Random Posts

Tuesday, August 31, 2010

A Little Known Improvement Secret

Kenneth R. Smith (Smith-Morra) offered this advice years ago: Play over hundreds of unannotated master games while trying to guess the next move. He said you should spend no more than 5-10 minutes per game and advised you are going after quantity, not quality because the quality would eventually come. The idea was that you were gaining skill in pattern recognition.

Chess for Dummies by James Eade has something to say on pattern recognition.

Read what Jeremy Silman has to say HERE

The Psychology of Chess Skill by J. Corey Butler, PhD

Chess Intuition on Science Blog

Does it work? Sometime in the late 1960’s a veteran local master asked me to edit his self-published tournament book of master games, so I decided to try Smith’s method and after editing the typewritten games in the book, entered a small weekend event where I finished 4-1. I told the guy about what I did and how I thought it really helped. His reply was a simple, “Of course.”

The next step was to order an Informant (cost $5.00 in those days!) containing 500-600 games and start plowing through them and playing in an occasional weekender. 300 or 400 Informant games later it was off to a big event in Chicago (my new rating hadn’t been published, so I was still mid-1600’s). In round 1 I defeated an 1800 and then won 3 more games against 1800-2000 rated players. Finally in the last round I lost to the Chicago legend, Morris Giles.

I think this secret is so little known because in my experience most strong players have either been unwilling or unable to say exactly how they actually improved or exactly what they did that brought inprovement about. Then of course the guys like Silman who do know are not willing to expound on it. Why? If all you have to do is get hundreds of games free off the Internet, have a sheet of paper or card to cover moves and maybe an Excel spreadsheet to track your progress, who's going to buy a $25 or $30 book?
When my new rating was finally published it was nearly 2100. So,yes, I think it works. Of course I had always studied the usual stuff, without much improvement I might add, so the background was there. To make things more interesting I kept track of the percentage of correct guesses. I can no longer remember the exact percentage but think it started out at maybe 20-25% and ended around 70%.

This is method is a well kept little secret that most average players are unaware of. For some strange reason when I tell it to them, they ignore it or do as one know-it-all 1500 told me, “It won’t work.” I think it is this pattern recognition that separates the good players from the rest of us. It’s also what makes writing good instructional chess books so difficult. Most strong players simply are unaware of how they choose a move.

At one US Championship I was watching Super GM Tony Miles (he was living in the US at the time) analyzing with an IM who kept suggesting moves and Miles kept telling him, “It’s no good.” When asked why all he could say was, “It just isn’t.” I don’t think Miles was being condescending; he just knew it wasn’t a good move, but he couldn’t explain why. Intuition, judgment, pattern recognition…call it whatever you will, he just knew.


  1. Interesting post. I remember when I had some free time and spent it going through the book "500 Master Games of Chess". Just reading through the games and playing the old DOS game USCF Chess I raised my estimated rating on that by a few hundred points.


  2. Silman tried to explain that he played over thousands of master games to get better in an article on chess.com a few years ago and a bunch of c and d players told him that he didn't know what he was talking about in the comments. If people refuse to hear the truth sell them what they want.

  3. If people refuse to hear the truth sell them what they want...that's priceless!