Thursday, May 01, 2008

While I Was Out

Occasional breaks from chess and blogging have a refreshing effect; if you're open to it, you can sometimes see things "with new eyes," the way you see a place you've never been when you arrive there for the very first time. If you work at it and pay attention, you can partially attain this "refreshment" without actually going anywhere, by actively remembering that first time, the way it looked then, the feelings you had, the kinesthetics and the smells and the sounds.

I've read and digested a good number of fine posts in the last few days, and combined with my vacation and a week without playing (a full-point bye in the Reno CC Ch. Swiss) I have some new insights about what I'd like to do, what I need to do, and what I'm actually going to do to learn, grow, improve and have fun in chess.

Originally I planned to put a whole bunch of stuff into one megapost, but after hacking away at that it seemed to me much better to break up this Epic of Our Times into more manageable bites. So first, let have a look some recent work by the chess blogosphere's resident neuroscientist.

Blue Devil Knight, during his extended retirement farewell tour, unleashes some of his best-ever posts, a real feat considering that he has one of the best-ever chess blogs. Check out:

Part the Last (8) of his book reviews.

Chess memorization and seed planting--let the neurons do their job!

Two academic papers of great interest to chess players--as far as learning 10,000 pictures, and why it might be much more difficult to learn 10,000 chess positions, well that's my next post. Perhaps most directly related to our favorite subject (winning more!), Chess Masters' Hypothesis Testing; if you've never heard of "confirmation bias" you really need to get out more often.

I thought I'd posted about this somewhere previously, but for the life of me I can't find it right now; anyway, the important point is that stronger players usually make the successful effort to "falsify" their own plans, finding the opponent's best responses, while weaker players see a lot of variations that are good for themselves, often missing good moves for "the other guy." While this isn't a brand-new insight that's never been noted before in the history of chess, it is one of the main reasons for the differences in strength between players, and it's not exactly an ability to calculate better or to "find" winning combinations, it's in a different category; it's what we mean when we say we don't "see" a one-mover that loses a piece, in a certain way it might be said that this is the basis for almost all blunders, this failure to seek and find the opponent's best reply to the move we want to play.

So if it's that important, how do we go about working on it? It occurs to me that first of all, solving "White to move and win" positions, where we know a crusher is there, might almost be counterproductive in this regard! We're given that Black has no good defense, so all we have to look for is the "shot." As much as I enjoy, for example, Chess Tempo, this is the only kind of problem provided there, and I'm wondering if it's not doing me that much good to find all those wins...

Seems to me that what's needed here are exercises of the type, "White has three plausible moves, Nb5, Bb3 and Qxd4. Find Black's best response to each." I don't know if I've ever seen a set of exercises in precisely this form, though GMs Hort and Jansa had a book out many years ago called The Best Move that might have been close. Currently Ray Cheng's Practical Chess Exercises has gotten a lot of great reviews, and though I don't own it (yet) it might suit the bill as well as anything out there.

Otherwise, we're just going to have to do it ourselves, Dear Readers, playing through games, whether our own or masters', and looking at everything with a critical and "falsifying" eye. Almost all of us, even masters, do have this weakness to some degree, always on the lookout for what we can do to the opponent; it seems we'd be better off spending a lot more time and effort finding out what the opponent can do to us!


Francis W. Porretto said...

In his little classic How Not To Play Chess, Eugene M. Znosko-Borovsky wrote that "It is not a move, not even the best move, that you must seek, but a realizable plan." But plans are most useful as metrics by which to detect and measure change -- deviations from the expected. It therefore seems that real chess vision consists in seeing the plans possible from a given position, both for yourself and for your opponent -- and which ones are likely to beat which others.

A plan is, of course, made for a few moves only, not for an entire game. But the strategic shape of the game constrains what plans are plausible. Thus, we study pawn structures, the quintessential shapers of strategy, and openings, which give rise to particular pawn structures, and choose from among them according to which ones lead to positions that favor the sort of plans we prefer.

Or as my diminutive darling Fetiche once said, "If I can't see any possibility of ripping him to bloody bits, I start to lose interest. I might as well be playing shuffleboard."

Blue Devil Knight said...

Thanks for the kind words. It turned out there were quite a few loose ends I had forgotten about, piled up in my "draft" box at my blog. They are now all gone except one, so she's almost done.

Good to see you expand on the notion of falsification. Soltis talks about it some (using different words)--typically people say find the 'worst case scenario' for each move and use some kind of process of elimination (this is number three of Soltis' four methods for picking candidate moves, described here).

Your criticisms of tactical sites/puzzle books and the like are right on. I think you are right that Cheng is a good antidote, as is Ward's book 'It's your move'. Neither is quite what you are saying, but is along the same lines (I review Ward in video 6).

But your idea is different from anything I've seen, and is a really clever idea for a set of problems. A great way to work on practical move selection too.

Hank said...

I'm sure everybody knows this already, but anyway... At the risk of sounding like a noob by dropping his name, which gets more than its share of play in the chess improvement blogosphere, I thought I'd mention that Dan Heisman's book "Looking for Trouble: Recognizing and Meeting Threats in Chess" is often regarded as being innovative (for a puzzle book) for its focus on developing "defensive vision" (not his term)... And his notions of "Hope Chess" and "Real Chess" seem very closely connected to these issues of confirmation bias, and specifically the "failure to seek and find the opponent's best reply to the move we want to play" (= "Hope Chess")... -- Hank

Anonymous said...

Your objection to Chess Tempo (and the like) is why I enjoy Larry Evans' "What's the Best Move?" in Chess Life so much. Yes, there's the choice of three moves, but you're not given what's supposed to happen.

Some of them are even "last stand" situations where you're looking for the saving move. Defense puzzles are rare, and I haven't seen a great resource for them.

Murphyman said...


You talked about a type of problem where which of the 3 moves is the best.

I would heartily recommend Igor Khmelnitsky's Chess Exams which has this quality of question to some really devilish puzzles in it?