The Chess Mind

In presenting the game Andersson-Ivanov in yesterday's ChessBase show, I benefited from Jacob Aagaard's notes to the game in his 2004 Everyman Press book entitled Excelling at Technical Chess. As helpful as his notes were, I found the following post-game comment even more valuable:

It is one of the most important points in technical chess that an advantage does not have to be decisive in order to win. Obviously it is better to have a winning advantage than a clear advantage, but as it is harder to defend in the endgame than to play for a win, a clear or even a tiny advantage often has a tendency to increase over time and prove sufficient to win the game.

It is important to understand this as a defender as well. I know many people would have though that there was nothing much wrong with 17...Rfd8, 21...f5 and 40...f6 in this game, and that 46...Bg1? was entirely to blame for Black's defeat. [DM: Or that 46...Bg1 was innocent, and that 50...Be3 rather than 50...Bd4 was the culprit - see last night's show in the Playchess archives for details.] But this would be missing one of the simplest truths about chess - that chess is a game. The defender has to find the best defence all the time, and if you go through the annotations, you will see that his problems are multiplying as the game proceeds. At move 17 the improvement is one half move long, while at move 46 the proof that Black was still not lost has increased to half a page, and most of the lines are drawn by the smallest possible margin of a single tempo...[pp. 24-25].

I find this commentary valuable for at least two reasons. The first is the reason Aagaard himself gives, concerning the nature of playing technical positions (both for offense and defense). The second reason, and the one that immediately captured my attention when reading the text, pertained to self-improvement. In my experience, many amateurs seem to think they've explained a loss when they've detected one of their errors - perhaps a blunder at the end, or a mistake in the opening - and promptly declare the analysis complete. Maybe they made other mistakes, but "if only" they hadn't made the one error in particular, then everything would have been fine.

Maybe this sort of "Ockhamism" is psychologically useful when one is still in the midst of a tournament, but as a strategy for detecting one's weaknesses and improving it's a dismal failure. The other errors reveal something too, and it might be that they represent a problem that's more likely to recur and cause problems in the long run. Further, there can be an integral link between the earlier errors and the one that officially tips the evaluation from bad to lost. They might all be part of the same general plan or motivated by the same (mis-) understanding or evaluation of the position, in which case the last error really isn't independent of what came before. Also, as Aagaard notes, the earlier errors serve at the very least to reduce the margin of error, so their contribution should not be dismissed.

So when you analyze your games, especially your losses, look for and reflect on all the mistakes. Some are more serious than others, but each has its own story to tell. It might not be much fun at first, but it's a lot better to fix the errors at home than to repeat them in a tournament.