Sample Size And Chess Success

Taking a look at the Anand – Gelfand World Championship website I see that the poll makes Vishwanathan Anand a massive favorite.

Yet something I agree with Susan Polgar about is that it might be far tougher than most people imagine. Besides the arguments about Boris Gelfand’s determination, any edge that Anand may have is going to be randomized by the relatively small number of games.

It’s well known to those in the gambling fraternity that those who have an edge need to sit at the table for as long as possible in order to maximize their chances of winning. This is useful for chess players to know as a higher rated player should therefore try to grind his opponents down over time (increasing the sample size of moves played) whilst the lower rated one should try for in short sharp battles. To some extent this may be mitigated by stylistic considerations but these are usually overrated.

Strangely we often get the opposite effect, higher rated players trying to wipe their opponents off the board after which the underdog just defends and tries to prolong the game. Psychological factors are in play in such cases but they lead people into doing the opposite of what they should.