Sunday, April 28, 2019

Trial and error rehab

It is not that there is nothing happening in my chess realm lately. It is more that I lacked the energy to write about it.

I adopted three new openings, for instance: The London System with white, The classical dutch and The Sniper with black.

I have become webmaster of the local chess club, and I have written two long articles on chess tactics in dutch.

I'm asked to developed a tactical course for adults by my local chess club. Which is good thing, since I am now obliged to reformulate what we have found sofar. While trying to do so, I am forced to iron out all inconsistencies in the theory.

It would be helpful if you comment on my reformulation of theory. Please don't spare me if you find contradictions in my musings. Expect some controversy. Take for instance the title ;) So, in no particular order:

Tackling the bulk consumers of time
In order to get better we need to tackle the bulk consumers of time. To name a few:
  • Tunnel vision
  • The flight of the vulture
  • System II
  • Repeating trial and error
  • Thought process
  • Redundancy in thought processes

Tunnel vision
There is nothing wrong with tunnel vision. As long as we enter the right tunnel. When we have entered the wrong tunnel though, there is no clear way to get out. At least, I haven't found a method which isn't time consuming in its own right. Stopping the mind when entering a dead end should be a task of system II.

The flight of the vulture
The flight of the vulture was designed, to prevent tunnel vision. There are two problems with this:
  • The flight is time consuming by itself
  • There is no guarantee you don't miss important details
  • It doesn't help against bias. You see what you expect to see

System II
System II is a big spender when it comes to time. We must transfer as much tasks as possible from system II to system I

Repeating trial and error
When trial and error doesn't lead to a result, we tend to repeat it. The stopping of endless repetition of trial and error is a typical task for system II

Thought process
Since a thought process is typically powered by system II, it is notoriously slow.

Redundancy
If a thought process is too generic, it fails to cover a lot of special positions. Covering more positions can only be obtained by adding details to the thought process, which in turn are redundant for other positions. There is no workable "sweet spot".

System I should do the trick
When pondering all of the above, my conclusion is: adult progress can only be achieved by system I. System II adds reason and understanding, but it is way too slow.

The conundrum to solve is the following:
How can we educate system I by system II, while both systems speak a totally different language, and work in a total different tempo?

I will elaborate on that in a following post.

6 comments:

  1. Btw, I started afresh with my training at Chess Tempo.

    ReplyDelete
  2. To say it different: At each puzzle where we fail or underperform was something we did not make right, did not see or did not understand ( quick enough ). A system of improvement should? give us tools to detect the reason for each failure and methods to overcome them.

    "To solve a problem" is (more or less) the same as "to understand the position". You prove that you did understand the position by making the best move. You understand the position by learning = become aware of the most important aspects. So you need to know a lot of pattern, recall them fast, keep them in your mind and explore the position systematically

    But until now: It seems that a tactics beginner benefit a lot from just solving the first 4000? puzzles no matter how? they do that. After that, a further progress becomes very very hard for an adult.

    It seems to be necessary to have something like "talent??" to be able to reach a certain level in tactics as an adult, just work seems to be not enough. "Talent" might be a matter of "trainig" in the childhood.

    ReplyDelete
  3. @ Tempo: I'm glad to see you continue your studies and your blog posts. I hope Margriet is doing well.

    On a recent The Chess Improver (see links on the right sidebar on this blog) titled Problem Solving, Hugh Patterson gave a general "streamlined method" for solving problems for adults. I noticed immediately that the gist of his suggestions for a "thinking process" followed the general pattern that Tempo, Aox (and others) have suggested on this blog: (1) Identify the goal of the problem. (2) Use the instructions for the problem ("advantage:, "Decisive Advantage", "Play to Win") as a clue to what to look at first. (3) Identify who is to move - this is a clue as to what to look at first in conjunction with (2). (4) Identify critical squares and LoA associated with the goal. (5) Look at the entire board and consider bringing new pieces/Pawns into the solution. (6) "Use a logical sequence of steps and look at the entire board."

    On a different (but related) front: I just finishing reading Erik Kislik's book Applying Logic in Chess. It's somewhat repetitive and meandering, but there is a lot of good, specific ideas for chess improvement in it. IM Kislis is also a well-known trainer. He asserts (without proof, similar to Dr. Lasker's claim about teaching based on a 200-hour program) that anyone who studies appropriately can reach national master level, or perhaps even IM level. There are concrete suggestions for HOW to go about improving as a child and also as an adult. He also addresses the "talent" question.

    Here's a link to Amazon, where you can read some of it:

    Applying Logic in Chess

    ReplyDelete
    Replies
    1. There is a "new" form of tactic puzzles at chess.com:Puzzle Rush you have to solve as many puzzles as you can in 5 minutes. My peak performance so far is 28 thats about 11 sec per puzzle. Some GM's have 59 thats 5 sec per puzzle AND the last (59-28) puzzles are more complicated, the puzzles are getting harder as you solve them.
      Again: as better the player as better the performance in this type of training. There is no time for logic an thinking..you let the position sink in and wait for the hand to do the move.
      It might be possible that Applying Logic in Chess can improve a player somehow ( maybe it helps to generate chunks?? ) but the difference between good and bad players is the "thinking speed", the instinct, the reflexes, the subconcious.

      Delete
    2. I have just bought a premium account at chesscom and I have been testing the tactics puzzles pretty much. I do Puzzle Rush, but only after the tactics training. It looks like it is a way better solution to the problem with tactics.

      Nowadays I have big problems to solve the puzzles even though I can solve some of these quite fast. My present level is about 28-30. We will see how it is going to change in 4-6 months.

      Cheers, Tom

      Delete
  4. Re: thought process, during a game I find it most helpful to think of it as a structured reminder of what questions should be asked with each move, rather than trying to make the process include a flowchart of all meaningful possibilities. This is a somewhat different concept qualitatively than the type of thought processes you have been working on for purely solving tactics.

    For example, the most important questions for me are: 1) what did my opponent's move change about the position? 2) are there any CCT that (now) work for either him (as a threat) or me (as an opportunity)? 3) in the absence of immediate forcing tactics, what opposing targets (weak squares, loose pieces, etc.) can I work towards exploiting? 4) how can I best improve my pieces?

    It's worth noting that pattern recognition (system I thinking) can be built via exposure to tactics in an explicit learning process - which for beginners is going to use system II thinking, by definition, since everything is new and requires conscious thought to understand. Once the brain is able to categorize/recognize patterns better, then it can switch to calculating how to reach the known favorable pattern, which is a far more efficient process overall. It's also basically how Carlsen describes his thinking process at the board - his intuition (system I) tells him what should be done, then he calculates how to do it in that particular position.

    ReplyDelete