Sunday, September 10, 2006

Terra incognita

The last 3 weeks I focussed on repeating old sessions of CTS. I memorized about 1200 positions. If our old theory of 1 problem=1 pattern and 34 patterns=1 ratingpoint is correct, that should lead to an improvement of 35 ratingpoints at CTS. But I noticed no improvement whatsoever. Even a little decline. There are two possible explanations for this. The first is that this is a normal statistical anomaly. I have seen that in the past. But I have never seen that it lasted for 3 weeks.
The other explanation is that the theory isn't correct.

So we might have found extra circumstancial evidence that massive repetition isn't the way to go.

I reworked my system of mates with The Mother Of All Mates (MOAM).
I found 29 MOAM. They are derived from King of the Spills unsurpassed Fundamental Checkmates. If you look at the 29 MOAM I think that everybody would agree why there are exactly 29 and not 28 or 30.
From this I derived 55 variants of MOAM by adding pieces to the positions. This happened on a somewhat subjective base, and possibly I missed a few important ones. I tried to define the positions that the mind sees as unique. Since we don't know how the mind works, this is under investigation.

From the point where we are arrived now I can go in two directions:

I can go in the direction of more diversity. The 55 mates are end positions. There are probably a few hundreds (pre-) positions that lead in a forced way to these 55 end positions. These "pre-positions" are possibly the ones to store in our system.

The other direction is to search for "vision". MDLM had a few vision-drills. If the geometric essentials of the 29 MOAM can be found, maybe sophisticated drills can be developed which would help to let the mates "pop up".

If I look in the direction of more diversity, I can see no end. So I start with research on "vision".


  1. I humbly disagree with the idea of memory as the key to success. Of course you can boost your rating at CTS just by learning all problems and solutions by heart. Remember that cheater at the top of the list some months ago? He claimed to have just done this. But I am sure hundreds of guys with hundreds of CTS points below him would beat him at the board all the time.

  2. When you returned to CTS, did you return to your previous level of activity, and how many days did you do this? I can easily imagine that being away from CTS for awhile can cause to a rating decline. But after you get back in to the swing of things, the rating will go back up.

    That's just a guess, of course it could be wrong.

    While everyone improves using different means, for me, what has helped the most has been focusing on general weaknesses, not necessarily specific problems. I gave some examples in my blog not too long ago (double attacks by queen, overlooking defenses) and I'll probably post on a few more in the future.

  3. no talk now, but staying in touch. i need to get better, so need to sit way in the back.

  4. Mouse,
    I don't know what to think. But I'm going to find out.

  5. Loomis, you can't say I "was away" from CTS since I reviewed my old sessions at CTS, with the aid of CTS and under the same conditions. I only solved "old" problems in stead of "new". BTW I did 40 new problems a day anyway. Which is 10-20% of my usual amount.

  6. Of course I do not claim to have found out why you have improved at CTS. I just have some ideas. Some may be true, some may be false. Just consider these: One part may be due to a better chess vision after massive repetition. A very small part may be due to better calculation, very small because you use to move fast at CTS. One part, I do not know how much, may be just cheap points without any effort, because you remember the problem and its solution from an earlier session. In conclusion: You may have aimed your training at storing patterns in your memory, but the real improvement may be due just to better chess vision. This also would explain why your rating curve does not follow your memory curve.

  7. Mouse,
    I was thinking along the same lines. I have to find a way to prove how it works and if there is a better method than massive repetition to improve vision.

  8. tempo, do you visualize the pieces shuffling around when you solve 'known' problems? I try to keep doing it myself, and also noting all the relevant structures, even when I already know the correct move by recognizing the position. sort of like calculating, only a lot faster.

  9. to be bare a_s honest, jokes at wormstars post aside, just now, i am honestly more excited by hearing that you feel no progress than that it is all maybe figured out. im serious.

    in the Gurdjieff work, to know that we do not know is the first step in inner awakening. utill we find outselves fast asleep and acknowledge it, we cannot begin to wake up, if not find a teacher, a true teacher, who they themselves have already traversed these same paths.

    think about that.

    fear and doubt are the worst, my guru said. but to think we know and overconfidence perhaps worse?

    written on 9-11 in the usa. in my country, our certainty that we know is our biggest problem. we know nothing here.

  10. Worm,
    by times I used the method of visualising you described. It might be a clue indeed. More experimenting is needed.

  11. Trans,
    chess study is finding out how the mind works. The results can be used on a much broader area than chess alone. As Gurdjieff said, when someone has mastered a skill, really mastered, for instance making coffee, we can talk with him. The truth is that nobody ever masters a skill, we are all jacks of all trades and masters of none (free translation).

    I'm in a maze, going left on every occasion. This is just another dead end and a little closer to the distorting mirror.