Sunday, September 28, 2008

Combat the king

Now we know what the goals of the enemy king are, how can we prevent that he realizes his goals?

King vs King.
First there is the combat of king against king. These are the weapons of your king:
• Shouldering away
• Opposition
• Zugzwang
• Triangulation
All these weapons have time or tempo as common ground. Shouldering away means that the hostile king needs more tempi to get around you. You lengthen the path to the goal which costs time. Opposition dito, this works only if there aren't any spare tempi (pawnmoves) around. Zugzwang again means there are no spare tempi around that don't harm you. With triangulation you lose a tempo intentionally, thus bringing your opponent in zugzwang. Thus the parameters to combat the king are:
• time
• pathlength
Which are closely connected.

Pawn vs King.
The king has proven to be vulnerable when his parameters are manipulated.

Diagram 1.

It is possible for pawns to keep out the king. This is done by lengthening the pathway of the king to get to the pawns. In the diagram this pathway is infinite, so the time for the king to get there is infinite too.

It hasn't to be so extreme, of course. Even when fewer pawns are standing abreast, the king needs a lot of time to get there.

Diagram 2.

Of course the pawns have an extra weapon besides their ability to keep the king out for 4 moves in the diagram above: the king must stay in the square of the pawns to prevent promotion. But that is an old weapon which we already discussed.

Diagram 3

If we think of holes in a position, we usually amagine that a knight arrives there at a certain moment. But the story above culminates into this: you must put your pawns in such way that they keep the king out. Here white can simply walk to b5 to penetrate into the enemy position. It is the pawn structure which allows this.

Extrapolation to the rest of the game.
Again we can extrapolate this principle to the rest of the game. When pawns are preventing each other mutually from promotion, which is the case right from the opening until the creation of a passer, often not before the endgame, there can be only one method to decide the game: penetration into enemy territory in order to attack the pawns from behind. It is the pawnstructure that allows or prohibits this. This is in no way contradictory to my findings in the past about the middlegame: the importance of piece activity and invasion squares.
As I have said in extenso, I'm talking about the ideal game here, that is with no accidents like gaining wood or mating the king by tactical mistakes. Tactics are overrated LOL.

1. I hesitate to publish this new post while nobody has commented on the previous one. But I must keep up the pace before my thoughts evaporate.

2. I'm an old codger who is making progress after forty years of no improvement. Played tournament chess off and on during that time without any gain in rating points.

Now after retirement and some health issues overcome, I'm serious about improving. Analyzing my past games for my numerous mistakes, it became clear I needed to understand and improve my combination skills and basic endgame understanding. Too many points lost or wins drawn because of these deficiencies.

Grampmaster

3. "I hesitate to publish this new post while nobody has commented on the previous one."

OK, I will take on the role of Devil's Advocate to your ideas. ;)

"As I have said in extenso, I'm talking about the ideal game here, that is with no accidents like gaining wood or mating the king by tactical mistakes."

Don't forget strategic mistakes. A player who makes a strategic error may be mated or lose material 20 or 30 moves later (beyond the tactical horizon of humans or even machines) without any blunders. See Vliet - Znosko-Borovsky, Ostend 1907 as an example.

Also, players may make strategic or tactical errors that don't lose material or result in mate but mean the endgame is not decided by penetrating and then attacking their pawns from behind (although very often this is the way to achieve victory, especially after sac'ing a passed pawn). It may also be decided by pushing a more advanced passer so that the opponent must sac a piece to stop it from queening.

Your precondition--a game without any tactical or strategic accidents--focuses on only a subset set of endgames that one must know. After all, most people make "accidents" of one sort or another, and so in the ensuing strategic endgame one side or another has an advantage--perhaps an extra pawn but sometimes only a better king or rook position--and the key questions are whether or not it's enough to convert to a won theoretical endgame and then whether the players know the proper technique for it.

If the Reti study had arisen from a real game, the answer to the first question would be no and the answer to the second question would depend on whether the "weaker" side had sac'd enough time to Caissa to know how to hold the draw against best play. ;)

4. LF,
A player who makes a strategic error may be mated or lose material 20 or 30 moves later

OK, logically you have a valid point here. If I prevent my opponent at move 7 to castle I might mate him at move 30. A lot of the gambits work this way. The question is, how unavoidable is the final disaster? After this investigation I will have to look into that.

Also, players may make strategic or tactical errors that don't lose material or result in mate but mean the endgame is not decided by penetrating and then attacking their pawns from behind

If your opponent offers you a shortcut by making errors, you grasp your chance of course. Even if that means that you don't penetrate in his postion and attack his pawns from behind.

Your precondition--a game without any tactical or strategic accidents--focuses on only a subset set of endgames that one must know.

Before you can play the ideal game, you must of course learn how to play the non ideal game. That is, you must learn to punish the errors of your opponent first.

But if your opponent doesn't make an obvious mistake, what then? You will have to make a move, but on what grounds will you select one? These posts are directed to give an answer on that question.

The ideal game must be a draw anyhow, I suppose.

5. Your precondition--a game without any tactical or strategic accidents--focuses on only a subset set of endgames that one must know.

That was not what I meant. I meant only without tactical accidents. Strategy is the area I'm investigating right now.

6. I understand Like's points, but I think what you're saying here based on the assumption that there has not be a significant gain of material. Chess would be much easier if we knew that every game our opponent would give us something.

In those games where everything trades down evenly or maybe one player has a one pawn advantange then this type of information becomes invaluable. Sometimes one can be down a pawn, and still win because he's able to exploit the holes in the pawn structure and penetrate.

I hope you'll follow up with some examples. Excellent posts!

7. It is completely unclear to me why an "ideal game" should end with an endgame. It might well be possible that the ideal game (played by the super computer that "solves" the game of chess, is already decided in the middlegame.

As your assumption of an ideal game is the foundation of this quest, I do not know if you have struck gold here. Yet I am very interested to see where it ends.

8. Wow, two new posts already, you are motoring along at quite a pace.

Your previous post, incidentally, is excellent. I love how you tie it in at the end with Philidor's position. It is the best motivation for his claim that I have yet seen.

The ideas here I have seen over and over again when I start to cross over into the endgame--subtle pawn structure difference that leads to a win for one side because of the blockading potential of the pawns.