Thursday, August 30, 2007

A reaction to a comment from Loomis

In a previous post I contended that it is not good to start to think at the begin of a line, since that means trial and error. It is impossible to start at the end of the line, since that is not yet physical visible. Hence I consider it to be best to begin at the second motif and work your way back to the beginning. The second motife is often already partly visible. In yesterday's post I showed you how that works. I'm very happy with the discovery of invasion and overloading, since that is a common second motife in complex positions. Often being responsible for the complexity in the first place.

Loomis showed how difficult it can be to work backwords to the beginning. This is my first attempt to formulate a narrative to assist in this part of the job. The art is to formulate it in such way that it starts to look simple.

White to move and win.

Yesterday we found:
  • g7 is the most vulnerable invasion square.
  • the queen is (can become) overworked since she has to defend g7 and e6
So the task to accomplish is to clear the long diagonal. In order to reach that we have to step up the pressure against e6, to remove the blockader of e5.

There are only two moves which accomplish this: 1.Rxe6 and Bh3
Which one of them is best?

1.Rxe6 Qxe6 white has invested an exchange, so the attack must work
2.Bh3 this clears the pathway for e5-e6

Now what is the crucial question to ask here? In fact a stardard blundercheck:
Is there a counterattack that works?
Since a counterattack is always to prefer above defense. The counterattack
2. ...Qxa2
works. It attacks the bishop. The power of whites attack depends on this very bishop.

3.Rf2 Rf8 attacking the defense of the bishop immediately.
4.e6 Re7 now black is defended well. g7 is protected and e7 is prevented. Ofcourse white can win his exchange back, but at the cost of his strong bishop. Once whites second rook is traded it becomes evident that whites king is vulnerable too. Which equalizes the position.

Now let's have a look at the winning line. Why is it so winning?

1.Bh3 Bxh3 in stead of taking the semi-sacrifice black can try to reinforce e6. More about that later.
2.e6 Bxe6
an interesting diagram.

Black to move.

The material balance is restored.
The black queen can't take on e6 since she has to protect g7.
The rook hangs but cannot move because it must protect the knight.
h7 is pinned. The only chance for black lies in a counterattack.

3. . . . Nf8 now the black rook can escape if white saves his rook
4.Qf4! counter-counter attacking the black rook at b8 and defending g3.

To be honest, even the grandmaster didn't find this move and would have continued with 4.Rf6 Rxg3+ 5.Kh2 Rg7 6.Rd6 which wins easily too.

Now let's have a look at the last part where blacks decides to reinforce e6.
1.Bh3 Re8 The alternative 1. . . .Nf8 doesn't work: 2.Rxf8+ Rxf8 3.Bxe6+
2.Rxe6 Rxe6
3.Rf6 Rge7
4.Qg5 Qe8
5.Bxe6 Rxe6
I haven't been able to formulate a narrative for this line. Since there is no investment of matter involved the line seems to lead to a very good position without risk.

Now what can we make of all this? The most eyecatching fact is that for every defensive move you have to ask yourself is there a viable counterattack possible in stead of the straightforward defensive moves? So that adds up to 3 basical motifs where I have to look at in complex positions:
  • Invasion square
  • Overworked piece defending the invasion square
  • Counterattack
That should have prevented the mistake 1.Rxe6.
I guess that these 3 motifs cover an immense amount of complicated positions. Because it are this kind of motifs that are hard to see, hence they make the position complex.
I have looked at my latest problemset and this is what I found:

25% of the problems is busted by Rybka (!) indicating that even grandmasters are having trouble with complex tactical positions. Busted means in general that the position is drawish in stead of a win. Looking at the non-busted problems I found the following divison of tactical elements:

100% Invasion
56% Overloading
40% Counterattack
32% All 3 motifs

There is more to say, but I have to prepair for my choir now.

to be continued. . .


  1. What is the difference between a motive and plan?

  2. This is very good. As your blog has been for a long time.

  3. Blue,
    I use the word tactical motive for a building block of a combination.

    Such are:
    Pin, skewer, double attack, invasion, overworked piece, counterattack, zwischenzug, discovered attack etc..

    A plan is something like "I want to trade off to an endgame with BOOC", "I'm going to counterattack in the center" etc..

  4. Ahh, a motif!

    A motive is something that...motivates you....That's why I was confused.