Chess theory

The comedy of this thread is the blinkered fascination with "great players."

And the massive waste of keystokes that ensue.  Get over it.

There is no magic bullet to getting stronger.  After 1800 USCF only 10 percent of tournament players rise higher, and they are thoroughly obsessed with the royal game.

No surprises there.

@Ziryab has it right -- "Systematic Study," and all that entails.  Hire him to guide you.

You'll be a better man for it.

I think it's a worthwhile endeavor. Anything that might contribute to a deeper understanding of the game will get my attention. It doesn't strike me that such system is unlikely at all. Certain level of healthy skepticism is reasonable, but I'm willing to give eastyz the benefit of the doubt despite the very limited clues given so far.

My main doubt is how far into actually systematizing his approach eastyz has gone. Moreover, some of the things he has explained thus far seem to suggest that this system could only be grasped by someone endowed with superlative working memory (the ability to manipulate board pieces in one's mind is not trivial), which strikes me as rather impractical or even beyond the capabilities of most people.

I'd like to hear what eastyz has to say about this or correct me if necessary; I might have misread something.

It's an interesting discussion, nonetheless. And makes me overflow with ideas. I'm glad that geometry was brought up, at least to brainstorm some more. The concept strikes me as pretty relevant to any comprehensive algorithmic system for chess. Something akin to algorithmic geometry (aka computational geometry).

I wonder if the idea of cooking up a strictly geometric language/system for chess has any merit. Something that could be derived not only from distinctive positional/tactical patterns, but also from the algorithmic logic of the rules of the game so it could be applicable to any situation. It's applicability would be highly heuristic, efficient and non-redundant. The possible downside of this is that the game could become a mere exercise in linear and predictable calculation. Not much thinking would be involved because the game would be pretty much solved in a deterministic way. There would be no actual game to play.

It's kinda challenging for me to hypothesize a comprehensive system that lacks the potentiality of solving chess somehow. At least, a strictly geometric one. I could be wrong...

I partially disagree with the working memory comment.

Yes, working memory is critical to chess success, and the measurable decline in working memory (as well as the increased practical problems with declining ability to successfully 'wipe' that memory when necessary) after the age of 27 is likely one (of many) factor in older chess players not being able to reach the same heights.

Since the number of items that can be held in working memory is relatively constant (within a range) across individuals, the key to using working memory successfully is to group complex tasks into bundles that are held in long term memory, and then working memory can treat each bundle as one item.  For example, the people who can hold huge lists of numbers in working memory almost all use learned bundles of numbers to achieve this.

One form of bundling that we all take for granted is language.  We have single words which describe quite complex processes.  Once we are fluent in that language then we can hold quite complex ideas in our working memory.   

Another less obvious bundling technique is mnemonics: "Roy G Biv is a colourful man!"  (Red Orange Yellow Green Blue Indigo Violet)

The key to chess thinking would appear to be the development of language to describe a variety of complex analytical patterns which can be stored in long term memory, increasing working memory.  

The issue is that in many ways this is not a shortcut, in the same way that becoming fluent enough in German (assuming you are not a native German speaker!) requires a huge amount of time and work, so does becoming fluent in a  new chess language.

Of course, the pattern recognition/heuristics will always be faster than even a fluent chess language user, because heuristics uses the brain's 'machine language' and therefore does not need to be compiled into that basic working language of the brain.  But as eastyz says, the chess language is for use when heuristics fail.

The reason that pattern recognition does not make us all masters is because heuristics are only right most of the time (say around 90% of the time).  One truism in family practice is that we make the diagnosis in the first few seconds of a consultation with a patient, but we spend 10-15 minutes with each patient making sure we were right!

Yes, that's a valid point.

I guess it depends on how well-systematized this chess system eventually becomes.

Heh. Indeed. "Systematic study" and all that entails. I actually agree with this generally in terms of a beginner. Although this doctrine changes as a player gets progressively better. One begins to wonder, as they face stronger opponents or of relatively equal strength, go through a string of losses exactly what one must do to improve. Especially if one is working with a trainer.

It's easy for a 1800 player to laugh and say what we are talking about is, comedic. He already has his own system built through his own experience, knowledge, talent.

My story is, I started out as a book player when I became serious with chess (I learned the game at 8) which was around 13. I was successful because not many players (including my friends in my high school chess club) studied openings as much as I did. Heh. It was my "long" game. I just had to learn some endgame fundamentals, be alert for tactics. I had a very good memory capacity back then (I was also known as the Computer) and patterned my game around RJF. It's funny that I spent a lot of time in Community College's library (as access was easier back then) simply because they had a good selection of Fine's books. Games collection. Fischer's Complete Games. Tal's games. It would be a couple of years later (when I became eligible to work@15) that I would start my own book collection.

najdorf, when I was playing competition, my OTB rating was much higher than your online rating.  At the time, I was said to be a good tactician.  I don't play competition anymore.  My point is that with the much maligned "system" my tactical ability is much greater now than what it was when playing competition.  But you are right.  You can develop your own system.  The issue then becomes whether it is as effective as other possible systems.  That is always an issue with systems.

