Thursday, July 26, 2012

Getting the right idea

The following diagram shows very well where the exact problem manifests itself.

White to move. It is a forced mate.
You can find the solution here.

With trial and error, you will see that all logical lines peter out pretty soon. That's the moment you need an idea. Without the right idea, you can't find the first move. No tactical patterns from the past will help you here, since the position is way too specific. Of course tactical patterns play a role, but what I say is that they are not going to bring you the right idea.

Logic is destructive by nature. It plays a role by the error part of trial and error. It works by means of elimination. Every creative idea that comes to mind is tested and discarded when it doesn't work.

Logic reasoning tells you to look for an invasion square. A square where your pieces can outnumber the opponent. Which can act as a bridge head. That is a common theme when attacking. When all obvious invasion squares are eliminated, you have to look for the less obvious.

f6 is such less obvious invasion square. Your queen and rook can work together to get the upper hand on f6. The only thing is that you will have to find a way to get rid of the only gard of f6 i.e. Rf8.

There are a few standard methods to get rid of the gard. Reasoning this way you will find the deflection move Bg8 soon, which happens to be the first move. Once this idea is found, the rest is pretty straight forward. Trial and error will reveal the right line soon.

If you will be able to solve this problem or not completely depends on if you find the right idea.
My hickup manifests when I have tried and rejected all obvious lines. In stead of going on, I start to repeat myself, maybe I missed something?

The reasoning concerning an invasion square you will find in every mating attack.
The reasoning concerning removal of the gard is evenly common.
The only thing is: why don't these common reasonings pop up when you need them?
Do you favour the term pattern in stead of reasoning? No problemo. The point we have to solve remains he same.

The answer to this will be the root to serious tactical improvement.


  1. I think I know why I did not find the solution: I saw the pattern of "cutting off the escaping squares of the black king by the bishop".

    This pattern is not questioned. Instead it makes you think about 2 moves:
    check with the rook, and check with the queen.
    The advantage of the check with the queen is, that it keeps the black bishop away from the square h6 (which would defend the black king).

    Anyway, it speaks almost for itself that the wrong moves were almost only the rook move or the queens move.

    Patterns are sometimes very disturbing.
    Logical reasoning can filter for the right patterns, but admitedly this is more valid if you see all patterns. If you dont see the distraction pattern, you cant decide with logical reasoning which way to go.
    It is then like you said:
    If you dont see the distraction idea with the bishop-check, you hardly find the solution.
    And after you found the pattern "cutting of the black king´s escaping squares with the bishop", you dont feel like searching for an other idea. Somehow it just looks too convincing.

  2. This is a very difficult pattern, probably extremely rare. Fedorov was unable to find the checkmate during the game...

  3. @Laurent,
    Invasionsquares are quite common. So is the removal of the gard. Why do you think this is rare?
    It isn't even difficult, in fact. You have to find only one logical idea. The rest is straightforward.

    Making these elements work together is what we have to learn. Its the masters playground.

    The only thing I'm worried about is that it takes me two hours or so;)

  4. Great problem, great blog post and great comments! Thank you all for exercising my brain!

  5. There is one thing we can learn from this:
    Once we found a (wrong) idea, but we cant really "make it work" and we finally give in and just play our idea and "let's see"... at this very moment, we should take a last look at all other checks. For about 10 seconds.
    And only then we move.

    I know, in Blitz mode in chesstempo we would not like to "waste" another 10 seconds after we already wasted so much time in trying to make our idea work. However, we only do this blunder-checking in the few cases we dont really see the solution. We need to internalize this blunder-checking within our thinking.

    The other question that remains is: will 5-10 seconds of blunder-checking be enough to find the distraction motive? Sometimes not. Unless you trained patterns like hell. If you invest only a few seconds with the bishop´s check, you will find it if you trained "easy" patterns like "hell".

  6. @Munich,
    I did look at B88+. But since I wasn't aware of the idea behind it, I dismissed it soon as an "error". Trial & error is forward thinking. Starting with an idea and make it work is backwards thinking.

  7. I did not look at Bb8+ at all.

    The trouble you may have encountered even after checking Bb8+ could have had 2 reasons:

    a) impatience: you only looked at it for 1-2 seconds instead of 5-10 seconds.

    b) weak (slow) pattern recognition: you checked Bb8+ for more than 5 seconds, but you just did not see the simple fork Qh5+ afterwards.

    c) you miscalculated with the rook check or the queen check, so you believed that one of these moves would work. So you were sure you found the solution.

    If we could check afterwards what we were really thinking, we could find out what actually caused us to fail.
    Then we would know which of these points a-c caused us to fail, and if it was a or b you could try to change it.

    As for me, I was not even looking at Bb8+, so you did already better than me. (so for me it is point d): not even blunder checking)

  8. @Munich,
    it is a combination of all three, in a sense. The main reason is probably that I combined Bg8+ with my old analysis of Qh4+ and Rh4+. Not seeing that a new opportuny has arisen (Qxf6). It's bias. I see only what I expect to see. Just as in real life.