Saturday, July 08, 2006

Payday

Update
Please all give a warm welcome to another Knight getting his breech on: Daland.
May his rating rise to such heights that ICC gives him a free membership!



At the club I'm crushing opponent after opponent. Even much higher rated players are crushed or drawed. Yesterday I played two G60 games against an opponent which had the same rating as me a year ago (1710). I always scored 50% against him.
But last friday I had difficulty to belief that we were ever of the same level. It was clear I had much more overview. He just couldn't see all simple tactical buildingstones that made up a complex position. He was blown away two times.
I even let him take a move back because he would lose his queen (it wasn't for a competition, it's vacation time now). To no avail.

Where does this sudden improvement come from?
Since september 2005 until now I have done only two things: CTS and pawnendings.
Because I don't win by grinding out pawnendings, the improvement can only be attributed to CTS.

Why seems the improvement to concentrate in the last two months?
I belief that that has to do with closing the gap between screen vision and board vision.

So what have I found?
A. Improvement at CTS pays off in OTB play (which was a major worry).

B. It is possible to decompose complex positions into a lot of simple tactical elements. Learning these simple tactical elements improves your overview in complex situations.
To give an example: if a position consists of 10 tactical elements and you use 3 seconds per element, you will have an idea of the position within 30 seconds. Since all elements are handled by longterm memory, mistakes are not very likely.
On the other hand, if you have to think a little about every element and it takes you 10 seconds to process the data (the pre-CTS state of a pattern, so to say:), you wil need 100 seconds to get an idea of the position. And you will have to make use of the short term memory for the thinking part. The short term memory will be stretched to its boundaries with 10 elements.
Hence it is more prone to errors.

C. CTS is a steady and sure way of improvement but it is a hell of a job.

CTS indicates the direction where to go, but can a more efficient method be invented? The first idea is to limitate the # of problems to make more use of the system of spaced repetition. Any other idea's?

8 comments:

  1. That's great that it's paying off. But come on, when there is nothing else left those pawn endings are useful. Maybe you need to start playing tougher people!

    ReplyDelete
  2. very cool that you have done so well with cts! how many hours a day do you invest in cts (on average)? i wouldn't mind emulating your results. :) hope your crushing continues!

    ReplyDelete
  3. King,
    I see what you mean. In middlegame I've reached an intermediate level. In endgames I'm still a novice. Within a few years I will be there at intermediate level too. Until then I don't mind to win from my opponents by crushing them only. Stronger opposition flows natural from a higher rating.

    ReplyDelete
  4. GK,
    sofar I invested about 600 hours in CTS. I see it as an experiment. So it's not a bad idea to analyze what is exactly beneficial at CTS and what part is only a waste of time. How does it work?

    ReplyDelete
  5. Tempo, if I recall correctly you said you completed Polgar's 5334. How much improvement did you see from mastering the mate-in-two problems? It's a big undertaking and I'm trying to decide whether it's worthwhile, or I'm better off re-focusing on CT-ART. I liked his mate-in-one problems.

    ReplyDelete
  6. Likesforests,
    you should do them until problem 1470 at least. Because that are very common patterns you have to know. After that the compositions start. These have patterns that are less likely to appear in games, but they force you to calculate. You have to improve your calcution skills to the limit. That limit is not so far away. At a certain moment you will find it is not longer possible to make progress with your calculation skills. I reached that limit after about 3 years of training.

    The possible improvement on the pattern recognition front is limitless though. The amount of patterns is limitless and the storage capacity of your long term memory is limitless.

    The variables to play with are:
    The quality of the patterns.
    The probability you will get them in your games.

    Calculation skills are trained at the tougher levels of CT-art too.

    ReplyDelete
  7. "sofar I invested about 600 hours in CTS. "

    I would attribute your rapid improvement to this massive bulk of time, rather than to any uniquely advantageous features of CTS.

    If you had spent 600 hours reading Nunn and Bronstein books, i think you'd improve equally fast.

    (just my opinion)
    congrats and may the crushing continue! :)

    ReplyDelete
  8. Patrick,
    nope. I did just that (reading books) in the past for 1100 hours during a few years and though my understanding of the game has grown, it was to no avail when it comes to better play.

    ReplyDelete