Thursday, May 17, 2007

The Weakest Link

The rough draft of my personal study plan is directed largely at studying a large number of master games, for a few reasons. One is that I haven't played over that many annotated games, and when I have it's been sporadic. Another is that I feel the need to be exposed to new plans, ideas and moves (for an example of one plan I found striking, see "When Pawns Lie"). I can tell from my own play that I have a tendency to play certain moves stereotypically and superficially....for example, my opponent plays ...g6 and I play Bg5 and Qd2. I also have a tendency to resist changing the pawn structure and not use my pawns dynamically. I need to see more examples where counter intuitive moves are played, moves that make you think: "they can actually play that?!?!".

However, this plan is incomplete because I have not addressed the most important thing I need to do to improve as a player. The biggest gains come from fixing your most glaring weaknesses. I think that for myself, and for most club players, one of the best ways of improving your playing strength is to make sure you're playing "real chess" on each and every move.

I am using Dan Heisman's definition of "real chess" from his classic article archived at The Chess Cafe:

"REAL" Chess - You select
candidate moves and, for each, you anticipate and evaluate all your
opponent's main candidate moves (especially all checks, captures,
and threats). If you see a threat you cannot meet, you almost
undoubtedly cannot play that candidate move; instead, you must
choose a candidate move that allows you to meet all threats next

whereas I am prone to playing:

"HOPE" Chess - This is NOT when you make a move and hope
your opponent doesn't see your threat. Instead, Hope chess is when
you make a move and don't look at what your opponent might
threaten on his next move, and whether you can meet that threat on
your next move. Instead, you just wait until next move and see
what he does, and then hope you can meet any threats.

--not on every move....just a small number of them. However, it only takes one bad move to lose a game. As Heisman puts it:

A Chain is Only As Strong As Its Weakest Link
The best way to introduce the second part of my explanation is to
make an analogy. Suppose you build a home where the temperature
is -20 degrees outside. You decide on a one-room home with four
walls, a roof, a floor, and a heater. You decide to save a little time
and material by finishing the four walls, the floor, and half the roof,
but the other half you leave open. Even though you have completed
over 90% of the structure, the temperature inside your home will
still be about -20 degrees with half your roof open. If you want
your inside heater to be effective, you have to enclose all of your

The cold home analogy is similar to what happens when you play
Real chess for 90% of your moves, but not for the other 10%. You
think you are a good player, but weaker players beat you when you
let down your guard for that 10%. In order to be a good player, you
have to at least try to play correctly on every move, not just most of
them. Consistency is important: remember that your chain of
moves, in many cases, is only as strong as the weakest link.

I think the common misconception is that blunders are just tactical errors
. Tactics are involved, and better tactical vision will certainly help reduce the number of these blunders, but studying tactics is treating the symptoms and not the disease. Consider the errors you make in a game. How many of the critical errors (those that change the expected result of the game--win, lose or draw) were immediately obvious to you, and how many required some serious thought or computer assistance? If you had to think about why your move was bad, or why your opponent's move worked, then the problem is related to some other facet of your game. However, if your error provokes an immediate "D'oh!", then it wasn't your tactics that were faulty but your thinking. If:

  • shortly after you make your move you suddenly realize it was a stinker, or
  • your opponent makes an unexpected move whose strength is immediately obvious, or
  • your opponent makes a move that you didn't consider, but it's immediately obvious that you should have,

then your thinking process failed you, not your tactics.

Another helpful article of his is this one where he explains the critical importance of playing "real" chess reliably, and of using your time efficiently:

The interesting part about both Real Chess and Time Management is
that both have to be practiced 100% of the time – 98% does not nearly
work. For example, if on 98% of the moves (49/50) you play correctly,
but on one move you decide to just relax and “see what happens”, that
can be a disaster. By missing that one move each game you will
consistently play hundreds of points weaker than your strength would
have been if you had played every move carefully. It is similar with
time – if you play even one move fast that may be enough to cause you
to lose and, if you play too slowly and then have to play quickly
during time pressure (as many top players do), then again just one big
slip at the end may easily be enough to cost you the game.

I seem to be pretty good at time management in actual tournament games, so what I need to concentrate on is the "real chess" aspect. When I return home, I'll edit this post to give a couple of examples from tournament games of gross failures to play "real chess" reliably.

So What's the Solution?

If the solution were to just read Heisman's articles and *poof!* problem-B-gone!, I wouldn't have to write an article about this. The critical question is: how do I train this skill?

It seems easy enough: when considering candidate moves, and before making my chosen move, ask yourself what your opponent can do and whether you can handle his response. In practice, I have trouble doing this on every single move. For example, one error I constantly catch myself committing is "use my time analyzing A and B, be dissatisfied with the result, and play unanalyzed move C". It seems that being aware of how you are supposed to think is insufficient. What's needed is to reprogram your brain.

As an analogy, consider golf. One thing you train is to be able to consistently pull off a certain shot, such as sinking all 3-foot putts or chipping a ball onto the green and 3 feet from the hole. After practicing a shot again, and again, and again, the "right" way to do it becomes programmed in your "muscle memory" and you can just execute it with minimal thought. When a player decides to make a change in how they hit the ball (e.g. wind up a bit tighter for a more powerful drive), even if it's a small change, they need to test it again, and again, and again, to "reprogram" themselves.

