Greengard's ChessNinja.com

Jen Shahade: Clarity at Last

| Permalink | 12 comments

Usually, when looking at the games of the top players, it is hard for me to find instructive games for my students. However, Iíve found many in this year's Melody Amber tourney. The blindfold and rapid format of Melody cause the top flight Grandmasters to make more mistakes, and also offer less positional resistance, creating more instructive and lucid games. Because no ratings points are on the line, they often play lines they are less familiar with, or take wild risks. For those who are enthralled by the careers of the world elite, but often find their games impenetrable, I highly recommend looking through the Amber games.

The instructional value of the Melody games seems to be based on their lower objective quality. This adds support to the idea that studying your own games and those of your peers promotes practical results more than studying the top players games. Chinese National coach Liu Wenzhe, in his book The Chinese School of Chess writes : "Most players keep their eye on the games of the World Champion and other top players in the World. They believe that the more they study the games of famous players, the more their own skills will improve. In fact, this has no scientific basis. It is a fallacy reflecting the obsession with celebrities."

I would differ with Wenzhe in that I think studying top players' games can inspire their fans to study and play harder, even if they donít always know whatís going on. For chess players to all follow the same games and players also strengthens the chess culture. In that vein, I should mention that Vishy rules!

[2002 and 2004 US women's champion Jennifer Shahade of Brooklyn contributes monthly to the Black Belt newsletter, from which this is an excerpt.]


That's one of the main reasons I try to annotate as many reader games as possible in White Belt. When the player adds some of his own notes about what he was thinking during the game it's particularly good. Just about anyone with Fritz can tell you what. The why and why not are far more useful.

It's a very interesting concept. I think it might be almost a universal truth about chess learning. Namely, that if the games are too far above you, you can admire them from afar (and gain inspiration from them) but it's hard to pick up practical tips.

I always found that I learned more from watching/analyzing the games of players a level or two above myself than from the games of the very best players. Mistakes are instructive, and if the games I am looking at don't contain enough mistakes that I can understand after some effort, I am not learning much. Even heavily annotated, the games of the top players are still rather opaque to me. But I can usually understand the reasons why IM A. lost to IM B., particularly with the advantage of hindsight.

A related issue is the use of chess engines to analyze games. My personal belief is that if you want to get the most out of watching live games, you simply have to turn the chess engine off. You don't learn anything, really, from having Fritz or Shredder point out tactical flaws in human moves to you, other than that these flaws exist. I also think that, when analyzing your own games, running them through Fritz should be the very last step in the process. There is a lot to be learned by trying to do a deep analysis by yourself before turning on the machine, and then comparing the results.

I too seriously disagree with Wenzhes assertion. It is important to mention the aesthetic qualities that differentiate games of world class players from many of the rest of us. I find the displays of beauty through brilliant play far more inspiring than any other aspect of "top level" games. These displays appeal universally to serious and casual players and appear most frequently in the highest caliber games. For example, I have often shown Fischers queen sac against Byrne to relative beginners as an example of "beautiful" play. The moves that make the sac possible may be of little instructional value for average players but the sac itself is inspirational to anyone who understands how the pieces move. This would hold true regardless of the players "celebrity" status. The ability to visually recreate famous chess lore for educational and aesthetic purposes is a unique and vital part of our chess culture.


Wenzhe has serious problems supporting this argument that there is no scientific basis for the idea that studying the games of stronger players will improve one's own game. Even more difficult to support is the wild jump to the conclusion that this is simply reflective of some obsession with celebrities. First, chessplayers acquire a wide base of knowledge through conscious learning, but patterns are learned both consciously and unconsciously. Also, as Jen argues, there is a motivation factor which is not tangible and which social scientists often have a difficult time trying to guage. Thus, not only is there a difficult time proving that there is no scientific basis for belief in this approach, but there also may not even be a need for scientific data to prove it relevant. This argument is quite misplaced.



I think it's more specific than that. I would agree that a 1400-player does not learn more from a game between 2700's than one between 2400's. And probably much less than from an annotated game between 1500's. Patterns are big, but at a certain level you have to be able to relate to the game, the plans, and the mistakes. It could take hours to explain a GM combination to someone who is still having trouble with two or three-move combinations because of all the counting.

But that's a blind taste-test, if you will. Of course players are also fans and they like to follow the games of the elite. Since inspiration is at least half the battle, it's very important to make GM games accessible to amateurs, even if it's only the occasional combination or blunder.

"Thus, not only is there a difficult time proving that there is no scientific basis for belief in this approach, but there also may not even be a need for scientific data to prove it relevant."

How cool. Scientific conclusions without all the bothersome trouble of having to find proof. I wonder why scientists didn't come up with this technique years ago! Think of all the energy they would have saved!

Back on topic: I have heard a similar argument for studying the older masters, that their games are more lucid because their resistance to defeat was much less sophisticated.

For my money, even better than playing over an annotated game of players 300 points higher than you are is to actually play someone 300 points higher than you are. Wouldn't we see a boom in ability at all levels if chess programs ever got good at scaling their own power? You could play an endless number of games against someone at the optimal strength that would allow you to improve.

I prefer, as a WhiteBelt, to study my games and games from my peers. When I look a GM game I really don't understand much :(

I have just finnish "Carlsbad 1929" by Nimzovitsch. His notes to the games are great! I think everybody could benefit from his works because his is set out to teach from the beginng.
I have also recently discovered Edward Lasker, who also have teaching in mind when he annotes games. As dessert he often gives a little annecdote or two...

Thomas, I would like to study Nimzowitsch's Carlsbad 1929. I have always enjoyed studying the great Master's games. One thing is sure for me, that the great Nimzowitsch is my truly magnificient chess master. Studying his games is like enjoying a Mozart's Symphony. His art was true and sublime! Please let me know, if you can lend me for a while. Thanks

I'd really like to study the games of my peers, but I have trouble finding players that bad.

I will give the the book - or rather booklet - to you as a gift! if you send me your adress to morkore@gmail.com i will send it to you some time next week.

Thanks a lot! I am going to write you an email tomorrow. Have a good night.

Twitter Updates

    Follow me on Twitter



    About this Entry

    This page contains a single entry by Mig published on April 2, 2005 2:20 PM.

    Must Be April was the previous entry in this blog.

    Guardian Chess is the next entry in this blog.

    Find recent content on the main index or look in the archives to find all content.