Saturday, December 4, 2010

Test and Improve Your Chess

I bought this slim (128 pages) volume by Lev Alburt a couple of years ago. I’m not familiar with Alburt’s other works but know he has written many books revealing “once secret Soviet training methods.” That’s nothing more than sales hype because there never was any such thing. GM Alex Yermolinsky said so and many years ago, I don’t remember if it was Bruce Panolfini or Jeremy Silman, writing in Chess Life, told of a visit to the Soviet Union where they visited a chess school. They were disappointed because it turned out the training methods they used were no different than those used everyplace else.

So why did the Soviets have so many masters and GMs? It was a function of numbers. Ratings follow the bell curve and only a very small percentage of players will be at the top. Back when I started playing chess, the USCF had about 5000-6000 members and there were only about 50 players in the whole country rated over 2200. Compare that to the Soviet Union which had a hundred thousand or more players…they had a LOT of masters.

The same was true of minority players, blacks and women. In my home state there were 3 or 4 women active in chess and they were rated 1200-1500. There was one Black player rated about 1750 and one Chinese player rated around 2000. With the Fischer Boom of the 1970’s minorities started playing chess and with the increase in numbers, their ratings began to climb so that today their ratings are comparable to the general population.

Back to Alburt’s book…he starts out talking about evaluation of positions and states, in passing, that pattern recognition is a helpful aid. He states, “Without a recognizable pattern to use as a frame of reference, the average competitor is forced to rely upon intuition and subjective reasoning.” No argument there, but then he begins to prattle on about something no reader of his is going to comprehend. He discusses the meaning of various symbols used to show the evaluation of positions and points out there is no standard. For example, does “=” mean dead draw, practical equality, theoretical equality or dynamic equilibrium? He points out that a major flaw in computer evaluations is that they assign precise values to positions formulated on values given them by fallible humans. To each element a number value is given and computers then synthesize everything to come up with an exact number and humans do not think that way...so far, OK. He goes off the deep end by recommending what he calls “The System of Predicted Results.”

The premise is that any position can be assigned a numerical value based on the estimated number of points that White is predicted to score out of ten games played from that position between two GM’s of equal strength. For example, a dead draw is rated at 5.0, an absolute win is 10.0 and the starting position is rated at 5.5. Another example: the “=” sign could mean an evenly balanced position that may result in +1 -1 =8 or it could mean dynamically equal where it would be rated +4 -4 =2. He points out positions have little or no statistical background to draw upon so you will be calling on your powers of intuition and abstract reasoning which you can improve through study and play.

Maybe you can accomplish it through study and play if you are on your way to becoming a GM. For the average player this is pure nonsense. Seriously, does Alburt really believe the average player will EVER be able to distinguish between a position that is +1 -1 =8 or +4 -4 =2 or +6 -2 =2, or whatever? The whole point of this is that you use this information to construct a graph of the game which will tell you how the game evaluations swung back and forth…or where the game went South on you. This tells you the critical point in the game so you know where to look for errors. This part isn’t too bad, but it’s all beyond, way beyond, the ability of his intended audience. Actually the graphing idea isn’t bad, but the average player is going to have to let Fritz do it…and all that information is interesting to know, I suppose, but useless for anybody below GM. That is, except for the brief explanation about using graphs. Just let Fritz construct them for you.

He has a section on “specialization” where he talks about specializing in a position until you know it well. In this case he discusses the Isolated QP. Then there are 4 chapters where he analyzes games and positions and asks you questions about critical points. Of the games, which are well annotated, 5 are Alekhine Defenses and the final chapter is on the Benko Gambit. Throughout the book he keeps referring to his assessment of the positions using his System of Predicted Results.

You can buy it on Amazon for $1.87, but even at that, I think it's overpriced.

No comments:

Post a Comment