Yes, thanks for the clarification, Katar.  I don't think that we disagree in principle; after studying a Master game in depth I can remember it perfectly for a time, then it begins to fade.  I wouldn't call this "memorizing," however.  When I think of having committed a game to memory, I think of one of the few games (The Opera House Game, for instance), that I can recall perfectly and completely on any given day, at any time, without looking at any notes beforehand, etc.  There are only a few games for which I can do that, but on any given day there may be a handful of other games (usually my own!) that I can also recall perfectly.

Excellent point about studying with an actual board and pieces, as well.  I always do this when I can, and at the very least it helps me to "get serious" and concentrate.  I suspect there are other benefits as well.

When tasked with "memorizing" a chess game, the brain naturally organizes entire sequences of moves into groupings that are personally logical or meaningful.  In the context of a chess game these groupings are intuitively represented by ideas and counter-ideas.  It is not an instance of "memorizing" a random string of unrelated individual moves.  After studying the themes, stages, and ideas of a chess game, I find it quite manageable to replay the individual moves, one step at a time.  Just knowing the opening variation, for example, "Ruy Lopez Breyer," easily gets you to move 9 or 10.  From there you should be able to figure out what plan each side embarked on.  And that should pretty much trigger the individual moves.  You may transpose moves here or there, but that is not a big deal because the ebbs and flows of the game are what matters.

http://en.wikipedia.org/wiki/Chunking_(psychology)
"It is believed that the assimilation of different items according to their properties occurs due to individuals creating higher order cognitive representations of the items on the list that are more easily remembered as a group than as individual items, themselves. Representations of these groupings are highly subjective, as they depend critically on the individual's perception of the features of the items and the individual’s semantic network. ....  Chase and Simon (1973), and later Gobet, Retschitzki and de Voogt (2004), showed that chunking could explain several phenomena linked to expertise in chess."

I suspect Eric's concept of "learning" or "absorbing" is more or less equal to what I have in mind with the word "memorizing".  For me, memorizing is simply a byproduct of learning/absorbing.  My problem is that if i don't "test myself", I have a tendency to read books passively like a zombie-- known as "reading and nodding".

Stigma wrote on 05/02/13 at 22:40:41:

Answers: 
1. Not very long, maybe an hour-- maybe a day or two.  But if I jog my memory, the whole thing comes back quickly.
2. I don't see any reason or need to repeat what I've already covered.  There is more benefit in moving on to new material, or going out to play frisbee.
3. Memorization is just a way to trick myself into being disciplined about paying attention.  For me this process can be a motivating and fun challenge.

I'm not sure there really is a finite limit to human memory. The main point anyway is that memory traces tend to fade over time if we don't revisit them (repetition!).

I found this interesting thread on another forum http://www.chess.com/forum/view/general/does-memorizing-masters-games-good?page=... where a certain SmyslovFan (!) claims that "Just about every GM has memorized every game they played, and most have memorized numerous complete games in the openings they play". I suspect this is an overstatement; I hardly remember all the games from my most recent tournament in full, much less my 1100 rated tournament games, but I can still give ordinary GMs a good fight and take the occasional half-point from them.

Now, opening lines and the essential endgame theory really is for memorizing. For everything else we rely on mostly unconscious pattern recognition.

@katar: For how long do you perfectly recall the games you have studied with your method? Do you ever need repetition? And are you certain you get a real effect of the memorization itself, over and above regular, unconscious pattern recognition?

whitecraw wrote on 05/02/13 at 20:07:40:

Probably 15-30 minutes per game.  Time spent is automatically regulated based on how long it takes to memorize the moves.  Something simple from the Art of Checkmate may even take less than 10 minutes.

I strictly only use an actual board and pieces.  There is a tactile component of physically handling the pieces that seems to help the patterns "stick".  For example, violent moves/captures can be done with a forceful snap of the wrist, while a subtle "Karpovian shift" like a Qd1-c1 can be nudged with a pinky finger.  In other words, these tactile clues can facilitate recall and help to get the moves into your bloodstream so you can easily recite or replay the game from memory.

This has worked for me, but it is only one approach out of many and I wouldn't claim there is a single "best" way.

Stigma wrote on 05/02/13 at 12:13:15:

Personally I've generally used an "abbreviated Bronstein method".  I will play through each game somewhat quickly, while looking at the notes.  I will then close the book and replay the whole game from memory.  If I fail, i will start from scratch until I can replay the game from memory.   If the game is fascinating or interesting in some way, I may spend some more time digging into it after having memorized it.  Otherwise, I just go on the next one.  After internalizing the game to the point that i can replay it from memory, I feel as though i've extracted the main lesson of the game without spending too much time on it.

For me personally, if I can't repeat the entire game from memory, I was probably just "reading and nodding."  The same could be said of a college student who after reading from a textbook cannot put the book down and explain the selection in plain terms.

I think this method does wonders for a person's pattern recognition, but may not do anything in terms of developing straight calculation ability.  Calculation is better addressed in other ways.  The above is my way of getting through game collections while maximizing the benefits / hours ratio.  And I have actually read a several games collections cover-to-cover this way.

kylemeister wrote on 05/02/13 at 14:41:40:


That must be it.  Thanks for that suggestion, most learned kylemeister!  I have Romanovsky; I'll check it out.  Romanovsky's work, by the way, is pretty good chess literature, even if it does contain this questionable claim.

nocteus wrote on 05/02/13 at 10:23:02:



Fine, but this really has nothing to do with my view that in general openings works do not rise to the level of great, or even good, chess literature.  I posted that in reaction to some of the books listed as favorites here.  We could debate my ideas about chess education in another thread, though it becomes tiresome to have to do so here, year after year.  I am increasingly willing to let those preoccupied with questions of the student's "taste" and "style" to wallow in their folly.  I am only sorry for their students.

One's journey does indeed begin.  The question is, when.

Quote:


I would not say "many."  There are probably about ten books devoted to the opening that I would consider good chess literature.  Polugaevsky's is among them.  I admire Watson very much and I find his works extremely useful (I buy all of them), but I think that the various editions of Play the French do not aspire to be chess literature (stuff you keep on your shelf forever and pore over on winter evenings). If they were chess literature, there wouldn't be four editions, you know?  

The problem with opening variations is that they are ephemeral.  They are very soon superceded by new variations.  You're right that openings works often include middlegame instruction (less so, endgame).  One recent standout is Marin's two-volume set on how to play Black after 1...e5.  But I wouldn't call Marin's work chess literature.  I recently discarded two or three hundred opening works that I had accumulated during the 1970s - 1990s (not all my collection from that period, but most).  I should have done it sooner.

And let's face it: most opening books are crap.

P.S. MCO was not a waste of time in it's day, nor was ECO.  It's just that nowadays, the scope of theory is too vast to fit in an openings compendium, and it changes so fast that any compendium would be obsolete when it was still in the galleys.