Saturday, August 11, 2018

Analyzing the analysis

Today I could finally devote a few hours to chess. This is the plan I concocted.

I devoted 2017 to unearth the tree of scenarios and the PLF (PoPLoAFun) system. These two things tend to improve the quality of analysis. At the expense of the use of extra time, though. So at the end of the day I didn't get the result I was looking for.

The holy grail of chess improvement is the transfer of chess thinking to the unconscious.

There are quite a few area's where I managed  to make this kind of transition happen. When it does, the unconscious kicks in and works its magic. For this transition to happen, I needed somewhere between 50 hours and 25 years, depending on the area. Usually there is no relation to what you study and the result the unconscious will come up with.

I learned in about 50 hours to drive a car. The magic my unconscious came up with, was to estimate how fast I could go through a bend in the road just by listening to the noise of the engine. I certainly didn't study that. But it was the magic of the unconscious that I got for free.

If the thinking mind was up to the task of estimating the best speed through a bend in the road, it probably would come up with some complex math. I'm pretty sure, that my unconscious mind doesn't make use of math to do the job.

So what do we know about this miraculous transformation?

  • There is little to no relation between what you study and what the unconscious mind comes up with.
  • Only the effort of studying doesn't count. The study has to be effective. But what is effective when there is no relation between what you study and the result?
  • The direct manipulation of the unconscious doesn't work. Since we don't know how the unconscious works its miracles.
  • It seems that the miracles are worked automatically when the conscious mind isn't overwhelmed by complexity.
tree of scenarios and PLF (PoPLoAFun) are a great way to analyze a problem. But to prevent my conscious mind from being overwhelmed so easy, I must analyze the analysis. How can I simplify my analysis to the bone?

The most logical thing to do seems to make use of problems that I have already analyzed in the past. I'm going to re-post the posts from the past on the current date. Complete  with the comments and all. That way, we circumvent our addiction to chess positions, and can we focus on the method of how to come to the solution in stead of finding the very solution itself. Let's see how that works out. I wilt start with the post of Dec 13th, 2017.



    Sorry, but I have a quite long response in four parts.

    PART I:

    I will agree with you that we do not have DIRECT manipulative access to the unconscious mind (System 1). However, we DO have INDIRECT access and the capability to DIRECTLY influence what gets stored there, by using System 2 to control what is presented consciously for (potential) inclusion into the subconscious.

    Using your example of learning to drive a car: I have forgotten exactly what training process I used. However, I do know that I learned to drive a manual transmission tractor on the farm when I was only 7 years old. My Dad put me on his lap and demonstrated how to start it, how to shift the gears (4 forward speeds and reverse) and use the clutch and brake, and how to steer, as well as how to operate the various tool attachments. I don't remember how long it took to reach an acceptable level of proficiency, but I do know it was not very long – there was work to be done; I suspect it was much less than 50 hours. At age 9, I was driving a 3/4 ton 5-speed truck around the farm. By 16 years of age, I had already been driving several vehicles for some time.

    It starts with a certain minimum body of KNOWLEDGE. If you have no idea what the controls DO, then you have no grounds on which to base a skill. So, we have to start with that minimum knowledge: there ARE controls, and there is a FUNCTION for each control. Having a "coach" who can describe HOW to coordinate the controls is a time-saver. Else, a longer, perhaps MUCH LONGER, period of "trial-and-error" is required to acquire SKILL; in the mean time (hopefully) without adverse consequences (such as wrecking) which tend toward winning the Darwin Award but not the Skill Award.

    Your example of controlling speed through a turn is apropos. The SKILL arrived through conscious effort with feedback. It came as a consequence of trying (usually SLOWLY) to control the vehicle. First you became familiar with the individual controls. Then you progressed to trying to coordinate the controls to accomplish a goal (driving the vehicle under control). During that process, you start with gross motor movements (think of the herky-jerky "fits and starts" involved in trying to coordinate starting to move with a manual transmission while “slipping” the clutch). You get feedback FROM THE VEHICLE as you progress, which in turn "programs" the subconscious to perform whatever has been learned at a higher level of skill (smoother response).

    We read of "judgement" and ASSUME (perhaps erroneously) that this is the kind of evaluation usually made CONSCIOUSLY, as the result of a logical sequence of steps, arriving at a conclusion. There is an alternative way of making judgements - by “listening” to System 1.

  2. PART II:

    From Dr. Lasker:

    "[The chess student] must examine, he must do his own thinking and by conscientious work HE MUST FORM HIS OWN JUDGEMENT."

    "But it is not the multitude of examples that is instructive, for the multitude is confusing; IT IS THE METHOD WHICH CARRIES VALUE AS INSTRUCTION, and the method has been sufficiently illustrated above to be thoroughly intelligible. THE READER MUST NOW WORK BY HIMSELF SO THAT HE MAY ACQUIRE THE ABILITY TO APPLY THE METHOD HOWEVER THE CIRCUMSTANCES MAY VARY IN DETAIL."

