Every move has its logic
Now I have a method to transform knowledge into skill, it stands to reason that I have a closer look at in which areas I can use the method as well, beyond the obvious.
The obvious areas are:
- Opening
- Middlegame
- Endgame
- Tactics
- Positional
- Strategy
And I have installed a daily training regimen for every obvious area already. Observations of my own games, and the playing of grandmasters have shown me there is another area as well, though.
Every move has its own logic. That logic is often very trivial, since that logic is often why we move a piece in the first place. To bring a piece out, to protect a pawn, to restrain an opponent's piece et cetera. What I noticed though, is that not all simple logic is seen by system 1. Not all simple knowledge is transformed into skill. It is more like learning a second language, where you often need to use a dictionary or have to google how an expression is used.
The latest discovery is the change in order during training. For the actual skill forming, it is necessary that system 1 takes the lead. System 1 sees the salient cues, while system 2 follows by verbally speaking out loud what the move actually does and why it is played. I'm experimenting with different formats, but it seems that looking at analysed master games comes very close.
Only when the knowledge of even the most trivial move is absorbed as a skill, you can hope that you can visualize a line into a decent distant future. Trivial knowledge that is not absorbed, can not be visualized with accuracy.
The test is simple: can you visualize it, then you have absorbed it. When it takes time, you know that system 2 is at work and not system 1.
The recent example I gave (the LINK to the YouTube video of the Khalilbeili-Lipnitsky 1954 game) seems to follow the approach you outlined above. During the opening moves, the narrator tends to give a simple reason for each move, with very little in the way of alternate variations. After White’s 16th move, he then goes into significant analysis of WHY Lipnitsky played what he did. FWIW: the name of the YouTube video is How to Find Greatest Chess Moves: Secret Strategy and Tactics.
ReplyDeleteTemposchlucker wrote (in part):
The latest discovery is the change in order during training. For the actual skill forming, it is necessary that system 1 takes the lead. System 1 sees the salient cues, while system 2 follows by verbally speaking out loud what the move actually does and why it is played.
Perception MUST occur first – if you can’t SEE the salient cues, you can’t train the necessary level of pattern recognition. Without perception, you are limited to surface-level analysis and can never penetrate to the essence of the position.
One aspect of perception that is ignored is the surrounding context for the cues. Emphasis is placed on the positive image (the active agents; foreground) with little (or no) mention of the surrounding context, which plays the role of negative space (background). Why? Because the surrounding context is passive and unnamed, and logical step-by-step training is focused on the positive images. After recognizing the context, the positive images tend to stand out much more easily in any given position.
Totally off-topic: Do you use the Logicly app to produce that diagram or did you copy it as-is from the Logicly site? Seeing that logic circuit diagram took me WAY back in time: I cut my professional electronics technician chops (11 years in the field) on digital/analog circuits used in flight simulators. My most ambitious and successful project was single-handedly redesigning a twin-jet utility aircraft (Lear jet) simulator to fly like a B-52 for the USAF. That was prior to learning trigonometry and calculus in engineering classes.
Usually, system 2 must PREPARE the material first before you can present it to system 1. That's why an author often says "you can pause the video now".
ReplyDeleteWhich is an invitation to system 2 to produce its word salads.
By simply telling what is going on, the author has done the preparation, giving the student the chance to absorb the cues.
This invites a passive approach in the student, though.
So somehow the student must engage in ACTIVELY seeing the salient cues and describing them verbally.
On the one hand we must avoid losing time by PREPARATION. On the other hand we must engage ACTIVELY in seeing the salient cues and describe them.
I spend 23 years to PREPARE a method how to acquire a skill. This was done by a combination of observation and thinking. After these 23 years I had only a THEORETICAL method how it must work. Logical thinking is destructive by its very nature. It can only tell you what does NOT work. You have to experiment (fiddle around) to be creative. To generate new things that MIGHT work.
After that, it took me 1.5 year to convert the THEORETICAL model into a PRACTICAL one by experimenting with it and seeing what works.
So we must AVOID the preparation BY OURSELVES as much as possible, by making use of the preparation of other people (or Stockfish, for that matter) as much as possible, and we must ACTIVELY copy what they have prepared.
I just copied a picture from the Logicly site by chance. I used to have electronics as a hobby when I was eleven.
ReplyDeleteThe video stresses the importance of the INITIATIVE as a way to prune moves. We must either find a way to maintain the initiative (CCT) or accept the position as it is.
ReplyDeleteThe initiative doesn't guarantee that there is a way forward. It is just a matter of trial and error. What we know for sure is that NO initiative means that there are no tactics.
