Wednesday, April 20, 2016

Counting

During solving a position, the following diagram came along.

White to move
1rr5/4pp1k/p2p2p1/3P4/3bnPPp/B6P/PR4BK/2R5 w - - 0 2
solution

It took me about a minute to work out the best sequence to capture. That shows a clear weakness of me, I have the feeling that I should see this kind of sequences immediately. This slows me down in every position with multiple mutual captures. How to improve this?

Something really weird happened just yet. The following position took me 5:58 min to solve:

White to move
 6Q1/pb4r1/1p2N1k1/3p1p2/2P5/7P/1q3P2/2R3K1 w - - 1 1
solution

5:58 min?? Well, I was tired after doing a lot of exercises, but that's a poor reason. It shows a fundamental flaw. The fact that the problem has a blitz rating of 1715 shows that this flaw is pretty common:

Problem Blitz Rating:1714.7
Blitz Av Seconds:00:51
Blitz Attempts:200
Blitz Success Rate:64%

Problem Standard Rating:1517.4
Standard Av Seconds:02:58
Standard Attempts:790
Standard Success Rate:63.04%

Intuitively, I feel that having it correct in, say, under 30 seconds would be much more appropriate. Anyway, it's definitely an area that needs much, much improvement.

12 comments:

  1. Diagram 1: White to move.
    This time I think that position is quite easy to solve. I did the following things: compare the attacked pieces vs defended ones. It looks like white attacks 3 pieces and black just 2! And the IMPORTANT feature is that nice horse being undefended! And for me the sequence Rb2xRb8 is the logical sequence because if Rc8xRb8 then the unprotected horse falls down. Anyway I checked the second variation Rc8xRc1, but it looks like Black blundered a whole Rook!

    In gerenal just 3 captures to consider: Rb1xR, Rc1xR and BxN. No more checks, mating threats, traps, etc. At least I did not notice these.


    Diagram 2: White to move.
    Quite a difficult one, but I hope I did (solved) it correct. The simplest scenario: TO EXCHANGE all the pieces and transpose into the ending R vs B (with pawns at both sides).

    1.QxR+ (necessary as it is check and double attack at black Queen) QxQ 2.NxQ KxQ 3.pxp Bxp 4.Rc7+ K(any) 5.Rxp and there should be a winning position for white.

    All the other options may be hidden or not that good (at least I did not find a better solution withing 5 minutes time).

    To sum up: the last position is really tricky one. After crunching he variations - I accidentally noticed that black K+R are the same OPEN line as my King! That's what we can call 'a discovery'! ;) :)

    ReplyDelete
  2. I think its necessary to be able to use "all" methods to solve a chessproblem and switch to the right one according to the "type" of the puzzle.

    Diagram 1 :
    Material: White has the Bishoppair for a pawn so White is ~~-0.5

    Last move of black: direct threat of BxRb2 and RxRc1

    The puzzle is not about Mate or Pawnpromotion and not about winning 3 pawns so its about winning a piece

    Candidate moves: Ba3,Kh2 and all pawns have no good moves

    The problem with these "endgame" puzzles is that its common that more or less all pieces are tactically weak. Endgame puzzles are more based on clalculation then middlegame puzzles
    ( Weak are the rooks and the knight and some pawns but not the bishops )

    So we just have to calculate sequences of takes. And that can be done quick.

    Diagram 2 is the nasty one
    Material: white is down a pawn
    Black is threatening to take the Qg1 or the Rc1+ so we know we has to do something very drastic at move 1
    Looking at this position as middlegame position the decision of Mate/Material/Pawnpromotion would lead us to the conclusion: Gain of material but: this is an endgame
    The seemingly "forced" exchange 1.Qxg7+ Qxg7 2.Nxg7 Kxg7
    results in a position: RPPP vs bpppp which is, as we all know, usually hard to win : https://en.wikipedia.org/wiki/Chess_endgame#Rook_versus_a_minor_piece

    But the advanced white c-pawn , a potential Rc7 and the free pawn at h3 will do the job

    So this puzzle is about pawnpromotion
    Puzzle 2 is rated that low because there is sinply no other move to do, its "easy" to see that every other move cant be the solution. Calculating shows us that we have to chose between Queen takes of Knight takes...
    In a way this is an endgame-judgement-puzzle and not (absolutaly) a tactical puzzle

    ReplyDelete
  3. The point I wanted to stress, is that in diagram 2, it took me 5:58 min to notice that I'm an exchange up after 1.Qxg7+ Qxg7 2.Nxg7 Kxg7

    I haven't looked at the endgame at all. Which, now I look at it, doesn't seem too difficult, for reasons mentioned by Aox.

    What I mean is, that apparently every now and then it goes terribly wrong when counting a (simple and forced) sequence of captures. I think it should be normal to intuitively know that you are an exchange up after the sequence of captures. This means that something is not automated what should be. Indicating that the short term memory system suffers from an unnecessary overload. Which undoubtedly influences my performance at all kinds of positions with sequences of captures. That is the essence of this post.

    ReplyDelete
  4. AS I looked at the positions, I tried to pay attention to what caught my eye first.

