## Educated Guesswork

Among first-class masters the capture of the adverse king is the ultimate but not the first object of the game and by best play on both sides a draw ought to be the legitimate result. – W. Steinitz, The Modern Chess Instructor

Thus did Steinitz enunciate his principle theoretical intuition, one still unprovable even using the computer. Chess is intrinsically difficult, it requires search to all terminal positions to validate the utility of any single move.

Computational complexity theory tells you there is no clean answer for games like Chess, Checkers, and Go. If you believe in the perspective of complexity theory, it tells you that all you can do is sit there and do the bookkeeping and exhaust the exponential possibilities. – Erik Demaine, interview

What indeed are we doing between the moment we sit down at the chess board and the moment the game resolves itself to a calculable endgame? We are exercising what are sometimes called heuristic techniques which Wikipedia defines as “any approach to problem solving, learning, or discovery that employs a practical methodology not guaranteed to be optimal or perfect, but sufficient for the immediate goals.”

Various chess masters (notably Steinitz and Nimzovitch) have believed themselves to be enunciating fundamental scientific principles of the game, but most, if not all of their thought has been directed to expanding and refining our store of heuristics. Bronstein and most problemists delight in the irony of the positions which defy the approved heuristics.

This variation, given by Tchigorin, shows that masters of the last century were capable of accurately calculating deep combinations. – Imre König, Chess from Morphy to Botwinnik

Heuristics in chess are effectively prods to calculate certain variations, the concepts of the struggle for the center, the advantage of two bishops, overprotection, pruning the search tree, but none answer the fundamental question, what’s my next move?

A teenaged expert recently expressed to me in a voice close to tears his frustration with opening study. “There are no openings,” I offered, “calculate from the first move.”

Jacques Delaguerre

## Fischer and Me

Last week in “Fischer and You” I touched on Bobby Fischer and Endgame, a biography by Frank Brady. Brady writes that Fischer played “12,000 games a year, most of them speed games” from age 11 to 13, and as is well known, within two more years Fischer qualified as a World Championship Candidate at the Portoroz Interzonal 1958 with a +4 score (not +3 as I wrote last week).

In the days before computers many players (at least according to their own accounts) reached master strength mainly through concentrated play night after late night in smoky chess clubs, with little book study. Most Chess Improver readers outside of New York and London don’t have this opportunity these days, but now there is the Internet which provides a similar opportunity (though without the stimulating cigarette/cigar/pipe smoke, and the stimulating presence of a live human across the board).

Since I did have some spare time recently I decided to experiment, if only for a week. with some concentrated blitz play. I have a rather convoluted history with fast chess; 30-odd years ago when I first started playing serious chess at the Reno, Nevada club one of my mentors was Ray Wheeler, a retired railroad engineer who rather specialized in speed chess. Ray was tournament-rated around 1700 USCF but at his preferred 5-minute time control he was more like 1900-2000. I played quite a few games with him and would usually lose on time but did learn something about chess in the process. When I came to Alaska there were some serious tournament players but I only played occasional blitz for many years, until Internet chess came along.

In early 2012 as Nigel posted here I decided to swear off of online blitz and engine-assisted analysis for awhile. I’ll admit the interregnum didn’t last the entire year but at least online blitz was out for many months. And I did stay away from computer analysis for most all of 2012, so there’s that. At any rate,this latetst experiment, prompted by the Fisher biography, proved quite interesting and even instructive. You, the reader, have probably played blitz chess sometimes and I would be interested in hearing feedback comparing your experiences on Nigel’s Facebook page.

I set out to play 300 blitz games in seven days, on the Free Internet Chess Server where I’ve had an account for a number of years and had an all-time high blitz rating of 1464 (probably comparable to 1764 at “standard” time controls of 15 0 or more). I played most of the games at the very rapid 3 0 time control instead of 5 0 (which I suspect Fischer played most of the time). “3 0” allows only three seconds per move to get through a 60-move game and places a real premium on playing by feel and subconscious knowledge rather than “If i go there, he goes there, then I go there…” calculation. I made it a point to play higher-rated players whenever possible.

