Friday, August 05, 2022

Summary

Rating progress 

The both tournaments are over. What have I learned from the preparation and the execution? The tournaments themselves will deliver me a 15 rating points or so, I expect. But given the fact that I don't like to play against nervously moving agitated boys of 12 years old, so I try to draw them as soon as possible, and my two blunders when I was hacking two 2000 rated players off the board, I don't attach much value to the rating result. In this special case.

Switch of area of attention

During my preparation I made a major switch, due to the fact that I was hacked off the board in under the 20 moves twice. I analysed what went wrong and talked to my victors about the games. That convinced me that the area of concern is the realm of simple tactical themes and motifs. The ABC of tactics, so to speak. So I abandoned the opening preparation altogether. And even my endgame study was put on the back burner.

Stopwatch as guide

There is a brilliant method to measure whether a pattern is absorbed within system 1: measure the time. Do you see what the position is about in under 3 seconds: you have absorbed it. Are you slower: there is work to do. Be comprehensive. include for instance the administration of the wood. That way you never have to count wood during your calculations in a game. Include the possible defences. That way, a blundercheck will not cost you time, be aware of the tempi (CCT) et cetera. Don't be satisfied easy, only when you see everything in under 3 seconds, you are ready to move on. Otherwise you have to repeat the problem. Use spaced repetition.

Expectations

I talked a lot to my usually higher rated opponents. So I got an impression of what they saw and how fast and what not, during the game. I'm convinced I must be able to reach a rating of 1950 (now I'm 1700) within a year of two. Then we have a solid proof.

In search for the method

What I have not found yet is a method to transfer knowledge from system 2 to system 1. I contemplate the position, I write things down, in short I do all kinds of system 2 stuff and hope for the best. I aim to find a method in the coming period. In order to save some precious study time. One thing I'm sure about: don't add speed to your training!

"Strategy requires thought, tactics require observation." Max Euwe


15 comments:

  1. C'mon, guys, we are writing history here! We almost reinvented the Polgar method for adults! We have defined the area of concern with certainty: tactical motifs and themes. We have a method to judge whether a pattern is absorbed by system 1.: a stopwatch. We even have some sort of method to transfer knowledge from system 2 to system 1. The only thing left to do is to optimize the transfer system. The estimated 940 hours that we need to absorb the whole area of concern is too much. We must do a better job. Ideas anyone?

    ReplyDelete
    Replies

    1. I guess 940 hours to devote to obtain a solid level of tactical awareness is not that big cost (time investment). It is a bit less than 3 hours a day (165 minutes = 2h 45mins / per day) for a year (I calculate a year of training as 340 days, as at least 25 days we must have to extract - for any urgent situations like diseases, burn out, family events, etc.)

      Maybe the optimization is just the limitation of the BASIC motifs (themes) up the level of difficulty as well? I cannot imagine deacreasing drastically the number of hours that we need to devote to chess tactical mastery. Of course instead of 940 hours, we can try 200, 300, 500, 600 or 800 hours... to see what are the results (effects) when we limit the numbers of hours to these hypothetical (random?) smaller tests.

      Delete
    2. I don't consider 940 hours for learning the ABC as very much. It should be possible to do it within a year, as you said, and I intend to do so. What bothers me, is that I cannot explain what I exactly do in those hours. I transfer knowledge from system 2 to system 1, I can measure that, but I have no idea how I do that. With more insight in what I'm exactly doing, an efficiency upgrade might be possible. But now I know not how the transfer from system 2 to system 1 works at all.

      Delete
    3. I think it is the constant (!) change in some mini-processes, the ones you explain and describe. There are some types of positions I can recognize without thinking. Then I can use my pattern recognition and solve the puzzle up to 3-4 seconds (the puzzle that contains 2-3 full moves). Therefore when I play blitz games (especially 3+2, 5+0 and serious blitz games as 5+5, even if they are rapid according to lichess classification) I can see that some of the positions seem a way easier to play on (navigate) and when I am focused enough, I can exploit tactical opportunities and more and more often they are very similar to the positions I practice as Puzzle Storm.

      One of the idea that I really understood is the idea of "length (chain?) of captures". Another one is capture with check or make the move with a mate threat, attack the most valuable pieces on the board (Q or R) or if the pawns are advanced - think of the promotion on the pawn.

