- What is piece activity and how is it measured?
- What is exactly a weakness?
- When can we consider a piece as being developed?
Once I played a knight to a beautiful outpost on c6. I thought it would be a super monster octopus there. But it accomplished nothing. The pawns it could attack were long gone, and the king was on the other flank where my knight could not reach it. So all the effort to get my knight there, was in fact wasted. That leads to the following rule:
Activity of a piece should be measured by what it accomplishes. A piece must do something concrete. Now what are the concrete things a piece can do?
- It can attack a backward pawn
- It can attack a piece
- It can attack a defender
- It can invite an enemy pawn to forsake its duty by occupying an outpost
Although other grandmasters seems to look differently at this point, both Nimzowitsch and Bangiev seem to consider the pawns on the 7th rank to be backwards. Bangievs invites us to look at the least defended weak pawn.
Putting a piece on f5 can invite your opponent to play g6. Thus weakening f6 and h6 in its wake. This is a different way of looking at matters.
Somehow, the idea of the center must be integrated into this. A weak pawn or outpost in the center prevails over one on the flank.
The first question is also highly relevant when developing a chess engine. Currently we have Stockfish 11+, where the latest knowledge in measuring activity is implemented?
ReplyDeleteThe second question: I guess a non-defended or suboptimal-defended square, no matter if a piece is on it or not - that is then a weak square. You can have structural weaknesses, too. like an isolated double pawn. This is a weak structure, because the isolated pawns will never have the option to defend each other (like 2 connected pawns). Also, the front pawn blocks the pawn behind it. You can argue that the squares of the pawns are weak, but that is not all. The blocked up feature is a structural weakness.
If that weakness is relevant is a different question. For instance you can want such a isolated pawn like f2&f3, because you get a free g-file as compensation (where a rook can attack a black pawn on g7?)
Third: pieces do not develop. It is about placing pieces actively. The "development" is just a guideline, but it is not desirable on its own. What is meant is: activate your pieces.
The problem is, that my brain is not a chess engine. So I'm looking for a more applicable way to estimate the activity of a piece somewhat accurate.
DeleteWEAKNESSES
ReplyDeleteParaphrasing Justice Potter:
I don’t know what a chess weakness is, but I know it when I “see” it.
Or, much more likely, we do NOT “see” the weaknesses as weaknesses.
A weakness can take many forms, categorized broadly into either static or dynamic terms. Usually, static weaknesses are associated with the Pawn structure. Piece weaknesses are associated with either (a) being undefended [LPDO] or lacking in harmonious coordination with the other Pawns/pieces.
Lasker notes the following:
"... the target for the attack has to be a weakness in the hostile position. . . . On the Chessboard there are no lines [in spite of our formulation of LoA - Lines of Attack], only points; Hence Steinitz speaks of weak points. The most successful, the most effective combination as well as the widest-visioned and deepest plan of attack - thus his idea – proceed, as if by a miracle, in the direction of the weak points, . . ."
"In other words, as the stability of a position is gauged by its least stable point, attempt to achieve at every point the same degree of stability."
Points (squares) are weak or strong regardless of whether there is a Pawn/piece on the "weak" square. (Nimzovich referred to this idea as the "dematerialization" of weak squares.) Given that a Pawn cannot retreat, any square left behind or left potentially unprotected may become a weak square as the Pawn moves forward.
Dynamic weaknesses occur because of restraints on movement. A piece may be unprotected, may lack cooperation with other pieces, may lack “safe squares” to move to, or may be hampered in its potential movement due to obstructions or by having acquired one or more functions. For example, an overloaded Pawn/piece (having multiple functions) is usually a dynamic weakness.
"A Chess piece is supported by some comrades, it antagonizes some enemies, and thus it has tasks and functions to fulfill. . . .the pieces ought to be supported according to their usefulness, according to their value. . . . In return for the labour which a piece is required to do for the common cause, it enjoys safety and support, and the more valuable the piece, the greater the task assigned to it and the greater the obligation to safeguard it."
"A defect of cooperation is often indicated by the great distance between the points which the pieces occupy-often, not always. . . . But considering the simple geometry of the Chessboard and the laws of motion of the pieces, distance is a first, though inexact, measure of the degree of cooperation of the pieces."
"It is wholly in harmony with the Steinitz theory that the computation of the value of cooperation takes stock of the value of the points: the weak points, the strong points, the important points, the less essential points. Particularly are the weak and strong points of account. The effect of these is decisive: all else is of secondary importance."
Steinitz Rule of Attack: "Attack the weakest points in the enemy camp."
Steinitz Rule of Defense: "Improve the worst weakness voluntarily."
piece activity is: how much is a piece doing, does it defend many important pieces / squares, attack many important squares and pieces. a piece is as more "active" as more it makes moves of the opponent bad. A bishop at f1 behind pawns a e2 and g2 ist not very active, a bischop at c4 ist active attacking a weak f7...
