- Blind mode tutorial
lichess.org
Donate

One half-move chess intuition. How to teach that?

#1

No unique answer expect. Be polite to each other. Try to see the others points of view. Be magnanimous of expression levels.
Assume that there might be interesting partial points in a debate across the debaters. Schools of thought are acceptable, as long as their axioms are allowed to be discussed as well.

How to teach or train the 1-ply intuition over a diverse set of positions one could be expected in a full life of chess games from any possible walks of chess life?

Also, maybe through debate, and muffled egos (put the ideas first), and cross-talk, imagination and logic, some parts can be brought together into a new point of view. or compatibles set of points of view that could lead to improved intuition training and teaching (and one of my own favorite, cohesive and robust chess foundational theory, but that is a muse).

No unique answer expect. Be polite to each other. Try to see the others points of view. Be magnanimous of expression levels. Assume that there might be interesting partial points in a debate across the debaters. Schools of thought are acceptable, as long as their axioms are allowed to be discussed as well. How to teach or train the 1-ply intuition over a diverse set of positions one could be expected in a full life of chess games from any possible walks of chess life? Also, maybe through debate, and muffled egos (put the ideas first), and cross-talk, imagination and logic, some parts can be brought together into a new point of view. or compatibles set of points of view that could lead to improved intuition training and teaching (and one of my own favorite, cohesive and robust chess foundational theory, but that is a muse).
#2

"Things Silman says!"

"Favorite axioms of Pandolfini!"

"Things Silman says!" "Favorite axioms of Pandolfini!"
#3

What is "1-ply intuition" ?

What is "1-ply intuition" ?
#4

I hope it works better than kleenex.

I hope it works better than kleenex.
#5

People need to know where they are going; the goals. Without the goals and the ideas on how to achieve the goals, training people by feeding them positions and the candidate moves (the 1-ply intuition I think you are talking about), will not work. Everyone knows this, so no teacher would teach chess (possibly teach anything) in this manner.

The old masters taught chess by teaching, in order, the rules, endgame, middlegame, and then opening. There are key positions and key ideas in each of these. Tactics was taught in each phase. It is important to realize here that the key ideas are much more important than the positions being used. This isn't to say that there are not good exemplars; there are. But if people are to be able to recognize in a position in front of them that is not exactly the exemplar that the same goals and same ideas work, they have to pay attention to the ideas more than the exemplars.

However, very few people want to learn chess that way, because they are eager to play the game. That leads to a number of problems. It also leads to teachers teaching rules-of-thumb, which, by definition, are going to have exceptions. In chess, these have many exceptions.

As an example of what exemplars are used in one setting, consider KPk and compare Dvoretsky and Silman's exposition. There are just a few exemplars really needed in KPk. Dvoretsky's Endgame Manual has just 4 positions for that endgame. However, Silman's endgame book has over 20 positions on the same topic. The answer to why there is a difference here is, I think, obvious - people are different!

People need to know where they are going; the goals. Without the goals and the ideas on how to achieve the goals, training people by feeding them positions and the candidate moves (the 1-ply intuition I think you are talking about), will not work. Everyone knows this, so no teacher would teach chess (possibly teach anything) in this manner. The old masters taught chess by teaching, in order, the rules, endgame, middlegame, and then opening. There are key positions and key ideas in each of these. Tactics was taught in each phase. It is important to realize here that the key ideas are much more important than the positions being used. This isn't to say that there are not good exemplars; there are. But if people are to be able to recognize in a position in front of them that is not exactly the exemplar that the same goals and same ideas work, they have to pay attention to the ideas more than the exemplars. However, very few people want to learn chess that way, because they are eager to play the game. That leads to a number of problems. It also leads to teachers teaching rules-of-thumb, which, by definition, are going to have exceptions. In chess, these have many exceptions. As an example of what exemplars are used in one setting, consider KPk and compare Dvoretsky and Silman's exposition. There are just a few exemplars really needed in KPk. Dvoretsky's Endgame Manual has just 4 positions for that endgame. However, Silman's endgame book has over 20 positions on the same topic. The answer to why there is a difference here is, I think, obvious - people are different!
#6

#1 "How to teach or train the 1-ply intuition over a diverse set of positions one could be expected in a full life of chess games from any possible walks of chess life?"

