Tuesday, June 07, 2005

A game in a game

I'm starting to become enthousiastic about pawn endings.
But I'm feeling a little weird.
When I talk to my chessfriends about it then I get a reaction like "o yes, rook endings, very important stuff. You have to learn that."

Rook endings? I don't understand pawn endings at all, how can I begin with rook endings?
It seems that a lot of people look down on pawn endings.
Well, maybe I am stupid, but I find it to be very complex matter. OK, I don't have feeling for it, but still I wonder how much people could solve the problems with for example Kpp against Kp well in all cases.
I worked my way for the first time through Euwes pawn ending chapter.
I have now a global understanding of opposition, triangulation, Reti manoeuvre, zugzwang, rule of the square, passed pawn etc. .
But that is something totally different than having the skills and technique to play the positions correct.
Because sometimes opposition doesn't matter.
Sometimes there are hidden resources.
Sometimes etc. . .

I get the idea I can work miracles if I master these skills.
Only I have no idea how often this can be applied.
Maybe I am over enthousiastic.
Maybe in 98% of the games end play is just straight forward and in only 2% of the games these skills are paying off.
But I can hardly believe that.

Here is a diagram from a game of I. Rogers (white) - A. Shirov (black) Groningen 1990.
Shirov has just traded everything off because he thought this position was a clear draw.

White to move and win.

After the first next move of white Shirov resigned.
If even he makes such mistakes in such seemingly simple positions, then a lot of half points could possibly be gained.

Why did I never came up with the idea to study endings?
I have for about a meter books on openings. I even worked my way thru most of them.
But I have only two books on endings.
And I haven't read them both yet. The Christopher said that most people find it boring. He is probably right.
I find it to be fascinating.
I have developed the plan to study pawn endings the rest of this year.
I think that the problem is that you have to study for about six months and that you then just have mastered only the basics.
Probably few points are going to be gained by just mastering the basics. So the pay off is going to be in the long term. When you master the different types of endings AND know how to trade pieces down towards a favourable ending.
You actually have to learn a game in a game.

Luckily I found that Jeremy Silman said:

A solid knowledge of pawn endings is essential for those wishing to understand endgames in general (Rook Endgames, Minor Piece endgames, etc.). In fact, if a player wants to seriously study the endgame, pawn endings are the place to start because many endgame positions are evaluated on the basis of the underlying pawn ending that might be reached.

So I assume I am on the right track. I'm looking forward to see how the results of my method compare to the system CD is going to follow.


  1. osmitz has a pgn of Pandolfini endgame problems. You might find it helpful.

  2. I agree with everything you said. But how many pawn endings did you get in the last 20 games you played? I got one.

  3. Mouse,
    If that resuls in a half point extra per 20 games, it would already give a gain in ratingpoints.

    The last 8 years I have had about 10 - 15 endgames in total(OTB). Not to mention PAWN endings. But I avoid them. I agree with a draw because I'm afraid to loose, or I start an unmotivated attack which leads to zero.

    But I know a man who have a Knight versus Bishop ending about 90% of the time. Ok, he don't know how to play it, so it doesn't help him, but still. Since 2 years he has a girlfriend and at the last Corustournament I saw she had often the same ending too. So it happens to be contagious.

    Of course I have to learn to steer things into MyWaters.
    Which is far from easy, because on FICS I forget consequently to open with 1.d4 and play my gambits before I know it.
    And most of the time I become so involved in the middlegame that I forget to trade the pieces off.
    But that's a matter of time.

  4. Tempo, did you solve tons of tactics puzzles just to end with trading pieces to pawn endings? I do not want to be a kill-joy, but I have to ask this question. I agree that training pawn endings gives better empty board vision which is good anyway. But I fear that 90+ percent of all games are won or lost with a fairly crowded board. I can speak only for myself, and for me it seems better to spend most of my time for situations that are most frequent, namely to eliminate the blunders that are most frequent.

  5. Mouse,
    Studying tactics has long been a goal in itself. I even had my whole openingsystem adapted to this one goal: playing tactics.

    The last tourney revealed that there are two serious flaws in my play, middlegame (positional) and endings.
    The fact that I created a mate in eleven reveals that my tactical play is reasonable now. So now I have to adress both of the middlegame and the endings.

    Which one do I have to do first?
    Both studies will take a few years, I presume. I don't want to divide my powers. Since middlegame play is highly influenced by the endings, the logic says that "the end is a good place to start".

    If I want to learn endings I have to take the means to do so. That means playing closed openings (the ones that I used to hate because of the lack of tactical possibilities!). And I have to learn to trade down pieces in a favourable way. Further I am not going to draw by negotiation anymore. Without these three measures I have no chance to get enough experience in endings.

    I do this only for study reasons.
    But I'm sure it will enrich my game and enhance my possibilities. You don't believe I let a mate in eleven go if I stumble upon one, do you?

    If I have to survey DLM's program:
    Repetitio mater studiorum est.
    For the non-polyglots:
    Spaced repetition is the mother of all study.
    I don't think the essence of his method lies in tactics. Or we can talk semantics, endings are for 95% tactical by nature.

    So I will do 7 circles of pawn endings the coming time.

  6. It should be somewhat similar, I think, Tempo. Chess Endgame Training is organized by theme. You start with pawn endings, proceed to rook endings, and progress through minor piece endings all the way to things like queen vs. rook and other odd endings. The themes appear to be arranged in order of importance. Well, next week I begin, and we will definitely have to compare notes!

  7. "Not draw by negotiation anymore." Good idea. Last year I played against a guy who refused my draw offer in a very closed pawn endgame. He moved his King zigzag with no visible plan which made me sort of impatient. So I blundered and he won.