      As for now the biggest challenge to me is to find the quiet moves (no check, no capture), especially at the positions (puzzles) when there are various moves and many possible captures. And from technical point of view - my visualization is very bad, therefore sometimes I suffer at solving checkmate in 3-4 moves if they are based on knight moves and I cannot visualize them good enough to be able to solve it (even if I know what type of mate can be the final solution).

      Delete
  2. That improvement in chess is a lot about improvement in tactics we already know. We know since years that the speed is an indicator of the quality of understanding. For me, the big question is: how to transfer from system 2 to system 1.
    My guess is: Focus on your own blunders, see what there is to see ( tactical motivs, chuzhakin system..). Find the right thought process use spaced repetition and improve in visualisation / calculation

    ReplyDelete
    Replies
    1. I agree that it looks not new. What is important though is where the accents lie. I proved, subjectively for now, that I have found a way.to transfer knowledge from system 2 to system 1. What I studied in the study room, I got on the board during the tournaments. Not once, but many times. The problem is, that I don't work systematically in the study room. I do all kind of things to study, and I don't stop until the stopwatch tells me that I have absorbed the matter. It works, but it is not efficient. The absorption is not sufficient. It takes too much time. 75 minutes per problem, at average. Meaning that there are problems that take me a few minutes to absorb, and some take several hours before I got the essence.

      We certainly do need system 2 activity, for instance like the Chuzhakin system, in the study room. But we are ready when we don't need the Chuzhakin system anymore behind the board. We need a blunder check method (system 2) in the study room. But we are ready when we have absorbed it and don't need it anymore over the board. Not that we don't blunder check over the board, but we do it by seeing (system 1), not by thinking (system 2).

      Delete
    2. After solving the Puzzle Storm puzzles at lichess for 18 months, devoting (at least*) 450 hours to this activity, I recently took part in the tourney (a local one, 7 rounds, 12+2 time control) and won it with 6,5 pts out of 7.

      And it was one of the great feeling that I do not need to calculate a lot of variations to be able to SEE the tactics instead of CALCULATING these! The simple tactical lines (up to 2-3 full moves) I could see (spot) pretty easily and I could focus on any other parts of the game.

      * - I have solved some tactics as normal puzzles at lichess, some puzzles as Puzzle Race (very similar to Puzzle Storm, solving but with/against other players with the form of a car race). In the meantime I solved about 40K puzzles at Blitztactics (mostly the puzzles of 1200-1800 difficulty).

      Delete
  3. We tend to underestimate the importance of the simple things. The ABC. We are too easy satisfied. We have the feeling that we master it when we can reconstruct it at will. But reconstruction is a system 2 activity. We need absorption into system 1.

    Think of the priest who read a prayer each day for 25 years and still needed a paper to read it from. If he had taken one evening and made a serious effort to learn it by heart, he would have remembered it for the rest of his life.

    One serious effort is enough to ignite the transfer of knowledge from system 2 to system 1. But we are inclined to do effort more often but less serious.

    ReplyDelete
  4. Tempo said: "We certainly do need system 2 activity, for instance like the Chuzhakin system, in the study room. But we are ready when we don't need the Chuzhakin system anymore behind the board."

    I think that is wrong. When I was a beginner in chess, the adults told me: Look at every move before you make it: Is it save, can it be taken? I was following this instruction for years .. and now its completely automated, usually I just know if the piece is en prise. THIS is the proven method to get things into system 1.
    The actual thinking process should be always according to the present level of system 1.

    ReplyDelete
    Replies
    1. The blunder check is part of a more elaborate system, the tree of scenarios check. If you mate the king, you check if there is a square to which he can escape. You check all escape scenarios. You can do that by system 2, or you can do that by seeing it. (system 1). There is a limited amount of tactical scenarios. It can be fully automated. When something isn't automated, you need to do it consciously over the board. Which is a system 2 activity. Which is slow and error prone. The blunders I made during the tournament, were tree of scenarios blunders. Not leaving pieces en prise.

      Delete
    2. This "blundercheck" did become an automated awareness of unprotected pieces

      Delete
  5. I devoted 450 hours to solve puzzles at lichess - Puzzle Storm type. It is the timed puzzle sets and each next puzzle is harder then the previous one.