I suggest that to get an idea of the difficulty in answering this question that initially the question be limited to a simple case. The question I suggest is:

How to teach or train the endgame of KPk such that a person so trained can be given any KPk position and play the correct move without look ahead turn-by-turn calculation?

Playing a correct move without look ahead turn-by-turn calculation is what I assume was meant by "1-ply intuition".

That endgame has been extensively analyzed and there are many references for how to teach it to people and computers.

#1 "How to teach or train the 1-ply intuition over a diverse set of positions one could be expected in a full life of chess games from any possible walks of chess life?" I suggest that to get an idea of the difficulty in answering this question that initially the question be limited to a simple case. The question I suggest is: How to teach or train the endgame of KPk such that a person so trained can be given any KPk position and play the correct move without look ahead turn-by-turn calculation? Playing a correct move without look ahead turn-by-turn calculation is what I assume was meant by "1-ply intuition". That endgame has been extensively analyzed and there are many references for how to teach it to people and computers. - "Endgame Manual" - Dvoretsky - "A guide to chess endings" - Dr. Max Euwe and David Hooper - "Silman's Complete Endgame Course: From Beginner to Master" - Jeremy Silman - King and pawn versus king endgame: Wikipedia https://en.wikipedia.org/wiki/King_and_pawn_versus_king_endgame - Chess Programming Wiki: KPK https://www.chessprogramming.org/KPK - My study on Lichess: King and Pawn Endgames I https://lichess.org/study/kbSEH1nW
#7

Did I ask how to teach or train that intuition? yes.
Did I say that it precludes anything already done? no.

I would suggest that there are some blind-spots though. in teaching. And that just waiting for one individual to randomly accumulate enough games experience meandering one branch of lines, and always getting bogged down in every consecutive positions without being able to skip some ranges, to make distant associative memory between positions, without needing every time to go the slow way, ratchet read the variation with nose stuck to each and every ply.

The intuition has that strong ability to make statistical associations, why not use that. It does not exclude the other tools. To the contrary, for people like me, those other tools mentioned above provide the hope for some glue to use to make such associations.

Also, being immature, in needing early gratification at the 10 min scale (or less), i need to put my learning into game chunks, like puzzles, but then again puzzles are again only making short term association because of the ply consecutive requirement.

I wish of an integrative "Spot the difference" kind of ranking games with whatever insighful crafting or selecting of positions pairs, more experienced chess players can provide (or teachers, they don't always mean the same thing, and may have not overlap at all, experienced player not experienced teacher).

But that is only one idea. I am putting it out here. Just keeping on the Hendricks 2 Sillman polarization or dichotomy, with self-satisfied status quo, and sense of truth ownership, does not seem like progress on the teaching front...

Did I ask how to teach or train that intuition? yes. Did I say that it precludes anything already done? no. I would suggest that there are some blind-spots though. in teaching. And that just waiting for one individual to randomly accumulate enough games experience meandering one branch of lines, and always getting bogged down in every consecutive positions without being able to skip some ranges, to make distant associative memory between positions, without needing *every* time to go the slow way, ratchet read the variation with nose stuck to each and every ply. The intuition has that strong ability to make statistical associations, why not use that. It does not exclude the other tools. To the contrary, for people like me, those other tools mentioned above provide the hope for some glue to use to make such associations. Also, being immature, in needing early gratification at the 10 min scale (or less), i need to put my learning into game chunks, like puzzles, but then again puzzles are again only making short term association because of the ply consecutive requirement. I wish of an integrative "Spot the difference" kind of ranking games with whatever insighful crafting or selecting of positions pairs, more experienced chess players can provide (or teachers, they don't always mean the same thing, and may have not overlap at all, experienced player not experienced teacher). But that is only one idea. I am putting it out here. Just keeping on the Hendricks 2 Sillman polarization or dichotomy, with self-satisfied status quo, and sense of truth ownership, does not seem like progress on the teaching front...
#8

Intuition comes from experience.
Playing games and analysing lost games.
Study of grandmaster games.
Training tactics.

