Takchess Chess Improvement

A Novice chessplayer works to get better at chess using an improvement program based upon the methods of Michael de la Maza and the teachings of Dan Heisman

Thursday, September 28, 2006

Ramblings on Pattern Recognition and Calculation

As I end my first circle of the 1200 CT-ART3 problem set, I am reading parts of MDLM Rapid Chess Improvement book. I am reminded that MDLM 7 Circles is a program of BOTH Chess Calculation and Pattern Recognition. The first 3 circles focusing on calculation while the second three on pattern recogition.

I have tailored the MDLM circles to my own temperment where I don't want to spend alot of time calculating. I had believed that I could through sheer pattern recognition plow through all these problems and seeing the answers develop strong tactical skills. In fact, I'd like it if the problems were in a Motion Picture format where I could just see an endless loop of all the tactics (removing my calculating brain from the process)

However I now feel that this is not the best methodology. Haste makes Waste. Chess improvement relies upon the combined strength of Pattern Recognition and Calculation.

On my first circle of the entire set, I am making my moves as if I am playing Blitz. This doesn't lead to good recognition but does help me with building Momentum which is very important to me. In the second set I plan to slow down especially where it comes to making the first move.I would like to soak in the original pattern.

I am interested to see if having done Level 90 problems has any effect on my Level 30 problems.This is a departure from my doing minicircle of a smaller easier set.

Last year when I did the Level 20, 14 times I had trouble building pattern recognition. I think this is because I moved too fast trying to solve the problems. Part of the pattern recognition is not only in the moves themselves but in recognizing the initial position structure . If I repeatedly made a wrong first move move, I may of mentally reinforced a bad pattern. Then next circle I will strive to slow down initially.

A benefit of repetitive tactical training is that it makes one calculation process less verbal and more visual thus more faster. My past level 20 training has made me more aware of the type of tactical solution it will be. Whether it will be a backrow mate, a kings march, a push of the opponents king into a wall of my pawn , a sacrifice causing a stalemate, pawn promotion. Sometimes I see a position and immediately see my opponents queen has Limited mobility and look to trap it.

More of this later as I move through the circles.............
Friday -1032


  • At 3:50 PM, Anonymous Anonymous said…

    I tend to get quite stressed when solving problems with CT-Art and try to do them in warp speed. I guess I'm more of a lying-on-the-couch-sipping tea-reading-book kinda guy. This is where I've found Combinational Motifs to be great, since it has almost exactly the same problems as CT-Art and it's harder to hurry through the problems without that easy click of the mouse. You won't get the speed, but you will get quality.

  • At 7:58 AM, Blogger takchess said…

    I have the CM book as well. I am not sure if I could follow it at the higher levels without a board. too many darn variations. I did once take out the book combinational art from interlibrary loan which had a number of the 5 X5 squares problems which I made flashcards of. Samuri did you formerly have a chess blog under Dragon Slayer or was that another Swede?

  • At 6:50 AM, Blogger Blue Devil Knight said…

    Since I just want to finish the damned circles, I've decided to stop spending lots of time on any given problem, to just answer within 2 minutes at the most (even on my first minicircle), and move on. While this isn't the best way to build calculation muscle, it will help me finish these damned things in my lifetime....

    I also have a hope that just learning the patterns will make me a better calculator. This is based on that whole 'chunking' thing from cognitive psychology: I can remember seven or so digits in working memory right after you tell them to me, but if someone tells me three dates with meaningful significance (e.g., 1492, 1776, 1789) I can easily remember 12 or more digits. The tactical motifs I am learning are becoming cognitive chunks that I can use as the basic units of calculation without having to calculate the details within those units.
    Of course, this doesn't mean calculation isn't extremely important, but to save time I have decided not to use my circles as a calculation exercise. I wish Quandoman were here: this is what he always recommended.

  • At 10:59 AM, Blogger Temposchlucker said…

    what happened in 1776 and 1789? (I even used copy and paste for the figures to prevent forgetting:)

  • At 11:14 AM, Blogger Blue Devil Knight said…

    1776: American Revolution
    1789: Ratification of the US Constitution.

  • At 1:21 PM, Blogger takchess said…

    I have the shared experience of knowing tactical training is what I need for improvement and the desire to get these d*mn things done. Quick first impression play works fairly well at the lower level problems the Bains book and the Ct-art Level 10. The trouble is when I got to a harder level even the small change to L20. I did not get any real benefit until I started reviewing the answers to the ones I got wrong immediately after I got them wrong and doing problem sets of just the ones I had difficulty with. Also when I made my first move there was almost no calculation at least less than 10 seconds. I never spent 2 minutes looking at a problem. Just wanted to see if I can do it on raw intuition. Please note raw intuition didn't work for me even on the 10th circle. So I have 100 more to go in my first circle of the entire ct-art3 then I
    'll spend some time scanning the position first before attempting to solve it. not alot of time but at least Some time 8)

  • At 7:44 AM, Blogger Blue Devil Knight said…

    I am really glad I am using CTB! One to three move mates and simple two-to-three move tactics.

  • At 3:07 PM, Blogger transformation said…

    i wrote likeforest extensively at his last post, and 'bob' also had many good comments, and basically said i thought that:

    it was a lot better to spend three months carefully and slowly studying the first ten games of chernev's Most Instructive Games of Chess Every Played (kind of an autodidact Simple Chess, Stean), rather than spend an hour a day to do thirty games of Nunns Understanding Chess Move by Move.

    what this has to do with takchess, tempos, blueDevils, and samurai’s discussion above is the idea of slowing down and doing it right, rather than rushing.

    ive done this too, but always regret it at CT-Art 3.0. i am not competing with any knights or trying to a RapidChess person in correct twelve step form, in compliance. i follow my own rhythm. i am not comparing myself to other CT-Art persons, and believe one circle slowly is better than two fast, as least to start.

    i plan to resume today, after a two week hiatus, and thirst for it, and at problem 476 have plenty more work, half way through level three, if that... so hard!

    fyi, i finally just hit 85.0% at CTS at 1500 elo, and wrote about this in context the last two posts not inextensively.

    nice to see you again Takchess. you are always so correct and so intelligent, i can always count on you for good protean. dk

  • At 2:51 AM, Blogger takchess said…

    I just a protean kinda guy 8)


Post a Comment

<< Home