a female opponent than if they were paired with a male opponent – even if the more aggressive opening is irrational and decreases their chances of winning!
Why? Is this because men subconsciously think women are weaker? Or perhaps because men are subconsciously – or bizarrely, in at least a few cases I know, consciously – trying to impress females with their risky chess openings?
Because that’s of course what a woman looks for in a man these days. Pawn sacrifices. (Insert pun of choice here.)
I have no idea. The authors themselves are pretty ambiguous about their hypothesised causality, stating:
“We believe that different outcomes across gender are not merely a question of deliberate discrimination on the part of men, but are at least to some extent due to deep-rooted mechanisms that surface in situations where competitors of the opposite sex meet.”
Of course, there are a number of ways to test the theories. You could ask players to report on whether they thought their opponents were over- or under-rated before each game, and match that up with the level of aggressiveness in the openings. To test for the effect of attraction, you could analyse a control set of the aggressiveness of openings chosen by homosexual males when playing other men – although, given the cultures of the most chess-populated countries, it’s unlikely you would get a large sample size of homosexual chess players willing to ‘out’ themselves. (I myself know of only half a dozen or so, somewhat surprising for such a male-dominated sport).
This aside, there are a number of fundamental flaws in the study. The choice of openings defined as ‘risky’ or not is poor. Even accounting for this, there’s no guarantee that the choice of opening is reflective of someone’s level of aggressiveness – many people play openings they were taught from childhood, or play them because they are easier to remember and require less updating. Furthermore, for many players, their opening repertoire is set, and they would thus not have the flexibility to be able to choose to play a different opening depending on the opponent.
Surely there is a means to check for ‘aggressive’ play overall in a game. After all, many computer programs these days have settings on them in which the computer is trained to play extra aggressively, even irrationally so. The formulas are already there.
These basic design flaws aside, the study is incredibly interesting. And, if nothing else, it gives me some excuse for my horridly poor performance rating against women. In my case it really is a mystery as to the root cause. I’ve lost enough games against women to know that they should certainly not be underestimated (given my experience, in fact, I probably subconsciously believe exactly the opposite). Neither do I believe that macho behaviour (on or off the board!) is a credible means of attraction in today’s modern society. So, given the rebuttal of these hypotheses combined with my continued weak chess performance against females, there really is only one logical explanation: