How to Analyze Your Games

      Do not just plug a game into an engine to see where you went wrong; you won’t learn anything. First, as soon as possible after the game write down quick notes. Be especially careful to make a note of positions where you had to make a difficult decision.
      When analyzing an opening write down your thoughts on the line played and then refer to your opening database or books to see if it’s “book” and, if it is, was the line considered good or bad for one side. You need to review a few annotated master games on the variation played so you’ll get a better understanding of the ideas and correct handling of the line played.
      After the game gets out of the book, you’re in the middlegame and you should write down why you decided to play the move you selected.  Don’t be surprised if there are times when you don’t know why you played a move.  For whatever reason, it seems that many time amateurs just play a move.  It’s the kind of thing that drives Jeremy Silman nuts and I once witnessed an IM going over the game with an average player and he kept asking, “Why did you play that?”  You’d be surprised how many times the answer was a blank stare and a mumbled, ‘I’m not sure.”  Be sure to list tactical ideas and possible sacrifices you considered. If you had a plan, indicate what it was.
      In the ending, if you reached one, try to come up with a winning plan or drawing plan as required.
      Only now are you ready to plug the game into your engine.  Now, this is important: analyze with more than one engine. Why?  Because engines evaluate positions using an algorithm and different engines use different values in their algorithms. One engine might value space more highly than while another engine will place a higher value pawn structure. The result is likely to be different results from different engines.  This is valuable information because the engines will be pointing out various lines that can then be checked out. 
      Pay special attention to the positions where you couldn’t find a good move.  Did you just blunder due to missing the obvious or were you just plain careless? Or was it because you did not understand the position or didn’t know how to play the ending?  The answers will tell you what you need to study.  If you are like me, more than likely you’ll find the problem to be more than just being a tactical dunce.

      This really has little to do with analyzing with engines, but even if you’re booked up on openings do you know how to play the position when your opponent plays something you haven’t seen? This can be a problem even in CC.  Despite my opening books on the Torre Attack and my 3 million game db a recent CC opponent got me out of the books after 3 moves! 1.d4 d5 2.Nf3 Nf6 3.Bg5 h6.  What?!  Black usually plays …e6, …Ne4, …c6, etc.  The move …h6 is usually played after …e6 but it’s not bad here.  I never saw his move in this position before and couldn't find it in my db; we are on our own.  White can either play 4.Bh4 or 4.Bf4, but must realize the two moves lead to completely different types of games.
      In another game in which I was Black against a CC opponent rated nearly 2500, the opening went:    1.e4 c5 2.Nf3 d6 3.Bb5+ Bd7 4.Bxd7+ Qxd7 5.c4 Nc6 6.Nc3 g6 7.d4 Bg7 A very interesting novelty played in Shirov-Kasparov, Erevan Olympiad, 1996. Now White has a hard choice between 8.Be3, which is completely unpretentious because the B normally aims for g5 in this system and the move played. 8.d5 Bxc3+ 9.bxc3 Na5 10.Nd2

     
      Shirov says that the text looks rather slow, and that for the moment is it not necessary to defend the P but gave no further analysis because he played 10.0–0. We were on our own after only 10 moves.
      Consider the following position:

       This position arose in a game played in the finals of the 1972 US Open CC Championship. My opponent was a former competitor in the US Closed Championship and at one time was one of the top rated players in the US.

       It was my move and the opening pamphlet I was using gave the moves 18.f5 Nc5 which leads to a position with equal chances, so that’s what I played without even thinking about it. Looking at the position while awaiting Black’s reply, I asked myself what would happen if he played 18…Ne5 instead. That move would centralize the N and gain time by attacking the Q. Then when the Q moves off the f-file, there is no attack along it after fxe6. If the Q goes to f4 the N is covering f7 and the Q is exposed to …Rf8 after opening the f-file and White has no attack.
       All I could do was wait and hope Black played the “book” move; he didn’t. He played 18…Ne5 and I tried 19.Qg3 and lost quickly. That’s when he asked me if I was using that particular booklet and said, “There’s a mistake in it.” This was a lesson I never forgot...don't blindly trust opening books.