Observations:

• After about 100 games I passed my previous high blitz rating and got into the mid-1500s for about 20 games, and though there was a “reversion to the mean” and normal statistical noise up and down, I ended the experiment a few points above my previous all-time best.
• One of the good things about fast chess is indeed the fact that there is little “thinking” required most of the time. Just make “good moves” (ones that don’t hang pieces or allow mate!) quickly and go with the flow. One of my own weaknesses in tournament play has always been taking each move so “seriously” and trying to keep everything “under control.” As a result I’ve often exhausted my reserves of mental energy before the game was over–many, many times I’ve played 15, 20 or 30 moves at a very high level and then started to make simple mistakes. Losing a tournament game has always been a big deal to me, but losing a 3 0 game to an pseudonymous non-entity is not. Something Nigel and I have touched on in private correspondence is that really strong players don’t have to put a great deal of effort into most moves–they see they correct move right off. This allows them to save that mental energy and will for the few critical moments of the game, and may partly explain why stronger players so often prevail in the end, even from equal or dubious positions; the weaker player has expended so much of his strength in getting to the position at hand that he is even weaker, relatively, that at the start of the game.
• At the 3 0 time control moving quickly is of such essence that it often takes precedence over the quality of the move, which is a real downside to this form of chess. If I have one minute and you have one second on the clock, I am going to win as long as I have mating material on the board. This leads to distortions like moving instantly or trying to capture all of the opponents pawns instead of going for mate, none of which are really useful if this form of chess is to be used for improvement.
• Since playing the same opening 100 or more times in a row would be rather boring, I began experimenting more, especially by playing 1. Nc3 as white. I was surprised at how often opponents responded 1. ..Nf6 when I would steer toward a Four Knights Game (an opening I enjoy with both colors) or 1. ..d5, 2. ..e6 and a French Defense (if I wanted). A ration of blitz is probably a good way to learn an opening, especially if one concentrates on the patterns and looks things up later to see where the improvements might lie. On the other hand, there are some strong blitz players out there who specialize in “The Hippo” and other weird stuff and count on getting ahead on the clock, then outplaying the opponent from a cramped position. It often works but doesn’t provide much instructional value!
• A short while after completing my “blitz week” I got together with some “real people” for the first time in months and played a little tournament at a 10 0 time control, scoring 5/5. While the field wasn’t too intimidating (the next strongest player was rated about 200 points lower) I found myself just having fun, relaxed and not worried to much about making the “best” move. It’s the feeling I want to carry forward into more serious future competitions.

Finally, a good question is whether the 30 hours spent on blitz would have produced more or less improvement if spent studying, say, Bobby Fischer’s best games and thoughtfully analyzing critical positions with pieces and a board, on one’s own or, even better, with a stronger player. I would say the analysis sessions would probably produce more improvement; but there are some valuable lessons and techniques that can be found in a modicum of blitz. You might want to try it out, with some specific intentions and goals in mind.

During my experiment I found that around the world there are people playing nothing but blitz chess day after day, whiling away the hours of their life on the chess equivalent of “reality” television. That, I think, can only lead to the opposite of improvement. Beware…

Robert Pearson

## Correspondence Chess Openings

An interesting question came in this week about the openings someone should use in correspondence chess and whether these should be the same as the ones in over-the-board games. Well, it depends…

If someone wants to play correspondence chess for its own sake then they should use the ‘most principled’ moves in the most challenging lines and not worry about either wild complications or lengthy theoretical variations. This approach will reduce both sides’ margin for error and lead to sharp struggles, which is rather necessary in this kind of chess, especially if computers are allowed.

On the other hand if someone uses correspondence chess for training purposes (a major recommendation of mine for over-the-board players) they should use the openings they intend to use in over-the-board play. Using them against super-GM opposition (which is what happens when you marry a 1900 player to a long time limit plus Rybka and Chessbase) is a wonderful way to iron out any wrinkles.

It is possible of course to tick both these boxes if one wishes to participate successfully at GM level and the following game was one of my own attempts to do so. It’s also one of my best at any form of chess.