Returning now to chess: I think that playing too much Blitz, especially online, programs your brain through repetition to play superficial chess. It's impossible to play "real" chess reliably during blitz. I suspect the psychology behind this is like rats pushing a button for a treat, or a couch potato constantly flipping channels with the remote. You perform an action, you get some stimulus as a perform the action, you get the stimulus....

The solution would appear to be to play more games at long time controls and to absolutely force yourself to play "real" chess throughout. Where I currently live, however, there are limited opportunities to play live games at standard time controls. After my move, I should be in an area with a lot more chess activity, so hopefully I'll be able to play proper chess more frequently.

Until then, the other solution would be to set a program such as Fritz to a decently low level and practice "real chess" there. I've only made a few attempts at this to date, and have had trouble sticking to my guns. One problem is that it seems hard to set aside several hours to play against a computer rather than a person. Another is that there's still the "TV remote syndrome", where it's too easy to just hit a button and find out the result of my chosen there's that "takeback" button where I can pretend that my overly-hasty move never happened. However, if I'm serious about improving this may be the most important thing I can try. I think I'm going to try and keep a game score and clock running, to try and emulate a tournament game more closely.

Another thing that may help is how I read over annotated games. I tend to go on autopilot and play through most of the moves uncritically, except where something grabs my interest and I feel like exploring the variations. Instead, I should probably make myself do a quick "real chess" scan after each move and try to anticipate the next move played.

I would be interested in hearing from anyone that has successfully employed Heisman's suggestions and turned their game around, and how they managed to do it.


Blue Devil Knight said...

It's sure hard to do it every move. I still don't. One thing that helps me is to be aware of times that I am tempted to play hope chess:
1) Tired (especially after a long think on a previous move).
2) Analyzed the move on my previous move, so I assume my previous analysis was good.
3) Cocky: the person is not as good as me.
4) Self-hatred: I perceive I am losing the game, and let it get me all depressed and I get lazy.
5) Excited: I am ahead.
6) He has low time: I want to move fast to force him to use up his clock and start playing sloppy. Thing is, I start playing sloppy.
7) Opening rhythm self-deception. The first non-book opening move I often feel like I am still 'in book', in a rhythm, and so move without doing the required analysis.
8) Time trouble: this may be the only time it is justified. If I am in horrible time trouble and simply have to move, there ain't no time to analyze.

Those are the main ones I can think of. Any others people have?

At any rate, being aware of the above has started to help my play: I remind myself that playing hope chess is always a mistake (unless I have like 1 minute left on my clock).

Wahrheit said...

This is an excellent post, and I'm linking to it in my post today about how I played 30 moves of Real Chess against a guy rated 2113 FIDE, obtained an even endgame, played 1 sloppy move and was lost...that's the way it goes, absolutely, and I thought my readers could profit from your thoughts.

By the way, is the name of your blog based on the book by Kenneth Mark Colby?

Grandpatzer said...

wahrheit: no, I had to google in order to find the book you were referring to. I was playing off the title of one of Nunn's recent books, although Soltis' "Grandmaster Secrets: Endings" was also floating around in my head.

I thought the blog title was appropriate for someone that understood more about chess than their OTB rating may suggest. Thus, "Do As I Say, Not As I Do" :D If I could just follow my advice I'd be a much better player. One goal of the blog is to coerce me into doing this.

transformation said...

i had to dig deep into my memory bank for the precise name--haunted all day today with the gnawing sense that i knew the origin of THIS idea but didn't have it on the tip of my tongue then had to recall it--and knew it as
Collins, just didn't know the first name.

perhaps most imporantly of all: this idea, that the erudition of a complete chess player necessitates their being familiar with some 2,000 chess games originates with Bobby Fischer's chess teacher John Collins (who Fischers mother asked to teach HER son!),

who recommended a noviate start with the early masters, through to capablanca and alekhine, and so on, progressing to bottvinnik, so that the student burns the progress of chess into their brain as an evolving phenomenon, that was

as you probably already know, this idea recurs again and again in modern chess, whether it be a soviet idea of 300 key chess positions, dvoretskys idea of first 40 key endings rather than 2,000 rook endings, or gm-ram, etc.

see my recent post for some elaboration of my own take on review of GM games among other comments, if i may pls suggest.

thank you again.

Ryan Emmett said...

Great post. I've recommended Heisman to others, but I've never gone through the archive of his articles myself methodically from the start. It's about time I did!

James said...


You could slightly alter Heisman's suggested method:

1) List your candidate moves;
2) Pretend it's your opponent's move and find his "killer" move;
3) If any of your candidate moves fail to this "killer", don't analyse it.

More to the point, if none of your candidates work, then you're not choosing candidate moves properly.

So, one might as well change the above order to the following:

1) Pretend it's your opponent's move and find his "killer" move(s);
2) Identify candidate move(s) against which your opponent's "killer(s)" fail.

This way, one of two things will be accomplished:

1) You're more likely to find defensive resources in a position where you're under pressure;
2) As well as 1) above, you're more likely to find offensive/counter-offensive resources.

After all, there's little point in spending time identifying a list of candidate moves for yourself, a la Silman, if you then discover - whilst analysing the first one on your list - that your opponent has a "killer" move; all you've done is waste valuable thinking time.

Kindest regards,