## Thursday, September 07, 2006

### Following the logic

Yesterday I developed a logical theory about the basic patterns and their lookalikes. Since I have to obey logic, the road ahead is evident as Hobson's choice. Today I analized a set of 68 problems. The set was derived from CTS by me doing them wrong. It's far from clear how general or biassed the problemset of CTS actually is, since they don't answer questions about it. But now I personal picked them myself by doing them wrong, it is at least clear that it is not a representative set of problems. But I do them wrong while good chessplayers don't, that's good enough for me.

I analysed the tactical key elements of the problemset:

 Tactic Key element Non key element Trap 24 8 Double attack 18 3 Pin 12 5 Discovered attack 7 1 Skewer 2 3 Overworked piece 2 Intermediate move 1 2 Stalemate 1 Invasion 1 1 Total 68 23

So a trap played a role in 32 of the 68 problems.
In 24 of the cases the trap played a KEY role and in 8 of the cases a NON KEY role. This division is rather subjective of course but hey, it's my blog.

For me a mate is just an instance of all possible traps. A mate is trapping the king.
It is possible to refine the traps by the piece which is trapped:

 Trapped piece Key trap Non key trap Total King 16 8 24 Queen 4 4 Rook 1 1 Bishop 2 2 Knight 1 1 Total 24 8 32

There are zillion possible traps. My theory presumes that there are a few hundreds of basic patterns and the rest are lookalikes. First I will try to identify the basic patterns. After that I will try to find a method to store the basic patterns into LTM by frying pigs on an open fire or whatever method seems suitable. Recognition of the remaining zillion lookalike patterns should be a piece of cake by that famous pattern recognition lobe of the brains.

A trap is one of the most complex tactics around. To make the start easier, I begin with mates. There are allready some lists around at the web which could be of help.
Wish me luck.

1. I am assuming that your definition of trap is a piece that has no safe place to go and there is a forced loss of material. In Ct-art sometimes these are called limited-mobility. I found these often hard to recognize .As I did them more often in looking at the position I got a quick feeling of extremely crowded conditions as well as my pieces covering large tracks of areas. Often these two conditions go hand in hand with a trap.

2. do you think there is any benefit for one to reconstruct the position in ones mind's eye ?
I wonder if that would impact retention/recognition

3. Tak,
trap positions are often very hard to see indeed. On the other hand, if it is possible to define basic positions here and lookalikes, there is a great chance my theory can work in practice. But I'm happy to start with mates only, which should be a little less complex.

I have experimented with reconstruction in the minds eye of positions, but it didn't appear to have special beneficial effects. In contrast with my expectation.

4. Wow another ambitious plan, I wish you all the time and energy required to complete it. BTW I have similar plans and I am curious where they will lead us.

I get most queen trappings right, but using much more time than for checkmates. Just because a queen is more mobile than a king, and I use to check all possible escapes before I move.

How do you trap queens? Just play a tempo and hope that she cannot escape? Or check all before you move?

5. Mouse,
there are quite a few queen traps at CTS. Doing much of them has improved my "queentrap-vision" greatly. That doesn't mean that there isn't sometimes a moment of gambling: "if there is something in this position then it must be a queentrap". But I don't think that's really bad. What I'm trying to train at CTS is that my unconscious procedural memory-braincells release the right candidate move as first without me knowing how I do it.

6. I see what you mean. For my part, I prefer to know what I do and why I do it. Relying on unconciousness would leave me suffering from paranoia, constantly fearing that my unconcious brain, in its own bad will, comes to the conclusion not to feed me with the right move.

7. Tempo: this is a really ambitious plan, but certainly interesting. Lately, I have been thinking about the question what a "pattern" really means, too, but have no good idea yet.

Anyway, concerning the traps: I also think they are hard to see. May be this is because recognising them requires to think more about which moves the trapped piece can do as opposed to where can I move my pieces to capture something. IMO, this makes traps a bit special.