    [Aside: this is one of many reasons why speeding through thousands of tactical exercies without thinking does not work for (almost) everyone: IT IS MORE CONFUSING THAN HELPFUL!]

    "This fundamental and universal principle may be briefly expressed as follows: THE BASIS OF A MASTERLY PLAN IS ALWAYS A VALUATION. To value, to judge, to estimate a thing, does not pretend to exact knowledge. But knowledge by estimate, by judgement, by valuation, though not exact, according to the principles of Steinitz is still an efficient guide for the master.

    What now is the reason for my valuation? Valuations again! . . True, in each instance the reason is simpler, more sure, more trustworthy than its consequence, but THE REASON OF A VALUATION IS ALWAYS ITSELF YET ANOTHER VALUATION. FInally, all my valuations originate from my experiences: my first losses and wins which gave me pain or joy; my first draws that called forth in me a variety of sentiments; my first analysis, which was crude and faulty. From then on I VALUED AND CONTINUED TO VALUE; AND WITH PRACTICE I BECAME CAPABLE OF MORE EXACT VALUATIONS. And FROM THIS ROUGH MATERIAL IS GENERATED, BY CONTINUED TRIAL AND INTELLIGENT CRITICISM, THE SERIES OF JUDGEMENTS BY WHICH THE MASTER ARRIVES AT HIS CONCLUSIONS.

    Please note that Dr. Lasker does NOT state nor imply that the ever-refined “judgements” are made logically by System 2. I submit that he is describing the process of intuitions into System 1 as a result of making good judgements.

  3. PART III:

    Let’s consider three totally different chess positions and see what (if anything) all three of them have in common. I’m fairly certain that you might be familiar with all of them. Try to “solve” them by the usual methods.

    POSITION 1: Puzzle byRichard Réti

    FEN: 7K/8/k1P5/7p/8/8/8/8 w - - 0 1

    POSITION 2: Game Ignatz von Popiel vs Georg Marco
    Monte Carlo (1902), Monte Carlo MNC, rd 1, Feb-03
    [ ]

    FEN: 7k/1b1r2p1/p6p/1p2qN2/3bP3/3Q4/P5PP/1B1R3K b - - 1 36

    POSITION 3: Game Borisenkov-Mezenev, USSR 1950

    FEN: 8/8/8/8/5k2/1p3pR1/bK5P/8 b - - 0 1

  4. PART IV:

    If your approach to solving these three positions is the usual "look for tactical themes/devices" and memorize them for future “pattern recognition,” I suspect that you will NOT find much commonality between them.

    Réti’s endgame puzzle (POSITION 1) is included in almost all general endgame books. It illustrates some important principles of endgame play, including a knowledge of “the square of a Pawn” and the non-Euclidean geometry of the chessboard (which is that the Pythagorean Theorem [C^2 = A^2 + B^2] is NOT applicable to the chessboard in terms of squares). So what does that have to do (if anything) with the other two positions?

    The second position is the final position of the game Popiel-Marco. White has piled up on the pinned Bishop (typical tactical response for attacking a pinned piece) and so Black resigned because he could not save it directly. But did he resign prematurely? Perhaps he should have looked at the entire board before throwing in the towel.

    The third position is from the game Borisenkov-Mezenev. Weteschnik does an excellent job in Understanding Chess Tactics in setting up the dichotomy of the position. If you analyze the two sides of the board INDEPENDENTLY, you will come to the conclusion that Black cannot win. In fact, it appears that White can win by advancing the h-Pawn, since he can get it moving with the help of the Rook while preventing Black from promoting his f-Pawn!

    The key takeaway is NOT just in “solving” the positions, and then moving along to the next set of puzzles. (Victor Henkin in 1000 Checkmate Combinations describes it this way: “You shouldn’t take ‘tactical medicine’ in large doses. Even the most beautiful combinations can set your teeth on edge if you swallow them with the greed of a pelican.”) More, more, MORE is NOT a recipe for successfully inculcating intuitions into the subconscious System 1!

    So what IS the commonality?!? We must learn to “look” at the ENTIRE BOARD, not just to an isolated local area. What is happening on one side of the board CAN have a huge influence on operations under way in a separate isolated area of the board.

    I assert that this demonstrates WHY the “hungry pelican” (or MdlM) approach to tactical training is significantly flawed. Rapidly running through a huge number of tactical problems will produce a “feeling” of familiarity, which is well-known in an aphorism: “Familiarity breeds contempt.” There will be no “Aha!” moments if you are rushing through a large number of puzzles to meet some arbitrary time constraint. There has to be time for contemplation of as many aspects as possible, finding analogies, commonalities and "rules" that can be internalized, not by scarfing up as many “fish” as possible in the shortest possible time!