Tuesday, March 20, 2012

When plan A fails, what would you do?

The reason that I introduced guidance is this post. You may think that this is a freak accident that will not happen to you. You are wrong. This post reveals a fundamental flaw in the human mind. This flaw is one of the main reasons that started WWII. I would describe it as follows:

People have the ability to see only what they expect to see. Which is a great way to simplify the world and to cut down on brain resources. But the downside of this preference is that people have difficulty to see what they don't expect to see!
No matter if I was able to see the mate in 3 in 3 seconds or in 20. Before you can recognize a pattern, your mind must be prepared for it first. You must guide your attention in the right direction first. The speed of recognition has no effect, plays no role, as long as you pay no attention to that part of the board.

Guiding your attention is plan B. Of course, when your attention points in the right direction, plan A kicks in immediately again. New familiar patterns will be recognized is this new area of attention. That's only natural. That is how it works.

Let me give an example.
Black to move.

I was focussing on the knight sac 1.Nf3+, but I couldn't get it to work.
Then I started with plan B. Looking for attacking squares and targetsquares. My attention was drawn to the main focal points of this position: h3 and f2. Then I asked myself how to step up the pressure on these focal points. That leads to investigating 1.Qh4. From there I could construct the different mates. White has to give up material to prevent an immediate mate.

Before I looked at the squares, I "more or less" rejected 1.Qh4 as being not forcing enough. That was why I was so focussed on 1.Nf3+.

So the right order is plan A, plan B, plan A again.

Aox immediately spotted the weak point in my plan: the realationship between the attacking squares and the targetsquares is rather vague. I find it too early to already worry about that right now.
I am already glad that I have identified what has to be improved. The how is the next step.

Speaking of steps, I defined a series of steps in my previous post.
The first step is: identify the potential attackers. 
  • The rooks are already attacking something.
  • The knight does nothing, but he could attack h2 and g1 with tempo.
  • The queen is doing nothing, but it can attack h2, g2 and f2 in one move.
  • The pawns play no role in the attack.
  • The king plays no role in the attack.
The second step is: identify the attacking squares.
  • The rooks are already on their attacking squares.
  • The attacking square for the knight is f3
  • The attacking squares for the queen are f6, g5 and h4
The third step is: Identification of the problems on the road from attacker to attacking square.
  • Nf3 is defended by g2
  • Qf6 is falsified by Nxd6
  • Qg5 has no follow up after Qc7
  • Qh4 steps on the attacking square without the gain of a tempo.
The fourth step is: Solving of the problems on the road from attacker to attacking square.
  • Can white cause me troubles when  I play 1.Qh4 without gaining a tempo? The answer is: not really.
The fifth step is: identify the targetsquares.
Since  1.h4 is left as the only attacking square, what squares will the queen attack from there?
  • h2
  • f2
The sixth step is: identification of the problems on the road from target to targetsquare.
The king is already on a square where it might be trapped (mated).

To be perfectly honest with you, I concocted this list after I knew the solution. Aox already expected I would do that. Yet it gives an idea what I intend to accomplish.

I don't consider the lack of familiar patterns, the lack of relevance of my familiar patterns or the speed of retrieval of familiar patterns to be my problem here. My problem is: how do I overcome my biased mind which causes me to neglect familiar patterns. The bias is here: I don't need to look at Qh4 since it is a move which gains no tempo and white is about to give me an unpleasant check.

Of course I don't say that improving on plan A is unnecessary. Munich has implemented my idea's about plan A very succesful, so I intend to post about that later. The point is: what will you do when plan A fails. More of A or switch to plan B? Even if we yet have to invent plan B.


  1. My first ideas where:
    Ne2+, Qh2+, Rhx#
    Rxf2, Rxf2, Re1+ (White queen must be deflected) Rf1, Ne2, and so on.

    So i was looking for the squares h2,f2,f1,e1 and most of the times e2!
    After i read "The pawns play no role": After a intense exchange orgie, c5 may queen...

    My "clue" to this position:

    Its the most natuaral move. The most passive piece is going to its most active position. Thats a perfect "Smirnov" move ( as far as possible in front ).

    I would put in my thought- process-list: consider the most natuaral moves, even if they seem to be not forcing.

    Maybe higher rated problems have a more positional/strategical touch?

  2. The higher rated problems seem to have the following in common:
    * You need common sense to solve them.
    * They have something that triggers your bias so you don't use your common sense.

    Hence they are high rated.

  3. I think this problems supports Munich experience. He said he would have been getting better in positional play by doing tactic-training.
    I think he might be right ( in many cases, not in all )

    By the way, your "guidance" seems to develop to a special case of thought process for tactical positions. I looks/want_to_be like a (universal) method to solve tactic problems. would be a dramatic novelty. I think Botwinik was working at something like that ( for chess in general )

    Smirnov says: that the thought process should result in a good move, if the position is tactical or not: "First calculate the forcing moves then..."