Sunday, June 03, 2007

"Ready to Improve?"

In today's New York Times chess column, Dylan Loeb McClain interviews GM John Fedorowicz and IM Danny Kopec, who offer some useful tips for those looking to improve at chess this summer. I especially like Fedorowicz's advice, which includes playing a lot of games, choosing and studying an opening repertoire that suits your style, and reading endgame books and game collections of the greats. Interestingly, he does not think studying books on tactics (or, presumably, CT-ART) as important as playing, and Kopec does not mention tactical study either.

Fedorowicz's disregard for the "study tactics, tactics, and more tactics" mantra of chess improvement gurus reminded me of a recent conversation I had with two-time NJ Open champ Tommy Bartell on the subject. I was trying to think of some books on tactics to recommend to a developing player at the club, and out of curiosity turned to Tommy, who was seated nearby, and asked what books he'd recommend. I was surprised when the only ones he could think of were the "1001..." series by Fred Reinfeld. "Well, what did you look at when you were a kid?" I asked. "I never really read books on tactics," he said, adding "I just learned by playing."

Hearing this young, 2400+ FM say he thought playing was more vital to his understanding of tactics than book study, I felt a little surprised. So when I read Fedorowicz saying more or less the same thing, I began to ask some more questions. Do strong players invest any time in tactical training or do they just play a lot? And if they just play a lot, is that better than study? And if it is better than study, why might that be?

One idea comes to mind watching a recent video at the excellent Chess Vibes blog that show Jonathan Rowson discussing his book Chess for Zebras, where, for instance, he discusses how the traditional way of determining the relative value of the pieces (by assigning point values, beginning with one point for a pawn) is both wrong and not how GMs really think.

Why don't GMs think that way? Maybe because they learned these things intuitively by playing a lot of games rather than "by the book." And, as Rowson argues in Chess for Zebras, one of the things that keeps developing players from improving often is the settled beliefs and habits of mind they have developed -- often from reading chess books.... So maybe the best advice is to leave most of those books alone -- except, perhaps, for the game collections of the greats, where you are likely to pick up a thing or two about how to think without being subjected to a lot of theory.... I'm still puzzling over this one, but in many ways it does fit with my view that there is chess knowledge and chess practice and the two are often unrelated. Perhaps the secret to success at chess is to just get plenty of practice.


Grandpatzer said...

I can tell you that NM Alex Betaneli was the same way. I think you can explain it by realizing that, whatever area of chess you study, tactics and calculation of variations are involved. So, if you're spending your time going over master games and really, really thinking about them--going over the annotations and variations of great players, questioning them, trying to find refutations--you're automatically training tactics, calculation and visualization at the same time. They probably also spend a lot of time analyzing endgames, and I personally feel that's another great tool for developing these skills. So, instead of just studying tactical problems, they're multitasking and addressing multiple aspects of their game.

Blue Devil Knight said...

Very interesting. My coach wants me to play a lot more (though he also likes me doing the circles with tactical problems).

I'm thinking of training for tactics solely by playing lots of blitz (e.g., 10 games a day) followed by blundercheck in Fritz to find the tactics missed by both sides. This would train one up on real-life tactics (there are a lot of other possible benefits of such a meta-blitz: opening repertoire development obviously).

Blue Devil Knight said...

Interestingly, I have noticed that my coach (an IM) knows very little about anything except games. When I ask him for books on X, he will hesitantly suggest some book that was popular like 20 years ago, and has no idea what is happening now. He just plays, plays some more, and analyzes games. Indeed, even as a coach he prefers to go over my games than to do any kind of canned lesson (which he is good at when he has to do them).

Every time we meet he says, "I think you just need to play more. Get more experience."

I have been, and I think it is helping.

Zeeple said...

bdk: what is "blundercheck in Fritz"?

Blue Devil Knight said...

Zeeple: you can have Fritz quickly analyze a game for 'blunders' (you define how far ahead it looks), and it will draw an evaluation graph (for each move on the x-axis, the evaluation on the y-axis is drawn as a bar, with +1 meaning white is ahead by a pawn, -1 black is ahead by a pawn). It is a great way to quickly find tactical mistakes.

There is an article about using Fritz full analysis and blundercheck here.

Anonymous said...

Hey check out this hilarious video about Chess by Mail. It's family friendly and may get them interestes in chess.