Tuesday, August 23, 2005

Does CTS work?

And again I am in the happy state that I can speculate over all the good fruits that might be yielded from my study. Not disturbed by allready known facts from harsh reality.
In order to yield results from problemsolving at CTS two criteria have to be forfilled.
The first is, that it is possible to increase your rating at CTS.
The first 10,000 problems were accompanied by a rating progress from about 30 points, from 1470 to 1500. Now I seem to plateau at 1510. On the other hand it is hardly to imagine that the rating would NOT increase by doing 400 problems a day. So I have to look again when 20,000 problems are solved.

The second criterium is, that increase in CTS rating leads to increase in OTB rating.
I have compared Dutch -, USCF-, and FIDE ratings and I found them to be about equal.
To compare CTS and OTB ratings:
1450 1700
1900 2300

CTS seems to make you fast. That is to say, your rating there can only grow when you become faster.
Everybody with a high rating is fast. But is this reciprocal?
It's a well known fact that playing much blitz doesn't lead to better OTB play.
Is there a difference between playing blitz and problemsolving at CTS?
I believe there is.
When playing blitz old patterns are repeated over and over again. You are not necessarily LEARNING new patterns. At CTS, when a pattern is familiar, your solving time will decrease to
less than 3 seconds. Which means that your rating improves. This leads to a shift of the problem window, so you get new problems to LEARN.
So sit on the fence and wait for results or give it a try yourself at Chess Tactics Server.

circle 0: 1470
circle 1: 1500
circle 2: 11,000 /70,000


  1. Have you recognized any problems that you went through in the first circle?

    As you've mentioned, CTS might be good because it forces you to recognize patterns rather than memorize individual puzzles.

  2. Blue,
    working thru CTS is a rather forehead slapping experience. Due to the time constraints and the relative simple problems you are constantly blundering, so to say. But besides that this build your character by making you humble it sharpens the recognition. The way how CTS generates problems makes that there is a lot of repetition allready for the problems which are close to your rating. Some problems I have allready seen 7 times.
    So the idea of 7 circles here is actually rather nonsensical. CTS gives you a good idea what playing a tempo really is. Making a move within 3 seconds is only possible if you don't think. Even 10 seconds can't be called a tempo, because there is some sort of validation going on. To recognize the pattern is the only possible way to move a tempo. And that is, according to prof. de Groot, what GM's are capable of. But if I say I'm going to do all problems at CTS untill I can do them within 3 seconds, guys like you are really going to phone the lads with the white coats, so I say that I'm going to do only 7 circles of 10,000.:)

  3. Here's some incentive (no joke).

    If you do the seventh circle in 3 days or fewer, I will send you a chess book of your choice (under $50).

    Note, this only applies if your rating doesn't dip by more than 100 points in those three days. I could go through 10,000 problems in 3 days and get them all wrong. :P

    I know I set the bar high, but heck, that's a lot of money!

  4. Blue,
    If you do the seventh circle in 3 days or fewer

    That's a strange fixation of you. Since the problem window is ever shifting, there can't be no 7 circle.
    The time constraints of DLM are arbitrary. So I use my own:

    1st. a pattern is assimilated in your brains when the problem is played a tempo. In CTS-terms: < 3 seconds.

    2nd. Doing problems must generate JOY. Look what happens to Dread otherwise. Doing 10,000 problems in 3 days is, besides impossible, NOT FUN.

    So the gauntlet is not picked up.

  5. :)

    It is an American put-down, often used amongst friends. It means 'Wimp' or 'Coward', to use two inoffensive synonyms. :o