Sunday, August 14, 2005

De la Maza's "Rapid Chess Improvement"

Today I read Michael de la Maza's surprisingly invigorating (though ultimately disappointing) "Rapid Chess Improvement." How could I have waited so long, you might ask? Well, I certainly was familiar with his ideas. I had read de la Maza's earlier articles, for example, available online at Chess Cafe:

I had also read quite a few reviews, including a skeptical one by Randy Bauer and a positive one by Hanon Russell. And I've been reading lots and lots of chess blogs by his followers.

Reading the book was a bit of an afterthought and it took me little more than an hour or two between weeding and watering my garden this morning. I have to say that being reminded of the basic premises of the de la Maza method has definitely inspired me, though I would not say it has made me a "believer." In fact, it left me somewhat a skeptic--though at the same time a skeptic who is much more enthusiastic about studying chess problems!

While I would agree with many of the basic premises of the book and recommend some of its ideas to developing players, I do not think I'd much recommend the book. For one thing, the book itself, like most books really, is just a lot of padding (and often poorly developed padding with language drawn from the genre of infomercials) layered onto the original idea--to the point, in this case, where it risks obscuring some of its best insights. Second, it is really the basic insight itself that is most valuable, in my view, and not de la Maza's specific program.

Basically, de la Maza tells us (as many have said before--as he himself reminds us) that all beginners and Class players could benefit enormously from a focused program of chess study that emphasized most of all tactics and board vision. The precise program he lays out is then described (in only a bit more detail than in the above-mentioned articles and reviews), and it involves repetitive tactical exercises of increasing difficulty (similar to athletic exercises, such as a pianist's fingering or the practice that a good golfer would use to hone his putting game). It also involves exercises to help you more fully grasp the whole board and the movement of pieces on the board, in much the way that Bruce Alberston's under-appreciated "Chess Mazes" attempts to do.

In reading the book, I developed some major reservations about the "De la Maza Method." On the one hand, I absolutely agree with de la Maza's very important insight that tactical study above all will help lower-rated players reach expert level. Of that there can be no question. But on the other hand, I think the Method falls way short of what you'd expect from a programmed course and even introduces some bad habits and notions.

My chief reservations about the Method can be summed up by two issues: first, that it suggests you need not study anything but tactics (including the openings beyond the most basic) and, second, that it discourages critical consciousness by, for example, treating the computer interpretation of chess positions as gospel. In both cases I am not completely critical of De la Maza, because I think he is right to place an emphasis on tactics and to recommend the use of computers and other methods of measuring progress in concrete, mathematical terms. But there is a point at which a basically good premise can become bad dogma, especially if it is presented without caveats and followed uncritically.

While I'm glad for a book that truthfully tells lower-rated players that openings alone are not going to cure what ails them (as many opening books at least implicitly claim and as GM Alex Yermolinsky, for one, has strongly railed against), I think the author is wrong to suggest that opening study will not have a very significant and speedy impact on performance. There can be no question that knowing at least one basic opening system as White and two as Black can make an enormous ratings difference when combined with tactical study and at least some knowledge of other aspects of the game. I can honestly say, for example, that in well over half of my games things were decided in the opening stages. And when I was a Class player, I think the number was even higher. While I think it is foolish for players to study "opening traps" (except as an introduction to tactics) or many different tricky openings or to think that memorizing opening lines alone is going to make the difference, I do think you need to know your basic repertoire in depth--as far beyond the level offered by various repertoire books as you can go--if you are going to create situations for yourself where your tactical preparation can prove itself useful. At the very least, you should study some online opening analysis and follow it up with some review of online games in your openings.

Another way of criticizing the "De la Maza Method" would be to say that it is too narrow. There is no question that de la Maza is correct that tactics are supreme and any unrefuted opening (and even a few refuted ones) are playable below 2000-level. But tactics alone are not going to take you very far if you are fighting to survive on move 10. There are definitely things that developing players need not study, and that includes arcane middlegame knowledge (such as "the minority attack") and even most endgames beyond the most basic K+P and R+K+P (especially with the advent of faster rated time controls). But opening knowledge and some other basics can pave the way to tactical success. What's more, you really need to start developing combinative thinking skills and a recognition of more complex strategic motifs to help organize your basic tactical knowledge. In a sense, de la Maza is suggesting you just study addition, subtraction, multiplication, and division, but these are really only the building blocks of more complex math.

A good place to find better ideas would be the Novice Nook by NM Dan Heisman, for instance.