I propose to convert that to "no salient cue, no tactic". Since there usually are way too much CCTs that are just redundant, and you just simply can't ponder them all.
If looking at this kind of videos is beneficial is a matter of experimenting with it, avoiding unnecessary preparation an avoiding a passive approach.
ReplyDeleteMy bet would be on actively fiddling around and describing what you SEE.
Somewhere I mentioned that 30 reps is the minimum and you quoted somewhere that 50 reps is normal. That sounds about right.
PART I:
ReplyDeleteI spent a lot of time yesterday and last night pondering the ramifications of your recent posts.
A lot of the KNOWLEDGE I’ve acquired over the years turns out to have a somewhat different meaning when I examine it again and reformulate it in terms of pattern recognition (PERCEPTION).
Consider the following excerpt (a repeat from a comment I made back in 2016):
GM Rowson (Chess For ZEBRAS) repeated something from GM Nigel Davies:
In case neither of these two distinctions [between "knowing that" {KNOWLEDGE} and "knowing how" {SKILL}] means much to you, I came across a third saying much the same thing, in an article on Chesscafe.com by Nigel Davies called 'The How and the What'. Extracts are copied below with the author's kind permission: "I recently saw a newsgroup discussion about tournament preparation. Everything under the sun was mentioned from openings to endings and strategy to tactics with everyone having their own idea about how it should be done. I just commented that 'the how is more important than the what', leaving anyone reading this guessing as to what I meant. In fact the comment was deliberately enigmatic ... It really doesn't matter what you study, THE IMPORTANT THING IS TO USE THIS AS A TRAINING GROUND FOR THINKING RATHER THAN TRYING TO ASSIMILATE A MIND-NUMBING AMOUNT OF INFORMATION. In these days of a zillion different chess products, this message seems to be quite lost, and indeed most people seem to want books that tell them what to do. The reality is that YOU'VE GOT TO MOVE THE PIECES AROUND THE BOARD AND PLAY WITH THE POSITION. Who does that? Amateurs don't, GMs do...
PART II:
ReplyDeleteThere are several important “hints” for PROPER TRAINING. The most important is your concept of “fiddling around.” But toward what end are we “fiddling around”? Just shuffling the pieces around the board using trial and error is highly unlikely to produce any appreciable gain in SKILL. The “clue” in GM Davies quote is in THINKING while you’re “fiddling around”—intentional and goal-directed rather than at random.
Unfortunately, due to the ambiguous nature of word salads, we often attribute an unintended meaning to the words, thus arriving at the wrong conclusions regarding the actual meaning. Since THINKING seems to be emphasized, I (for one) took that to mean that we should be following a logical step-by-step (formal?) thinking process. Ergo, look for as many “thought processes” as possible, determine what steps seem most applicable to me and then memorize that process. VOILA! We should then “think” in a structured way, just like the Grandmasters! FAT CHANCE!!
I’ve now come to a radically different viewpoint. “Playing with the position” means doing what you have described: perceive the patterns that are inherent in a position (System 1) and then formulate and verbalize (System 2) the pattern(s), what connects those patterns, and the context within the specific position. In short, SEE what surface level features exist, and then “fiddle around” to SEE what various combinations of patterns can be uncovered and thus connected together. That process will automagically ingrain specific patterns into long-term memory AND connect various pattern together into larger chunks over time with repetition.
Ideally, there should be some optimum presentation of training material (example positions) that would minimize the time it takes to ingrain sufficient patterns and interconnections to enable SKILL at a much higher level. A lot of tactical training material is oriented via criteria that are random at best. For instance, categorizing by “Mate in X moves” does not take into consideration the basic patterns that require exposure and repetition in order to form chunks. Since there is no focused attention on specific patterns [say 30-50 repetitions of the same pattern], System 1 may not form long-term memory connections through reinforcement of specific patterns. Without focused attention on all aspects of a given pattern repeated many times, the long-term memory may not coalesce into SKILL that can be readily retrieved by System 1 in a new position. System 1 automagically “connects” various patterns into “recognition triggers” and potentially larger chunks—but only if System 2 explicitly emphasizes those patterns as being important for System 1 to remember.
While training, System 2 has to choose what to present to System 1. The easiest way to determine what patterns are “missing in action” is to study and observe what has been overlooked as we solve problems and play over games. Another “clue” is to play over annotated games and note (and study intently) the points in the game where the moves are “surprising”—to US.
