Three sitting ducks

From study of middlegame strategy by means of games like the one of Karpov in the previous post, the following picture emerges:

There are three sitting ducks: The King, the pawn and the promotion square.

All these ducks have their own parameters around them which tell you when they are ready for the eagle to dive in.

Piece play is guided by which duck is currently the most vulnerable.

The King
For the king we have the preconditions of Vukovic. The parameters are geared around King safety, lines of attack that end next to the King, amount of attackers and amount of defenders. The battle of the LoA's.

The pawn
Isolani, backward pawn, doubled pawn, base of the pawn chain. Can they be blockaded? How about the open files adjacent to the pawns? Can they be attacked from behind? Encircling, undermining.

The promotion square
Where are the pawn majorities? Can they create a passer? How close are they to the promotion square? How distant is the opponent King? Is the majority mobile or fixed? One holds two. Two fronts.

The battle ground
The sitting ducks tell you something about the targets and the state they are in. The pieces are striving to help where they can. To protect, to clear the way, to blockade et cetera.
A special case can be made for the dual purpose piece move. Where one move brings you nearer to two ducks.
The pawns themselves are double edged. They can act as a passer, striving for promotion, as a cowbar, to pry open, as a wedge which divides the opponents forces et cetera.

Variation free analysis
I think that the points above form a framework to base my analysis of a game on. I expect that the main part of my future study will exist of fine tuning this framework with the aid of annotated grandmaster games. Have a look again at the game of the previous post and try to identify this framework in the game. Most comments are mine, btw, albeit inspired by the comments of the author and kept in check by GM Stockfish.


  1. First step: make a choice which duck to go for.
    Second step: chose between the standard scenarios.
    Sofar, everything is variation free.

  2. Word salad is on the menu, raw with no fancy dressing:

    PART I:

    There is a logical progression in strategy irrespective of the stage of the game. Each phase leads to the next phase, based on the strengths and weaknesses of the pawns and the (perhaps ‘accidental’) placement of the pieces relative to the pawns. We are accustomed to focusing on individual phases and often don’t progress to “SEEing” the forest because of the individual trees, or, at best, a small grove of trees.

    For example, there is a logical sequence associated with pawn tension in the center, creation of open files, creating an outpost on an open file, and penetration to the 7th/8th rank with the major pieces. We often fail to realize that there IS a sequence of operations involved, each phase dependent on successfully navigating through the previous phases.

    There is a logical sequence between pawn structure and piece development (from the opening into and through the middlegame) in accordance with that pawn structure, which leads to the “sitting ducks”: an attack on the king, the pawn structure and pawn promotion. In any phase, there are preconditions (commitments) that must be made in order to make progress toward the eventual goal. It is trivially obvious that the overall goal is to checkmate the opponent’s king. (After all, that’s in the formal rules.)

    KNOWING WHAT (in the broadest general case) does NOTHING to provide clues or signposts pointing toward KNOWING HOW.

    Here’s an example from My System, Part I, Chapter 4: The passed pawn:

    FEN: b5k1/P7/K3p3/4Pr2/8/8/8/1R6

    What is the overall goal? To checkmate the Black King. DUH! That provides NO information on HOW to continue in this specific position.

    If we look for a logical sequence (a scenario, if you will), we are much more likely to figure out WHAT has to be done and (more importantly) HOW to do it.

    There are insufficient pieces to attack the Black King with the intention of checkmating it and the Black King has considerable mobility. That kills the first duck.

    The pawn structure provides a clue or two, but those clues have to do with the third duck, not he second one. The second duck goes into the soup pot.

    However, the third duck leads to pawn promotion but “there is no joy in Mudville; mighty Casey [the a7 pawn] will strike out.”

    The first phase is to ”Changez les blockadeurs!” [Nimzowitsch]. How can we get rid of the BBa8 which is blockading our pawn promotion? Easy-peasy: give the Black King a check with the Rook, allow Black to interpose the Black Rook AND simultaneously protect the BBa8, then capture the Black Bishop. (Heaven forbid! Do NOT capture the Black Rook just because of that silly “rule”: “Always check; it might be mate.”!!)

    1.Rb8+ Rf8 2.Rxa8! Rxa8 3.Kb7

    The Black Rook will be much more amenable to allowing the a7 pawn to advance to promotion, or (in desperation) may just capture it before it can advance to a8.

    3...Rf8 4.a8(Q) Rxa8 5.Kxa8
    3...Rxa7 4.Kxa7

    There are some arcane bits of KNOW HOW that are required to make progress in the next phase. Assume (in the mind’s eye) that all the pieces are off the board and the White King is on either a7 or a8.

    1. "KNOWING WHAT (in the broadest general case) does NOTHING to provide clues or signposts pointing toward KNOWING HOW."

      Before the administration of the REMEDY you need the DIAGNOSIS. Diagnosis and remedy have a 1:n relation. My diagnosis is that my diagnose ability is below standard.