Intuition comes from experience. Playing games and analysing lost games. Study of grandmaster games. Training tactics.
#9

endgames training fine. sold to the idea. to build the mobility dimensions of each men on board. and start combining them, constructively with increasing complexity of those combinations (non-chess meaning, it still exist as an English word, i would say). I also think, that in teaching of chess or a lot of chess mentality, there are sacred cows of assumptions that are rarely revisited. One of them being that chess is a fully conscious look-ahead engine type of search. That is not either what Hendricks or Silllman say, but it still is pervasive in how stuck with consecutive ply didactic constraints everything is deliverd with.

I don't need to screw every screws in an "Ikea" furniture, to understand its model, and how it might look like. I will need to do it to make an actual instance of the furnitur. But i am a free thinker, and can view the whole, without having to worry about the 1,2,3,4,5. i can even look at many such models without having to make an instance of each. And even looking at each 1,2,3,....

I could make decisions based on those "static" models. (because i have internalized the screwdriver screwing scenette routine, seeing a screw with dotted into a dislocated version of the furniture, allows me to predict how the thing will look and what to do, without having even to follow the 1,2,3.

Other tenet: we all have the same type of brain, and cognitive set-up.. possibly all masters end-up in the same box, i don't know, but, at least in the adult non-chess world, and the subset of those that get an inkling of interest for the game of chess, I am sure that there is more diversity and chess naive experiences and ways to acquire knowlege autonomously from same chess game experience. The 1,2,3 at all scales of teaching (i include the ply-ratchet view of chess as only scope of chess stories), is platonic in the sense that it says, you are blank. let me be your windows 10 and pour my tea into your skull, you know nothing. don't even try those silly moves (a personal quirk of mine, i love my silly moves, i find it good hygiene).

Back to endgames, with that last tenet (or impression of it)
For some the number of actors on board is what needs to be simplified. But maybe others, find more naive ideas (i mean can themselves create solutions from challenge positions, without having to be spoon fed the consecutive recipe), when then have enough material to play with. i don't even know which is more natural to me. Sparse material, means probably long sequences, which i don't like to memorize and can't spontaneously generate. But then i will find pleasure in thinking about how to dissect the problem into digestible size problem chunks. instead of starting from the most foggy-to-me position with same material, i will take the board by its horns (or just the units on board), and put a position a like in same material class (TB classes). Not done yet for many, but i have seen examples of that on lichess. and i do have ideas to keep working on. So there my instant gratification is not winning and knowing how to repeat somebody else (or TB) solution as mere sequence of consecutive plies... but discover smaller problem which i can understand..

I am even sold to that idea for thinking about various categories of engine modules: evaluation (one or zero ply, depending on which position, before ply or after ply), tree search or no tree search (here it is answer to #3, also Cf. Carlsen own self-awareness). Comparing evaluation heavy engine, and tree search heavy one, also comparing RL and SL (self-play versus pre-existing database such as the holy grail ones from some alchemist database creators, or the fully filled TB databases , yes they can be viewed as position databases with plenty of Supervising lables not just categorical ones, but ordered ones, and whose to say they could not use floats or rationals as context or target also). What that has in common with human teaching, is that the combinatorial complexity is reduced (that's for machines and TB), which translates into increased isolation of features, emerging, on board, with concept of piece mobility dimensions ("space" for that piece), and how they can be put together with clearer geometrical intuition transfer from non-chess intelligent life (all walks of chess life). That is what complexity reduced means for teaching. d'apr`ès moi.

but, what about the full game? The problem with endgame first and only, is that one can skip the subtlety of openings. The or a problem with opening "theory", and its repertoire mastery offshoot, is that one can skip the subtlety of openings.