Anyways! Easy's assertion is pragmatic. Like patterns, what to do if your partner deviates in the opening? Memorizing a ton of lines doesn't do one good if one cannot think on his feet. Elaborating further, flash cards are great for learning the basic multiplication tables but alas there are none when you get into advanced mathematics (trig, calculus even geometry) and even then, various authors have written books on solving equations with shortcuts, thus eliminating the need for pattern recognition/flash cards. It's not soo simple to learn as one poster would have you believe but once you get it, it becomes applicable regardless of the situation.

Oh. Cool easy. Sorry I was writing my above post while you'd already posted yours.

Indeed. I don't doubt you're a much stronger player than I am. In fact, my online rating in this site is just relative to my playing experience. It's just a hobby for me. Like you, I don't play competitively anymore. But yeah, your "system" works for you because you'd taken the time to tweak it. It's your baby. I can relate because that's how I feel about my opening repertoire.

Like nature, there are eco-systems among many eco-systems within one another, magically all working together like a Swiss watch.

My system evolved in '92. I'd been out of competition for about 3yrs. I'd stopped into my favorite comic book store (which also sold used books) and found an old copy of "How To Think Ahead In Chess" by I.A. Horowitz & Fred Reinfeld. The methods and techniques of planning your entire game. Heh. Circa 1951(!)

If I could write some part of the Introduction (or most of the relevant pieces)

Many years ago, the mighty Capablanca engaged in a rough and tumble chess game with a New York expert. Momentarily forgetting that he was invincible, Capablanca lost!

At this unexpected turn of events, the spectators were amazed. How could the grand master lose to one who was, comparatively, a second-rater?

"How far do you think ahead?" One of them asked the Champion. And, as befitted the titleholder, the impressive reply came back: "Ten moves!"

That is nine moves more than usual for Capa.

"How far ahead do you think ahead?" the winner of this particular contest was asked. And he replied, "Only one move."

Now here was a confusing state of affairs! How could a chess player who thinks only one move ahead, defeat another who thinks ten moves ahead? The victor then explained:" I think one move ahead-but it is always the best move!"

It is really no secret that the right move in each position is the sure path to victory. To find the right move invariably, however, is beyond the power of ordinary mortals. What is within his grasp, is the right way to think ahead. Not far ahead. Accurate appraisal is more important than deep calculation.

najdorf, I agree except where the position is concrete you have to calculate accurately (by definition, there is a solution).

Heh. Yeah.

As it is, the authors mention the syntax, "patterns". But I reckon it's more of a language (for the readers to relate to) than what other posters advocate.

It is a common failing of all chess learners to want to become generals before they have no more of an inkling of strategy and tactics. Naturally, when put to the test, they wind up in a welter of confusion...Attack?...Defend?...Where?...When?...How?

All these are perplexing questions. And they will remain so, until the learner familiarize himself with guiding principles and patterns. (Which is to say, opening patterns-Naj)

Skipping ahead~

In our zeal to emphasize the strong points of our recommendations, we may leave the impression that White has a forced win and that Black has a forced draw. Definitely, this is not so. Only the astute play of a master, however, can challenge some of the minor conjectures. And this book is not intended for masters!

After you become thoroughly familiar with the patterns (opening play-Naj) of play recommended in this book, you can, if you wish, branch out into other openings (patterns- Naj). The principles involved in these patterns obtain in all patterns of play. (For me, nuances of each line learnt, contains principle applicable to other variations, notably the Colle, Torre Attack, Tromp, and surprisingly, QGD lines).

Indeed easy. That of course is a no brainer. Mating attacks, double attacks, discovery.

What they are saying is, very similar to you, is that not every position is assessed the same and that accuracy in appraisal of the position is paramount (paraphrasing).

A glance of some positions, the appraisal of them, are automatic. Familiarity with Legal's legacy for example can help facilitate quick appraisal. But if it were not the case, accuracy is essential as in practical terms, we only begin calculating when we know what lines to follow to be the most surest/efficient solution.

The key is to have a system that draws you to looking at the right lines sooner so that you waste less time.  OTB chess is correspondence chess.

Exactly easy. You have found it for yourself and although still a work in progress, obviously successful for you. I have my own system but it's geared toward searching for ideas. Not very suited for tactical trainer type puzzles. Although applicable, speed of solving depends on my mood.

I don't need tens of thousands of patterned drills to be successful. Heh. I guess if I wanted to but again, tactics aren't my main concern.

Of course, if I want to astronomically increase my rating in that, I would come to you.