As a recent example, take the position in the Khalilbeili-Lipnitsky 1954 game after White’s 16th move. I “SAW” the possibility of “attacking” White’s Queen with 16… Ba6; it’s fairly obvious although it looks nonsensical because it just throws away a piece. What I did NOT “SEE” was the concerted action of the Black knight, Bishop and Queen in a potential mating attack or, barring that, a significant gain of material. As soon as I saw 17. Qxa6, I then “SAW” the potential mating attack. THAT is what I emphasized as I continued to study that game, repeating the moves again and again, trying to determine if there were other alternative patterns that had not become “obvious” to me.
You can’t PLAY what you can’t SEE. You can’t SEE what you haven’t trained System 1 to SEE.
Imagine that you are a non native English autor, just like me. How can I surprise you? With the word "serendipity". I know the word quite well, but I always forget what it means. Just like you, probably. Hence I use the rarity of the word to surprise you (I had to look up rarity, since I was struggling with "rareness", which didn't sound quite appropriate to me).
ReplyDelete"Rare" means with a low frequency of occurrence. That has as a consequence that I will not have often the chance to surprise you with "serendipity".
If I want to have a greater chance to surprise you, I must find words that have a higher frequency of occurrence, yet cause me trouble. Inferring that what causes me trouble, probably will cause you trouble.
Hence I advocate to have a closer look at everything that happens BEFORE move 16. Have I really absorbed everything that happens before move 16? My latest findings seem to indicate there is still a lot to gain in that area which we tend to dismiss as "too trivial".
Of course this is not a plea to NOT study what happens after move 16. I missed the mate too (and saw the bishop sacrifice immediately too). The author of the video advocates starts with the unproven claim that there must be something in the position BECAUSE there is a move that keeps the initiative. Which is a silly claim, ofcourse.
ReplyDeleteWe must learn to SEE the mate. That is where it would start in an ideal world. From there we would work backwards, concluding that the queen prevents Nf3. Leading to the question "how to lure the queen away"
Preparation (=gathering the material): system 2
ReplyDeleteSeeing salient cues: system 1
Verbalisation: system 2
You are not ready after the verbalisation. You are ready when the verbalisation becomes obsolete.
In the end there is no such thing as a thought process. Because there is no time for that OTB. A thought process is only a temporary method for in the study room.
I think (pun intended) that when PLAYING (definitely NOT training), the focus of System 2 should be on becoming consciously aware of what System 1 SEES (recognized patterns) in each position and then connecting those nexuses of attention together (based on evaluation of various combinations of those patterns) as part of a logical Narrative that makes sense of the whole position.
ReplyDeleteThere is a certain amount of "fiddling around" involved to try to determine the relative best combination of patterns and the order in which those combinations should be translated into a move sequence. I suspect that SKILL is not just determined by the raw number of distinctive patterns known (as speculated by various cognitive scientists to be anywhere from 50,000 to 300,000) but also by the trained capability to clearly SEE the various connections and translate them to moves/variations. This subconscious process is NOT conducted at the level of individual moves, even though some of the cogitation contains some individual moves. Ultimately, the overall result of that “thinking process” results in individual moves.
The difference in SKILL is the ability to effectively enter into that cooperative process between System 1 and System 2 by conscious choice. Think of it as entering into a state of “flow.” As long as that state can be maintained, the player can utilize that skilled process essentially at will, accessing all of the patterns in long-term memory.
I recall one of my most exquisite games against an Expert-level player in which I played the Black side of of the Poisoned Pawn variation of the Najdorf Sicilian—with no memorized variations to rely upon. I was a safe piece ahead after nearly 30 moves. Unbeknownst to me (because I was totally in “flow”), a large crowd had gathered around the board to watch the game. Unfortunately, something (I do not know what) distracted me out of my concentrated state, and I became aware of the crowd. I was so startled that I immediately threw away the game. I couldn’t re-enter the state of flow and couldn’t SEE anything.
Sic transit gloria.
A L-O-N-G word salad follows.
ReplyDeletePART I:
Temposchlucker opined:
The latest discovery is the change in order during training. For the actual skill forming, it is necessary that system 1 takes the lead. System 1 sees the salient cues, while system 2 follows by verbally speaking out loud what the move actually does and why it is played. I'm experimenting with different formats, but it seems that looking at analysed master games comes very close.
A casual (or merely recent) reader of this blog might come to the cynical conclusion that after approximately 25 years, we have come full-circle back to the common suggestion of the masters for developing SKILL:
PLAY as much as possible and STUDY (annotated) master games.
That conclusion could not be more WRONG!
