Tuesday, August 07, 2007

Back to Polgars middlegame brick

Right now I'm doing exercises in order to get better at complex middlegame play. I stopped using Renko's CD ICT II since even the exercises at masterlevel are too easy. So I'm back by 28 exercises from Polgar's middlegame brick. I have done these before about 6 months ago, using 4-10 hours per problem. I want to check what I remember from the analysis and if I can create narratives to get deeper in such positions.

With narratives it is easier to divide the position in different parts. In this old post I had trouble with bookkeeping of all the possible trades in the diagram. What I used to do was something like:
  • White to move Nxe8
  • Add 5 points to white
  • Nxc2
  • Add 9 points to black
  • White can do several things now, take the bishop, add 3 points or play Nc7 attacking the rook 5 points or he can play Bg5 attacking the queen 9 points.
  • Eh, where was I?
Thanks to the narratives I all of a sudden saw that I can just count everthing that black can take in a row and what white can take in a row.
Black: Nxc2 (9 points) Nxa1 (5 points) and Bxc3 (3 points) together 17 points.

White: Nxe8 (5 points) Nxg7 (3 points) together 8 points

Or after playing Bg5 first.
White: Bxe7 (9 points) Nxe8 (5 points) together 14 points

This is a much simpler way to look at counting problems than switching from side to side every move, in the mean time adding up points and devising new moves.
I'm convinced that the chessmasters live in a much simpler world than I because I'm confusing myself all the time. Narratives help to dissect the possition in its constituent parts and to get a definite conclusion about each part. Thus diminishing the load of the short term memory.

For your convenience I repeated the diagram below.

White to move, black to win.


  1. That's a good method, but does it really depend on using a narrative at all?

    For instance, I've adapted to the circles partly by using a similar method ever since I had an awful time with counting problems. Sometimes I still get myself so confused I just have to count the remaining material on the board after a sequence of trades, and compare that the the amount on the board before the trades.

    However, perhaps the narratives help to find such heuristics faster.

  2. "Theoretical endings form only 5% of the total area of endgame study, and the study of the 95% endgame strategy that precedes the theoretical ending is much more profitable by far." -- Interesting. I've never read an Endgame Strategy book, but don't endgame texts often intermingle theoretical positions, tactics, and strategy? For example, any rook endgame book will tell you rooks belong behind passed pawns, don't swap down into a pawn endgame until you've calculated the result, and when you're ahead trade pieces instead of pawns. Of course, the more positions you study, the better you learn how to apply these 'sayings' in your actual games.

  3. Blue,
    while I was constructing narratives, the position started to look simpler. Since it looked simpler, I could see that this new way of counting was simpler and would give clear and correct results. If you look at the comments of Fierabras on that old post, you see that he doesn't see what the problem is. He is about 200 points higher rated than I am. He is right, of course, since there is no problem. But if there is confusion, you don't see the simplicity. I wasn't sure that this new counting method would yield correct results. Now it is obvious that that is the case. But first the confusion had to be expelled. As said, narratives helped me. Or maybe it was the fact that I already have studied this position for 10 hours before.:)

  4. LF,
    but don't endgame texts often intermingle theoretical positions, tactics, and strategy? For example, any rook endgame book will tell you rooks belong behind passed pawns

    In a way they do. But if you don't have a question, you can't get an answer. Even if you have read it 100 times, you don't realize what the importance is.

    A maxim like "put your rooks behind your pawn" triggered something in me like "behind? From which point of view? Seen from me it must be on the promotion square. Seen from the direction my pawn has to go it must be the first rank". That is what hapeens if you get a maxim without the idea behind it. Endgame strategy gives you a clear picture of the relationship between the maxims and their goals. When you have the overview, there is no longer the need for a maxim since you can reflect on it yourself.

    Endgame strategy is not very difficult once you see what it is all about. As said, the cover of Hansen's book shows it all. See
    But it took me a lot of time to get there.