Another well-rounded approach to chess study for amateurs is mentioned by Bauer in his review. It was laid out in an article by the late Ken Smith that he used to send to people who purchased through Chess Digest, and which was later published at his now-defunct company's now-defunct website. It is still available at the web archive and worth a look.

The main difference between Smith's course and the "De la Maza Method" is that Smith emphasizes opening knowledge and gambit play. He does not put as much emphasis on tactical exercises as de la Maza, but I think that's because he assumes he is writing to a Class player who has already studied tactics to some extent. Tactics are, however, central to his basic four-part program, which he lays out as follows:

"(1) Keep emphasizing 'tactics'. This part of chess will overcome a bad opening, a poor middle­game and lack of endgame knowledge. Only until you reach 'Expert' can you stop devouring everything on combinations and tactics. You put fear into your opponent when you are known as not letting anyone escape.

(2) Every chess book should be saved and gone over a second time. There was no consensus of how much time between readings. Only that you be at a different level of strength. There must be a balance between this study and play.

(3) Be exposed to different authors -- even on the same subject -- even on the same variation of an opening.

(4) Master a complete White opening system and a complete Black defensive system. It does not matter what they are---a complete simple one is better than an incomplete superior one."

I like Smith's basic idea and I'd simply add a few to his list to make the most well-rounded program possible:

(5) Study tactics and do exercises to improve your board vision at every opportunity (on the train, in the bathroom, on your lunch break, or pausing between tasks at work). If you have time to devote to a program of study using a chess program such as CT-ART, more power to you. But don't let the best be the enemy of the good. Any and all tactical study is beneficial and exercise of any sort is good for you.

(6) Play over as many games as you can in your favorite openings to get a feel for the deeper positions that arise. It's best if you do this with a computer database using a computer program like Fritz (just to know when mistakes are being made). The key, though, is to develop a good sense of patterns and ideas in your lines well into the middlegame so quantity can be as useful as quality. Simply looking at games at online databases such as ChessLab or ChessBase is a good start.

(7) Play blitz or speed chess (at ICC for example) only with a view toward cementing your opening knowledge and learning something about typical middlegames and endings that might arise from your openings. Nothing beats experience for cementing the lessons of study, but it cannot substitute for study or for serious play.

(8) Read a few classic books on the middlegame and ending and maybe a book or two on chess history. Be a literate chess player--it will help you at the board and socially. At the very least, if you want to improve your practical results most of all, read Nunn's "Secrets of Practical Chess" and Yermolinsky's "The Road to Chess Improvement." And to help your historical consciousness, at least read "The Development of Chess Style."

(9) Study very closely (and repeatedly over time) at least three books of well-annotated games: one from a major tournament, one from your favorite player, and one that covers a wide range of themes. You can't go wrong, for example, with Bronstein's "Zurich 1953" (also for download), Botvinnik's "100 Selected Games," and Nunn's "Understanding Chess Move by Move." You could substitute many others. Any World Champion (FIDE or pre-FIDE) or major contender would do as a player, for instance, and there have been many great tournaments and tournament books. Whatever you find suited to your style or level of knowledge will work best, but I don't think I'd recommend doing many more than three at first. The key is to look at these games in depth until you have them memorized. A national master I know still goes over Chernev's classic "Most Instructive Games Ever Played," for example, and it clearly has not hurt him to do so. One thing I'd say about choosing a book: sometimes it's a good idea to read something that's just a bit beyond you at the moment, because once you master it through repetitive study over time you will have made progress.

(10) Play as much serious chess as possible against stronger opponents and analyze your games afterward, first with your opponent (if he or she is willing) and then with the aid of a coach and/or a chess computer. But do not take everything any of them say as gospel.

I should probably add (11) that it can only help you to stay physically in shape by exercising, watching your diet, and avoiding smoking or other bad habits. I hope your mom told you that. I will say, though, that physical fitness is probably what makes the biggest difference over the course of a long tournament or even an intensive three-day event. There is no use studying your tactics if you don't get your rest and eat your Wheaties before Round 5.

I think many experienced players would recognize these additions as not original with me, and anyone would say these are all fairly common and good suggestions that it would never hurt to follow. I'm sure that even de la Maza follows them to some extent, though you'd never know it from his book. And that's precisely the problem. What de la Maza is pitching seems less like a good study plan for a well-rounded chess player than a rather dogmatic religion intended to create speedy self-transformation through a form of intensive prayer. I have no doubt that it will increase performance, but I'm not sure it will make you a better player--let alone a better person. The ten or eleven suggestions above would do both.