Consider the impetus for beginning this journey: a search for a method that is successful for continuing adult chess improvement. Let’s start with one of my favorite quotes:
The great chess teacher, author, and publisher, USCF Senior Master Ken Smith wrote, “Until you are at least a high Class C player [1500 – 1600 USCF rating], your first name is “Tactics”, your middle name is “Tactics”, and your last name is “Tactics.”
That advice is perfectly sound, yet totally unusable for training. There is no information regarding which tactical motifs/devices/themes should be studied, and zero information about HOW to study them so that the PATTERNS will be cemented into long-term memory and become accessible when confronted with new (not previously encountered) specific tactical positions while playing. It is as useless as GM Kotov’s advice to start “thinking” by identifying all the candidate moves in a given position—with no real advice for learning the criteria by which one chooses candidate moves.
So let’s go to the “maestro” of tactics training for more advice. Michael de la Maza advocated for the Seven Circles (of Dante’s Inferno): pick any (random) set of approximately 1,000 tactical puzzles, and go through them all 7 times, each time reducing the amount of time by half used to complete them on each cycle, ending on the last day by doing all of them in one day. Somehow, all that repetition will cement those PATTERNS into long-term memory and you will become a tactical wizard. After that, you can give some thought to strategy, the endgame, the openings, etc. Again, useless because there is no hint of how to embed those patterns into long-term memory. The result: no one (except MdlM, and that is debatable) who attempted to complete that process ever made the quick (within 6 months) and large jump in rating (100-500 rating points) that was supposed to ensue. That did not deter the authors of The Woodpecker Method in the least. After all, at least one of the authors is a Grandmaster!
Skipping over a large number of different approaches (ALL failing to produce the desired result; consult this blog from its beginning for the numerous attempts), we finally get to where we are now.
And, for the first time, we have not only a training “method” but the necessary DETAILS of putting that method into practice.
The devil resides in the details.
PART II:
ReplyDeleteOr, as GM Nigel Davies stated:
"I recently saw a newsgroup discussion about tournament preparation. Everything under the sun was mentioned from openings to endings and strategy to tactics with everyone having their own idea about how it should be done. I just commented that 'the how is more important than the what', leaving anyone reading this guessing as to what I meant. In fact the comment was deliberately enigmatic ... It really doesn't matter what you study, THE IMPORTANT THING IS TO USE THIS AS A TRAINING GROUND FOR THINKING RATHER THAN TRYING TO ASSIMILATE A MIND-NUMBING AMOUNT OF INFORMATION. In these days of a zillion different chess products, this message seems to be quite lost, and indeed most people seem to want books that tell them what to do. The reality is that YOU'VE GOT TO MOVE THE PIECES AROUND THE BOARD AND PLAY WITH THE POSITION. Who does that? Amateurs don't, GMs do…
Even that advice is enigmatic and incomplete. Merely tossing the pieces (using trial and error randomly) around the board in the training room will most likely NEVER produce significant improvement during PLAY.
Temposchlucker’s method is the only one that gives the details of HOW to go about following GM Davies advice while training.
It is also universally applicable to EVERYTHING (every aspect of chess) that is needed for improving SKILL.
As an example, consider the tabiya reached after the “standard” set of moves in the Najdorf Sicilian: 1. e4 (advancing a pawn into the center and opening a line for the WBf1) c5 (preemptively attempting to restrain White’s advance WPd4 and opening a line for the BQd4) 2. Nf3 (developing a Knight, adding pressure to e5 and preparing to castle kingside) d6 (contesting e5 and protecting c5, while opening a line for the BBc8) 3. d4 (contesting the center in spite of Black’s c5-pawn and opening up a line for the WBc1) cxd4 (gaining a center pawn (temporarily) while eliminating part of White’s center phalanx) 4. Nxd4 (restoring material parity and advancing the Knight into the center) Nf6 (threatening the e4-pawn, essentially “forcing” White to protect it) 5. Nc3 (a duplo move: developing another piece while defending the center e4-pawn) a6 (preventing White from playing a knight to b5 and waiting to see what White’s plan of attack will be based on his subsequent move). This is an example of what Temposchlucker meant when he stated that the reason for one or more moves may be trivial.
Upon reaching this specific tabiya, White must now choose a plan of attack. This will require choices as to where to place his pieces for attack and defense and the order of those moves. Black will then have to make his response based on the information (“clues”) provided by White’s choices. Neither player has totally free choices as to how to continue; each is constrained by what the other side does or does not do. The interaction of the various choices restricts the range of plausible responses.