Middle game. That has been one of my personal favorite, even before endgames. just to be plunged in unequivocal fog...

Yes i just said openings were the most subtle... contradiction? well, they don't look complicated do they the standard symmetric initial conditions of all full games (right, there is only one in standard). So while the move candidate (including silly ones, if any), look simpler, it is their longer-term consequences that are complex because subtle (with increased experience, or imagination or both). Middle-games are the most scrambled positions (hopefully still drawish) with all material on board. I think from endgame sharper "turns", to full spaghetti of middle-game position challenges, is a good contrast. Why wait for mastery of one to go onto the other. If one is to make associations, one has to look first at both situations in parallel, or closest (i.e. same time ranges , know to optimize various scales of associative memories). Back to the ratchet only inexorable pace.

How to get there?
teach ratchet chess always from initial standard position, hope the listener always glued, and when at endgame, assume still there, and remembering the decisions made at beginning and doing proper associations? does not seem very knowledgeable of psychology, or pedagogy for any age.

I guess i have a theme. "skip those plies" i don't want to get lost in them when looking at bigger pictures or story. Use them only when student or non-chess tourist does not see the associations, or is ripe for some grinding of the ratchet kind, having been through enough charades at higher scope level.

I don't know the judicious blend of low level proof and high level (chess time scale) association that would be a unique proven teaching regimen, one thing i am sure is that both Hendricks and Sillman opposite have been assuming that there is no teaching of it, other than having started chess very young, or have been born lucky to have a lot of free time later in life...

So back to my question, which i insist appear to be taken as off-limits to both views... back to alchemy for that one.