    In diagram 1, it was the convergence of the White and Black pieces on square b2 that first drew my attention. The next thing I noticed was the (apparently) undefended WRc1 and (hanging) BNe4. I realized that Black has a superiority of 2:1 on square b2 (this is NOT an example of the encircling motif because that motif requires the target to be immobilized), so there is a sense of urgency in figuring out what to do with WRb2. Then I noticed that the WBa3 would be protecting the WRc1 (geometrical motif) if the WRb2 moved somewhere else. In the meantime, the BNe4 still hangs (Loose Pieces Drop Off motif). By this process of reasoning, it seemed that 1. WRxb8 would be a good first move to investigate. It seemed that Black is forced to recapture 1. ... BRxb8 rather than 1. ... BRxc1 because after 2. WBxc1, Black is down a Rook and the hanging BNe4 must be addressed. After 1. ... BRxb8 the conflict around square b2 has been resolved. I then turned back to the still hanging BNe4, and it can be taken off. It took less than 30 seconds to go through this reasoning process.

    In diagram 2, I first considered whether there was sufficient restriction (a "box") around the Black King to allow two White pieces to mate. I intuited that the answer was "No" because there are too many escape squares and any move by the Black King exposes the attack from BRg7 on the White King. There is a hanging WR at square c1. All things being considered, it appears that exchanging off on g7 (winning the Exchange) puts White in a good position for the ending and solves the problem of the hanging WRc1. So, 1. Qxg7 Qxg7 2. Nxg7 Kxg7 3. cxd5 Bxd5 and we are safely in a superior endgame. At this point, the fork 4. Rc7+ K any 5. Rxa7 leaves White clearly an Exchange ahead in a won endgame. That thinking process also took less than 30 seconds.

    FWIW, I don't explicitly use a formal thinking process. I try to look at the entire board and get a "feel" (an intuition) about which area(s) of the board have something going on that piques my chessic interest. It is only as I look at those things that I begin to get some idea of what to do. Generally, I don't start looking for candidate moves until after I have that "gut feeling." I try to avoid the "if he goes here, then I go there, etc." calculation process until I am sure I understand where the conflict points are that must be resolved. After prioritizing those conflicts, I then begin figuring out the move sequence(s) to resolve them. In short, "look wide before looking deep."

    BTW, this is NOT indicative of any presumed "superiority" on my part. I'm quite sure there are many positions that you all can solve in a fraction of the time that it would take me to solve them. I'm just trying to share what I did in the hope that it might prove helpful to someone; that's all.

    ReplyDelete
    Replies
    1. Your first hand observations are HIGHLY appreciated. It helps to confirm or deny my own thoughts about how to approach these kind of positions in the future. In this specific case: to confirm.

      Delete
  5. At Diagram 2 you are not an exchange up at the end, a pawn is missing!
    With an exchange up that would be easy

    Counting: you have to count at every move
    You start with -1
    Queen takes rook : +4
    Queen takes queen:-6
    Knight takes Queen:+4
    King takes Knight: +1 which is usually not enough!


    ReplyDelete
    Replies
    1. I should have added: I'm an exchange up RELATIVELY TO THE START POSITION. But heh, I thought you were smart enough to add that yourself ;)

      Delete
    2. A move is ok with CT if the evaluation is more than +1.75
      If you are an exchange up then this is usually ok for CT, but if you are less than an exchange up then this is usually an error! at CT, you would lose points.
      There is need for further! thinking if your material balance is less than 1.75
      points at the end of your line.

      Delete
  6. No disagreement from me that after the completion of the exchanges on g7 (without any further consideration), Black temporarily has a Pawn as compensation for the Rook versus Bishop material imbalance. However, the position is not yet in quiescence; the c4-d5 Pawn vis-a-vis must be resolved. As soon as the exchange occurs on d5, the c-file and the 7th rank are opened for the White Rook, resulting in the gain of the a7 Pawn, with a clear imbalance of Rook versus Bishop. I am pretty sure that I could win in this position against any master, presuming I was not in severe time pressure. (Or, maybe that is just hubris on my part.) In any event, the position is now quiescent and my evaluation is White can win. Whether White will win is an entirely different question.

    Perhaps I am an oddball, but I don't usually count the material balance at every move during an exchange. Instead, I try to have an idea of which pieces to exchange and the order to exchange them that takes me closer to the goal(s) that I have in mind. The goal(s) may (or may NOT) be gain of material on the exchange square. I count the material balance at the end of the sequence of the exchanges. Maybe that's one of the many things that keep me at a patzer (less than master) level. . .

    ReplyDelete
  7. Perhaps I am an oddball, but I don't usually count the material balance at every move during an exchange.

    This post was in essence triggered by the fact that I felt that it is unnatural to need to count during captures. I feel it should be possible to just know whether I'm up or not without an intellectual effort. I don't feel qualified to comment on the oddball part.

    ReplyDelete
  8. Weeeelll,
    either you CAN do without.. or you HAVE to do with it
    ;)))
    At this easy! puzzle i can do without, i even see it before i calculate the line... but very often i have! to count

    ReplyDelete