PART III:
ReplyDeleteFor the middlegame, it is a matter of subtle (or blatant) attacks and defenses and the interaction between the various pieces. There are typical patterns to be learned that are position-independent, yet the only way we can learn those patterns is by exploring deeply approximately 30-50 examples that demonstrate that pattern (or connection of patterns). For example, a typical back-rank mate pattern has an extremely large of potential examples. However, once the basic pattern and the context of that pattern have been absorbed, System 1 can be triggered to SEE it in myriad positions.
For the ending phase, there are also typical patterns, many theoretical ones as well as practical ones. Again, these patterns (and ALL the considerations associated with them) are position-independent. For example, consider that Richard Réti explored a certain collection of ideas in specific King and Pawn endgame studies:
POSITION 1: Puzzle by Richard Réti (1921)
FEN: 7K/8/k1P5/7p/8/8/8/8 w - - 0 1
POSITION 2: Puzzle by Richard Réti (1922), based on Schlechter-Marco 1893
FEN: 8/1K6/8/3p4/2k5/P7/8/8 w - - 0 1
POSITION 3: Puzzle by Richard Réti (1922)
FEN: 8/8/k1P4p/8/K7/8/8 b - - 0 1
POSITION 4: Puzzle by Richard Réti (1928)
FEN: 8/6p1/k1P2p1p/7K/8/8/8/8 w - - 0 1
All four of these studies (which could conceivably occur in a game) are connected via the common patterns within them. I’m fairly certain (based on my own personal experience) that if you study these four endings using Temposchlucker’s training method, you will gain considerably in endgame skill.
The same type of training can be done with variations on the theme of the Trébuchet. [WHAT? Did you think I forgot about it? LOL!] Here’s an example [repeated] that (at first glance) has nothing to do with the textbook position given in most endgame books:
FEN: 8/5rk1/5Rp1/4K1P1/5P2/8/8/8 b - - 0 1
Interestingly, this position occurs several times in the Chess Tempo database, with White wins and Black draws (sorry, no Black wins). I saw it in a local game between two Experts; the Black player LOST the game. I merely glanced at the board, mentally noted the drawing line (based on the Trébuchet with NO calculation) and went on my way, patrolling as Mall Security. Imagine my surprise when I returned and found out that Black had LOST!
The point of this extremely long word salad is that:
THE METHOD WORKS IN ALL PHASES AND ASPECTS OF CHESS!
I‘ve been reading your blog off and on over the years and have always been impressed by your endurance and capability to come up with new training-approaches. Kudos! What you outlined in your recent posts resonates with me, with my experiences - I think you really found something here!
ReplyDeleteIn an attempt to understand the method you are suggesting I’m trying to rephrase it in my own words - I‘m not trying to say anything new or different - my intention is just to summarize your method as I understand it.
1. The starting point is a suitable training position, illustrating typical and relevant points about a training theme, e.g. a specific opening or a tactical theme.
2. You write about PREPARATION in your posts and it seems you mean by this just the selection / collection of suitable training material - if that is right then PREPARATION is the same as step #1. (However, sometimes it sounds as if there is more to PREPARATION than just selecting positions?!)
3. Looking at each training position the student needs to ACTIVELY see/perceive on his own the relevant cues of the position. Put differently: in the particular position the student has to DISCOVER the cues on his own. While looking in this active-discovery-mode at the position the student verbalizes in detail which cues he is seeing and what appears to be relevant about them in this particular position.
4. The conscious perception of the cues plus some fiddling around with them then sparks the solution to the positions (e.g. a tactical move sequence).
5. Repeat steps 1,2,3 for a specific position until verbalization becomes obsolete/automatic.
For a training theme like a tactical theme or a type of endgame etc. you estimate it will need 30-50 positions. So steps 1,2,3,4 would then be repeated for 30-50 positions.
Does that capture it? Where did I get it wrong? Anything missing?
Merry Xmas to you and folks here on the blog!
-Steven
Ad 1 and 2. Preparation is everything before the forming of skill by system 1 begins. You can save time by outsourcing this stage to a grandmaster and Stockfish.
DeleteAd 3. Ok. It is about seeing the cues and relating them to the associated logic.
Ad 4. The fiddling around is about getting the details clear by asking "what if?". You need absolute clarity. No gm can do that for you.
Ad 5. Exactly. In stead of verbalizing the logic, you simply SEE it. Which makes the verbalization obsolete.
About the repetitions. They are not a goal in themselves. If you focus well enough, one time might be enough. In practice, it takes time to dig deeper into a position.
DeleteFurthermore, it is entirely possible to repeat without getting deeper at all. So be forewarned!