endgames training fine. sold to the idea. to build the mobility dimensions of each men on board. and start combining them, constructively with increasing complexity of those combinations (non-chess meaning, it still exist as an English word, i would say). I also think, that in teaching of chess or a lot of chess mentality, there are sacred cows of assumptions that are rarely revisited. One of them being that chess is a fully conscious look-ahead engine type of search. That is not either what Hendricks or Silllman say, but it still is pervasive in how stuck with consecutive ply didactic constraints everything is deliverd with. I don't need to screw every screws in an "Ikea" furniture, to understand its model, and how it might look like. I will need to do it to make an actual instance of the furnitur. But i am a free thinker, and can view the whole, without having to worry about the 1,2,3,4,5. i can even look at many such models without having to make an instance of each. And even looking at each 1,2,3,.... I could make decisions based on those "static" models. (because i have internalized the screwdriver screwing scenette routine, seeing a screw with dotted into a dislocated version of the furniture, allows me to predict how the thing will look and what to do, without having even to follow the 1,2,3. Other tenet: we all have the same type of brain, and cognitive set-up.. possibly all masters end-up in the same box, i don't know, but, at least in the adult non-chess world, and the subset of those that get an inkling of interest for the game of chess, I am sure that there is more diversity and chess naive experiences and ways to acquire knowlege autonomously from same chess game experience. The 1,2,3 at all scales of teaching (i include the ply-ratchet view of chess as only scope of chess stories), is platonic in the sense that it says, you are blank. let me be your windows 10 and pour my tea into your skull, you know nothing. don't even try those silly moves (a personal quirk of mine, i love my silly moves, i find it good hygiene). Back to endgames, with that last tenet (or impression of it) For some the number of actors on board is what needs to be simplified. But maybe others, find more naive ideas (i mean can themselves create solutions from challenge positions, without having to be spoon fed the consecutive recipe), when then have enough material to play with. i don't even know which is more natural to me. Sparse material, means probably long sequences, which i don't like to memorize and can't spontaneously generate. But then i will find pleasure in thinking about how to dissect the problem into digestible size problem chunks. instead of starting from the most foggy-to-me position with same material, i will take the board by its horns (or just the units on board), and put a position a like in same material class (TB classes). Not done yet for many, but i have seen examples of that on lichess. and i do have ideas to keep working on. So there my instant gratification is not winning and knowing how to repeat somebody else (or TB) solution as mere sequence of consecutive plies... but discover smaller problem which i can understand.. I am even sold to that idea for thinking about various categories of engine modules: evaluation (one or zero ply, depending on which position, before ply or after ply), tree search or no tree search (here it is answer to #3, also Cf. Carlsen own self-awareness). Comparing evaluation heavy engine, and tree search heavy one, also comparing RL and SL (self-play versus pre-existing database such as the holy grail ones from some alchemist database creators, or the fully filled TB databases , yes they can be viewed as position databases with plenty of Supervising lables not just categorical ones, but ordered ones, and whose to say they could not use floats or rationals as context or target also). What that has in common with human teaching, is that the combinatorial complexity is reduced (that's for machines and TB), which translates into increased isolation of features, emerging, on board, with concept of piece mobility dimensions ("space" for that piece), and how they can be put together with clearer geometrical intuition transfer from non-chess intelligent life (all walks of chess life). That is what complexity reduced means for teaching. d'apr`ès moi. but, what about the full game? The problem with endgame first and only, is that one can skip the subtlety of openings. The or a problem with opening "theory", and its repertoire mastery offshoot, is that one can skip the subtlety of openings. Middle game. That has been one of my personal favorite, even before endgames. just to be plunged in unequivocal fog... Yes i just said openings were the most subtle... contradiction? well, they don't look complicated do they the standard symmetric initial conditions of all full games (right, there is only one in standard). So while the move candidate (including silly ones, if any), look simpler, it is their longer-term consequences that are complex because subtle (with increased experience, or imagination or both). Middle-games are the most scrambled positions (hopefully still drawish) with all material on board. I think from endgame sharper "turns", to full spaghetti of middle-game position challenges, is a good contrast. Why wait for mastery of one to go onto the other. If one is to make associations, one has to look first at both situations in parallel, or closest (i.e. same time ranges , know to optimize various scales of associative memories). Back to the ratchet only inexorable pace. How to get there? teach ratchet chess always from initial standard position, hope the listener always glued, and when at endgame, assume still there, and remembering the decisions made at beginning and doing proper associations? does not seem very knowledgeable of psychology, or pedagogy for any age. I guess i have a theme. "skip those plies" i don't want to get lost in them when looking at bigger pictures or story. Use them only when student or non-chess tourist does not see the associations, or is ripe for some grinding of the ratchet kind, having been through enough charades at higher scope level. I don't know the judicious blend of low level proof and high level (chess time scale) association that would be a unique proven teaching regimen, one thing i am sure is that both Hendricks and Sillman opposite have been assuming that there is no teaching of it, other than having started chess very young, or have been born lucky to have a lot of free time later in life... So back to my question, which i insist appear to be taken as off-limits to both views... back to alchemy for that one.
#10

@MrPushwood

You express the hope that 1-ply intuition work better than Kleenex.

My guess - would be "nope". Kleenex cleans up snot pretty efficiently.

Playing without calculating and evaluating game trees would be like showing up to Carnegie Hall to do a piano recital only armed with some rules like "Hit the keys hard, fast and frequently".

@MrPushwood You express the hope that 1-ply intuition work better than Kleenex. My guess - would be "nope". Kleenex cleans up snot pretty efficiently. Playing without calculating and evaluating game trees would be like showing up to Carnegie Hall to do a piano recital only armed with some rules like "Hit the keys hard, fast and frequently".

This topic has been archived and can no longer be replied to.