by Dan Heisman

This exercise was suggested to me by FM Steve Stoyko, and I strongly suggest it for intermediate players who wish to improve their visualization and evaluation capabilities.

First the reader should find a rich middlegame position. Suggestions would be go go to most any Kasparov, Shirov, or Speelman game, or perhaps from the books How to Think in Chess or Genius in Chess. Take out a paper and pencil.

For this exercise the idea is to write everything you can possibly visualize from the position, like you were playing the game without a clock and you had to see everything before you move. Write down every line that you look at, along with that line's evaluation. To be done correctly this should fill up 2+ sheets of paper and take 45 minutes up to 2 hours or so for just this one position; the exercise is only done on one position at a time (that is enough!). You should analyze that position without moving the pieces. Each line you analyze should end with an evaluation (such as "White is a little better or +/= or .3 pawns or however you wish to state it). As with any move, the evaluation of your best line would make that your Principal Variation (PV) and among all the lines that one should be noted as such.

When you are done, either take the analysis to a good instructor, player, or software program. Look at each line to see how well you visualized the position (any retained image problems, etc.?), and also compare your logic (was that move really forced?) and all your evaluations, as well as your PV.

Steve claimed that each time he did this exercise he gained about 100 rating points! :)

A summary of this exercise I posted for Chessville:

1) Find a fairly complicated position

2) Get out a pen/pencil and paper

3) You have unlimited time

4) Write down every (pertinent) line for as deep as you can see, making sure to include an evaluation at the end of the line. This will likely include dozens of lines and several first ply candidate moves. Evaluations can be any type you like:

a) Computer (in pawns, like +.3)
b) MCO/Informant (=, +/=, etc.)
c) English ("White is a little better")

5) At the end state which move you would play and it's "best play for both sides" line becomes the PV

6) When you are done, go over each line and its evaluation with a strong player and/or a computer. Look for:

a) Lines/moves you should have analyzed but missed
b) Any errors in visualization (retained images, etc.)
c) Any lines where you stopped analyzing too soon, thus causing a big error in evaluation (quiescence errors)
d) Any large errors in evaluation of any line
e) Whether the above caused you to chose the wrong move
etc.

Chessville question about this exercise: I don't understand your point: "The key is the amateur's evaluation of every line... you will have your instructor (or Fritz or whatever) compare your evaluation of every line, resulting in a really good evaluation test". How is it a "really good evaluation test" to analyze a single position from a Kasparov or Shirov type game for a hour or so?

I can see how it's a good calculation/visualisation exercise - totally agreed. I've done it in the past for this benefit and I'd do it again. But I'm just not understanding the evaluation benefit?!

Answer

: Your question is very good (I guess if you misunderstand that purpose of the exercise, that would help explain my observation as to why so many players are missing out on a valuable resource!).

Most players are very poor at even-material evaluation. Therefore they make bad moves because, assuming they evaluate potential outcomes of various candidate moves, they choose a move that is not best because they erroneously think the resultant position(s) from their chosen move are better.

The second (non-analysis) aspect of the Stoyko exercise is to evaluate EVERY line that you examine in the tree - that could be dozens or even possibly hundreds of lines for one position since the Stoyko position has unlimited time. By comparing your evaluations of these hundreds of lines with your instructors' evaluation, you learn to improve one of the most critical skills you have - what is good and what is bad and why and how much. It also helps you identify the all-too-common quiescence errors where weak players stop their line too soon and therefore mis-evaluate because they did not look to see what might happen with further checks, captures, and threats.

This capability is so important and its failure so critical that you would think everyone would want to work on it, especially since the amount of work is an hour or two, plus additional time for going over it with someone (or even at worst via computer evaluation).

Here are two common mistakes when doing your first Stoyko exercise:

A player takes a position that was far too balanced and almost still in book (too early in the game). Instead, you want to find "messy" positions which are unclear, usually because one side has sacrificed material for some other advantage. Also, when the line is balanced, then most of the evaluations are fairly balanced, and that is not good practice for evaluation. You can find these in the middle of lots of Shirov games, Kasparov games, Topalov games, or in books like Levitt’s Genius in Chess, or Przewoznik, Chapter 9 of Eingorn's Decision Making at the Chessboard, and Soszynski’s How to Think in Chess, likely even in some of your own games.

A player does not nearly enough analysis (which goes along with number 1, since a much more "regular" position has far fewer lines worth investigating). Usually a Stoyko position is unclear and you would have dozens of lines and hundreds of nodes in your analysis. On a sheet of paper, handwritten, this would take up around 3-4 sides of paper. Remember, since there is no time limit, you are trying to see as far as is pertinent in every possible line, so it can get fairly deep (which is why this exercise is good for testing your visualization).

Begin Chess

Super Admins