x
Chess - Play & Learn

Chess.com

FREE - in Win Phone Store

# How to Identify and Fix Errors

Sep 23, 2016, 10:32 AM 0

Chess is all about errors, and they can be costly.  One small error in a tournament can cost a lot of prize money, not to mention rating points.  I'm serious.  Suppose you are playing the last round of a quad with a perfect score. You are a pawn up, and then botch it with one simple error and the game slips into a draw. That half point could well be worth 30 rating points in the USCF system.

So, when I make an error I'd like to figure out why I did it.  I mean I want to identify the cause, and then eliminate it.  Besides going over my own games, and looking at the mistakes, what else could I do?  I thought long and hard about it and came up with this idea.

Why not work tactic problems, preferably difficult ones, and every time I fail one, write down what I was thinking, and compare that to the correct line.  Why not? Frankly, I'm balking at the thought, because it is a lot of work.  Let's see how much of this revealing and humbling toil I can handle.

Here is the first problem that I failed.   I approach problems like this as if I'm playing them in a game, and I can afford to spend several minutes calculating.  Here I burned a little over 12 minutes before playing the wrong idea and failing the problem.  The analysis between main line moves contains much of what I was thinking.

Diagnosis

OK.  The source of my error is pretty clear if you read the notes.  This happens a lot more than it should, too. I failed this problem because I spent too much time and effort on the wrong idea.  I need to search wider before I search deeper in positions like this.  A simple one ply search  of all legal moves would have solved this one quickly.

Treatment

I have to develop the discipline to make the one ply search.  One full ply on all first move options, and one full ply on all possible opponent replies to candidate moves.  Properly implemented, with the mentality of every move of every game, this discipline could easily be worth 200 rating points.

After that failure, I got religion, and applied extreme discipline of the one-ply search to solving tactic problems.  I reeled of a streak of about 10 correct ones, not counting one position that was ambiguous.  Then I got to this one.  After about 10 minutes I played a move that was OK, but nowhere near winning.

The main line is the correct line.  My thoughts and calculations make up the rest of it.  What went wrong here?

Diagnosis

I went too deep into the wrong idea, and not deep enough into the right candidate. I dug a post hole, when I should have just dug a shallow grave.

Treatment

Continue with the one-ply mentality, but try to be more broad in its application.  This should improve with practice.

Remarks

To be fair to myself, I might well have found the queen trap in a game, because I might have seen the idea during my opponent's move or otherwise as the game proceeded. Also, 1.f6 was fairly strong, and lots of defenders would have cracked in those lines, so I don't feel too bad about failing that problem on practical grounds.  It does seem, however, there is a pattern here.  of "too deep" and "not wide enough".

I maintained my one-ply discipline mentality and went on to solve several more problems, then I got this one.   I found the right idea, and then followed it up with a game ending blunder.  My thoughts are in the notes.

What happened here?  I'll tell you what happened.  I was so enamored with the brilliance of offering my Queen on e5 that I simply did not want to see that b4 was hanging.   Fatigue was probably a factor, too, but let's not use that as an excuse.

Diagnosis

I allowed my feelings to trump calculation and judgment.

Treatment

Be on the lookout for emotions while a game is in progress.  Elation and fear can cloud judgment.  Get out of whatever mood you are in and replace it with clinical dispassion. Learn how to do this.

Wow.  This is a lot of work.  I need to work on this one-ply discipline mentality until it becomes habit.  That might take a while.

Online Now