Tuesday, March 27, 2012

Broken chunks

Black to move.

I'm trying to figure out what is going on in my mind during the solving of a puzzle. Why doesn't the experience of one puzzle transfer to the next?

This diagram is very typical. The problem has a rating of 1904. Which means I'm not the only dummy in town.
After 15 seconds I played 1. ... Qxh3, which fails.
I had trouble to see why. But after 2.f4 black has nothing. Of course 1. ... Qf4 prevents that and white has to give up the queen to prevent mate.

Those patterns can't be more familiar:
  • Mate with the queen on h2. Played it hundreds of times.
  • Closing the diagonal with f4. Played it hundreds of times.
  • Decoying the pinned knight. Played it hundreds of times.

Yet, only the first pattern comes up, blinds the mind for the other two patterns and I fail.

It is evident that more plan A is not going to bring me any further. Since the patterns cannot become more familiar than they already are.

This is exactly the way I fail time and again on 1800-1900 rated problems.

A thoughtprocess can help me out. If I had taken more time, I probably would have seen the two other patterns. But somehow this is unsatisfaisant. Of course, if it doesn't come up automatically, thinking is the next best thing to do. But in my opinion, this should be automated. I think a better player than me has automated this. I think that a better player is better because he automated this.

What is the method to accomplish this? Since plan A is clearly not capable to have an effect on this. A pattern cannot become more familiar than familiar. After this little sidestep towards plan A, it is time to dive deeper in the investigation of plan B. How on earth are we going to automate exactly this?


  1. These 3 pattern where not learned enough. They would have popped up all 3, during the first seconds and you would have had much time to select wich is best and how to combine them best.

    The chunk "battery" is missing in your list:

    I recognised, that many better player like to play for a Bishop-queen-battery ( its a common maneuver for them )

    The missing of the "battery" in your list tells me : this pattern is not part of your chess-nature.

    Often is often, not often enough.

    It is evident that more plan A is not going to bring me any further. Since the patterns cannot become more familiar than they already are.

    That is not true. There seems to be "no limit" of the grade of "beeing familiar". As higher the difference "your rating - problem rating" is, as quicker you can solve a problem compared to the "average" tactician (At average!!). Masters are muuuch quicker at easy chess related tasks than we.

    Maybe you like to read this:

    It is evident that more plan A did bring you further.

  2. It is very difficult when you have the smell of blood in your nostrils to quickly perceive the adversary's defensive resources. Thus, your own mate threat and the timer cause you to play the move that fails before you see 1) why it fails, 2) your secondary threat, and thus 3) the correct move.

  3. I wanted to add: good that you failed this puzzle. As long as you remember it, you will know, that a possible response to such a check mate threat could be to get the f-pawn away, which block the diagonal and gives a white heavy piece such as queen or rook the possibility to defend the point h2.

    You failed it, because you where not aware of this defence pattern. You added intelligence by learning this defence pattern.
    You will keep it and apply it subconsiously in the future if you make sure you dont forget about this puzzle. Memorize it through repetition. Next time in your real game, you might be the defender. Than it is not only good to know the check mate threat pattern which fails here, but also to know a defence pattern.

    It is true, there are thousands of patterns. But this defense pattern you did not consider. You missed important intelligence. I mean, you are not going to repeat this puzzle over and over again for the check mate pattern. You are right, you know that already. No. You repeat it for learning the defense response. True, isnt it? If that is now plan A or plan B - I dont know. But what is important: by memorizing this puzzle as a warning example, you will know about it in your real games.

  4. I looked at Qxh3, saw f4 in response and then saw Qf4 as the remedy. No defence because the knight on f3 is pinned.

    Patterns are only one part of the tool box. The two most basic analytical tools are:

    (1). Look at all your moves, no matter how stupid.

    (2). Look at all your opponent’s replies to your chosen move, no matter how stupid.

    I always do both before playing a move. If you look at Qf4, it is obvious that it works. If you look at f4 in response to Qxh3, it is obvious that Qxh3 does not work.