In presenting his program as something to do rather than to think about, he is not helping to further a critical consciousness. He may, ultimately, be helping some players to greater success over the board. But he is not helping people escape the sort of autistic repetition to which unhealthy-minded chessplayers are too often prone. Nor will he help you to make the lessons learned from chess useful for life, unless you are preparing to become some sort of religious fanatic. The "Seven Circles" will not bring you closer to God. More likely, they will make it harder to do your laundry or have a social life.

I actually did like one idea from the book, which was to chart your performance mathematically. You can do that in a number of ways, most obviously by keeping track of your performance rating in tournaments. There are others, such as when solving puzzles: one of the appeals of the CT-ART program he recommends (as the followers of de la Maza in the blogosphere will tell you) is that it allows you to keep strict percentages of problems solved to measure your increased performance over time. These are good ideas, on the model of other forms of athletic training. After all, what aspiring track star would not keep strict track of his times in races, down to the nanosecond? I even think it is interesting to use a computer program, such as Fritz, to track the evaluation swings in one of your games to see where the critical moments were that you need to focus on to improve. But I also think there is a limit to how far this sort of math can take you, and believing in numbers too much is dangerous.

It's especially dangerous to believe in the evaluations offered by your computer program, or to think that what it says is right, or to believe that it might pinpoint exactly where you went wrong. To give an obvious example: due to the horizon effect, a computer might not see that something you did at move 10 is wrong until move 15, so its evaluation will not dip until 5 moves after the critical juncture, which could be quite misleading for a developing player.

What's more, his suggestions for "How to Think" strike me as a very primitive version of a computer evaluation function. At each move he suggests you seek to:
"1) Improve the mobility of your pieces" (a typical computer evaluation gauge)
"2) Prevent the opponent from castling" (ok, no problem there).
"3) Trade off pawns" (I don't get that one, except that it creates an open position where you are more likely to have tactical solutions).
"4) Keep the queen on the board" (obviously in order to create more tactical piece play).

And that's it. If you play like that you may get more opportunities for tactical shots should your opponent blunder, but you will not be playing good chess. The sample game of his that he gives on pages 63-71 proves my point. It is completely incoherent and reminds me of early computer games.

My overall point is that de la Maza may give you some tools to improve your rating, but, as even he admits, these tools can only take you so far. Meanwhile, his methods will encourage you to play less like a person and more like a machine, for both good and ill. Like a chess computer, you may see more tactical shots than your opponent and therefore be able to hit him hard when he makes a tactical mistake. But you will be completely reliant upon your opponent to make a tactical error in order to win. Once you start playing better opponents (or computers for that matter) you will have no clue what the hell is going on. And because you will never have the accuracy nor the brute-force calculating ability of a modern computer, no matter how hard you work at the Seven Circles, you will never be able to perform above the level of a 10 Mb machine running a 1000-positions per hour search function with a weak evaluation rubric.

Think about it. Do you really want to train to play like an old chess computer? Or do you want to play good chess and use chess as one way to become a better human being?

I can sum up the best lessons of de la Maza in a few words: study your tactics as often as you can and keep track of your performance. But if you really want to be a good chess player, you should read some books--and good, well-written chess books with some real substance and ideas. At the very least they will make you a more literate and well-rounded person.


Blogger Qaundoman said...

I consider this a fair and objective assessment.

I appreciate the fact that those who study The Method are all serious about improvement. I also appreciate, moreover, that some of the followers of The Method--the Knights Errant--have an esprit de corps and have contributed greatly to chess blogging overall. But I think those two positive correlations with the RAPID CHESS IMPROVEMENT method have less to do with the quality of De La Maza's method, and more to do with the excellent character of the Knights Errant themselves.

Mon Aug 22, 08:19:00 PM EDT  
Blogger St. Patzer said...

Excellent write up, and very balanced commentary Michael.

I take your recommendations as valid, coming from someone with a long and continuing association with the game..

when is your book coming out ?

The one I want to read...

Thu Sep 07, 11:42:00 AM EDT  
Blogger Blue Devil Knight said...

A great review. I have recently started working through master games (McDonald's 'Chess: The Art of Logical Thinking'), and it is exhilerating. To see into the mind of a great chess player, on every move in the game, provides a context for the tactics. To some degree, it has gotten me to think of tactics as a means to help me develop a good position, rather than vice versa. Clearly both are important, and they fed off one another, but it is nice to be seeing the big picture for once.

Thu Sep 07, 03:32:00 PM EDT  

Post a Comment

<< Home