    2. In an earlier investigation my conclusion was that a master has good diagnosis skills, but poor remedy skills. A grandmaster has good remedy skills too. I can't remember whether I wrote that in a post.

  3. PART II:

    Whatcha gonna do now, huh?!?

    It looks bad for White: his King is farther away from the two remaining pawns.

    HOW is White going to play at this point?

    One of those arcane bits of KNOW HOW is the Réti maneuver, more specifically, that the path whereby the King can traverse squares along a diagonal is just as short as along squares across ranks and files. Another is the Opposition, especially distant Opposition. Another one is Zugzwang in the form of the Trébuchet. A final one is the KNOW HOW to win a King and pawn ending when the pawn is on the fifth rank and the opponent has the opposition. If you do not have these bits of KNOW HOW down cold, you will most likely mess up this ending.

    In the next phase, both players maneuver to get the Kings in closer contact with the two pawns.

    Black is already at somewhat of a disadvantage because the two pawns are on his side of the board. He can either try (defensively) to “shoulder” off the White King away from his own pawn or (offensively) try to grab the White pawn. Both alternatives must be investigated.

    [The move numbers ASSUME that we are starting AFTER the pawn promotion and eradication by Black.]

    (1) 5...Kf7 6.Kb7 (Opposition) Ke7 7.Kc7 and Black must give way, losing his pawn and the ending because with the White King on the 6th rank and the [NOT a rook] pawn on the fifth rank, White wins. Mentally ”SEE” the position WKe6-BKe8-WPe5. Regardless of the side to move, White wins. [KNOW HOW!]

    (2) 5…Kg7 (going for the White pawn) 6.Kb7 Kg6 7.Kc6! (distant Opposition!)

    Black now faces a dilemma: (a) if he directly attacks the WPe5, then White can defend it AND attack the Black pawn, winning the pawn and the ending or (b) if he tries to keep his options open, White forces him into the Trébuchet position.

    7...Kf5 8.Kd6 winning.

    7…Kg5 8.Kd7! Kf5 9.Kd6 winning.

    Too often I have learned little bits and pieces (those arcane bits of KNOW HOW) without SEEing the broader strategic plan (logical scenario) that weaves them together.

    Nimzowitsch favored using endings as examples of strategic thinking because the important aspects were easier to SEE. That most definitely does NOT mean his ideas are only applicable to the endgame!!

    BTW, that same chapter in My System addresses pawn majorities as well as blockading and other issues associated with passed pawns.

  4. PART III:

    Amusingly, GM Stockfish gives Mate in 30(!):

    D36 Mate 30 1.Rb8+ Rf8 2.Rxa8 Rxa8 3.Kb7 Rxa7+ 4.Kxa7 Kg7 5.Kb7 Kf7 6.Kc8 Ke7 7.Kc7 Kf8 8.Kd6 Kf7 9.Kd7 Kg7 10.Kxe6 Kf8 11.Kd7 Kf7 12.e6+ Kf6 13.e7 Ke5 14.e8=Q+ Kd4 15.Ke6 Ke4 16.Kd6+ Kd3 17.Qa4 Ke2 18.Ke5 Kd3 19.Qe4+ Kc3

    1. In the past I had the Nalimov tablebases installed (a few computer crashes ago). Because I wanted to train against it. With the first position, the tablebase said: I resign, since this is a mate 32. It took me an hour or so to find the setting to set the automated resign off.

    2. Even more amusing, GM Stockfish "improved" his play:

      D46 Mate 24 1.Rb8+ Kf7 2.Rxa8 Rf1 3.Rb8 Ra1+ 4.Kb6 Rb1+ 5.Kc5 Rc1+ 6.Kb4 Rb1+ 7.Kc3 Ra1 8.a8=Q Rxa8 9.Rxa8 Kg6 10.Kd4 Kh5 11.Kc5 Kg6 12.Rf8 Kh6 13.Kd6 Kg7 14.Rf1 Kh7 15.Kxe6 Kg6 16.Kd6 Kh5 17.e6 Kg4 18.Rf8 Kh3 19.e7 Kg4 20.e8=Q Kh4 21.Qe5 Kh3 22.Qh5+ Kg3 23.Qg6+ Kh3 24.Rh8#

      It only took 10.5 hours of analysis. Chess Tempo must not be using endgame tables.

  5. (What is that with you and the Trébuchet?) Too often in the past I was seduced with little bits of HOW. But everytime it proved to be very difficult to abstrahere the what from the how. So I became a Jack of all trades and a master of none.

    My mind is always rebellious when confronted with too much how without being explained the what. When I know the what, I can figure out the how myself the hard way.

    That is why I didn't make progress with endgames in the past. The Trébuchet is the last piece of the puzzle. Just lose a game by not knowing the Trébuchet, and I will study it and will not forget it forever.

    1. There was nothing particularly important about the Trébuchet in isolation. I just saw it as a shorthand for figuring out one phase of the logical sequence to get from the initial position to the e-pawn promotion.


Post a Comment

Chessbase PGN viewer