    From my calculations there is the following data:

    1) 6200 attempts called "runs" at lichess (A run is a set of puzzles you solved and finished either due to making too many mistakes or crossing the time limit)
    2) number of various days I solved these puzzles = 186 (there were some longer breaks, I have not solved puzzles regularly)
    3) the period I have been solving these puzzles = 18 months (started 27th January 2021, and I have been solving up to now (11th of August 2022)
    4) the total time devoted to solving these type of puzzles = 450 hours (average 25 hours per month)
    5) highest solved puzzle = 1948
    6) average score of highest solved puzzles = 1815
    7) average time per puzzles = 2,18 seconds
    8) average scores at best puzzles scores 48-54 (I can reach this score regularly each time since July 2021 - as long as I make at least 6-10 attempts/runs)
    9) my highest score is 61, and a few times (6-8 times) I got 58-60 score

    Let me know Tempo if want to comment this and if you have any questions. This way probably we can save you a few hundreds of hours (to test various ideas and approaches) as I already did that :)

    ReplyDelete
    Replies
    1. I advocate the Thai Chi approach. Slow and concentrated. Quality versus quantity. We need to know more about the process of transferring knowledge from system 2 to system 1. From thinking to seeing. For the moment it is a black box. I need analogies. How did I learn to drive a car in only 50 hours? That is incredibly efficient. What other examples are there for the system 2 to system 1 conundrum?

      Delete
    2. I'll repeat part of a comment that I posted back on Oct. 9, 2013:

      Cognitive research has shown that (regardless of what is considered to be a "chunk") that it takes at least 6-8 seconds of PERCEPTION, followed by up to one minute of ABSTRACTION, in order to establish a "chunk" in long-term memory (LTM). Any time shorter than that will NOT result in storage of a "chunk" in LTM.

      In Isshinryu Karate-Do, there are 8 empty-hand kata (forms). The sixth kata, Sanchin, is introduced between green belt and brown belt (approximately 2-3 years of regular training). It is the oldest known kata, and also (mechanically) the simplest kata. It is called the “Three Battles,” and symbolizes the harmonization of the conflict found between body, spirit (breath), and mind. It consists of slowly (at Tai Chi speed) performing a series of three techniques, repeated three times. The first three forward steps are simultaneous step-block-punch. Then there are three sequences of punches, performed while stationary. Then the last three backward steps are a sequence of three defensive blocks. Easy-peasy – right? WRONG!

      During my martial arts training (17 years), Sanchin was the hardest kata to perform correctly – because it is NOT just physical. To perform it, you must synchronize all three aspects of your entire being. The instructor watches you closely as you perform, introducing corrections by simply repositioning a body part, or “whacking” an incorrect body part to get it away from the wrong place/position. The body is used to gain access to the mind and the spirit (through the breath). The instruction is NOT usually verbalized; you have to use the vulture’s eye to “see” what the instructor demonstrates.

      The test of the kata is conducted by one or two instructors. While performing it, the instructors continually “test” the harmony of the student’s body (the correctness of the technique used for each physical movement), the mind (the focus and attention to minute details) and the spirit (the ability to breath correctly, maintain calmness while under attack and not get frustrated or distracted). During the kata, the instructors will “attack” the student’s body, pushing to see if the stance is rooted, punching and kicking to see if the breath is maintained, and overall, observing the student’s demeanor to see that there is harmony among the three elements. If there is any faltering of attention or frustration, you flunk the test.

      Perception must be trained for a long time in order to “see” the requirements. If you “get it,” eventually you can perform the entire kata while being totally unaware of the attempts at distraction. In martial arts terms, “it” does the kata; you don’t. What is “it”? I think it is System 1. It’s hard to explain (impossible to explain to anyone not well versed in martial arts of some kind) but in some ways, it is a moving form of meditation.

      The training appears to be concentrated on System 2 (logically performing a sequence of physical movements) but is actually aimed at training the whole person by “pushing” the results of the training into System 1.

      Delete
    3. I have done Jiu Jitsu and Judo for six years.

      The most important thing is to be not satisfied easy. I verbally name the tactical elements and imagine how those work together. I write them down. Most tactics contain about three three elements. Sometimes that are moves, sometimes series of moves each aimed at a particular goal. The verbalization helps to keep the mind from wandering around.

      It is amazing to see how bad I am in simple tactics. You always know there is a tactic, but often my mind still denies it. Everything is properly covered, how can there be a tactic here?

      Often you hear the "aha's" and the "wow's" being yelled through the room. Close study reveals the beauty of chess!

      Last week, my mind shifted focus for a minute or ten. In stead of a sequential repetition of the three elements a combination exists of, I saw the pieces work together. All of a sudden the combinations were easy to see.

      I reckon that's the direction.

      Delete