ReplyDeleteor at b5 ist attacking and pinning a Nc6 and therefore is indirectly attacking the square e5...
Your comments are food for thought. So I updated this post in blue.
ReplyDeleteactivity can be defensiv
ReplyDeletethe piece just need to be part in many possible lines
I'm not quite sure whether I'm buying that. A defensive piece has a function, and is hence restricted in its mobility. That's the contrary of being active.
Deleteactive = doing something useful
Deletea defensive piece enable the other pieces to do something agressiv.
Or think about the overprotection idea: by protecting "too much" you gain mobilty even to all protectin pieces.. not that i think its an important idea..
Active = to have a function ( maybe even not now but in many near possible futures )
I'm inclined to gear piece activity around concrete targets. A pawn or a defender or a B.A.D. (Barely Adequate Defended) piece.
ReplyDeleteThe value of a square is measured by the fact whether it is situated on the path from the attacker to the target (on the line of attack). If it is not, its value is zero.
Pawn moves must be designed to enhance your own activity and to decrease your opponents activity.
Yes, true. The most important thing is the target. I have had a game, where all my pieces where active, and I was up material, too. And he had almost nothing - except I was threatened with check mate in 1, and there was no way to stop it.
ReplyDeleteThis isnt the general rule, though. Most of the time, the side with the better activity usually gets eventually a tactical shot - and wins the game. Bad positions tend to get worse. When you see a GM having a positional advantage of a stockfish evaluation of +1.7 - that is most of the time enough for the GM to win the game. (Tt seem that a strong GM needs to have more than +1.0 in order to win, but mainly because stockfish seems to exaggerate a bit in its evaluation-score).
It is all in all not easy to evaluate. If you can not evaluate, how can you know which move is the best move?
Aox had already the idea of positional puzzles, and he did some puzzles. But I guess it ended there. More puzzles would need to be given - and if we did such positional puzzles, our eval could improve.
My opening evaluation improved by looking at statistics: what kind of openings (or opening moves) are statistically promissing, and which moves are getting you rather into trouble?
I share what I found:
an early ...Nf6 is giving white often enough a target. Openings where the move ...Nf6 is delayed a bit are statistically more promissing for black.
Also: if you are black, try to aim for assymetry. If you are white - aim for symmetry.
Moving the f-pawn early is usually not so good.
It is interesting, that a lot of popular openings are thus not so good:
The Petrov Defense is bad (early ...Nf6, and black plays symmetry) - dont believe it? Statistics prove I am right!
All kings-indians (early ...NF6) are not so good, and the main reason why 1.d4 is thought to give white a longer lasting initiative. statistics can't be wrong!
1.e4 c5 - highly assymetric --> indeed, the sicilian is statistically on of the best defenses.
1.c4 e5 - if you are black, you should go for assymetry. So it isnt so great for white to see the sicilian (1.e4 c5), but surprisingly, if white plays 1.c4 --> black should go 1...e5 (assymetry)
If you like KID, better try "The Modern" set-up where the deployment of ...Nf6 is delayed.
If you like to play the ruy lopez as black (you should not play symmetrical with 1.e4 e5 in the first place), you improve your chances if you play a variation where you either dont play ...Nf6 at all, or where you delay it. Example: 1. e4 e5 2. Nf3 Nc6 3. Bb5 g6
Dont expect too much, because you should not have playe 1.e4 e5 in the first place, but you can get away with it just about good enough if you dont follow up with ...Nf6 anytime soon.
You can play around with an opening database and look at statistics. Prove me wrong, or have learned some positional play from me just right now!
This is not fallen astray from the topic. See - there isnt an easy way to evaluate the worth of a move. All depends on when you move it, the circumstances. But there are also truths to be found which seem to remain intact most of the time (like: do not play ...Nf6 early!), or "isolated double pawns" are almost always not desirable - they are weak, and you only embrace them if you get something in return (like, for instance) an attack).
"It is all in all not easy to evaluate. If you can not evaluate, how can you know which move is the best move?"
DeleteWhen I think in a game, I think about the best move. What emerges from all my readings about positional play, is that I should start with Find a plan!. Only when you have a plan, you should start to think about moves that contribute to that plan. Without a plan, you have no tooling to evaluate a move. (You can even do the best move for the wrong reason. Which isn't going to be of much help.)
"You can play around with an opening database and look at statistics. Prove me wrong, or have learned some positional play from me just right now!
"
Statistics explain the what. Which should trigger us to haunt the why?
well 1. where to play, 2.what to attack
Deleteif nothing found.. a.improve activety pieces or b.make opponents pieces bad