Next Tournament

How to study annotated games?

I have heard every top chess player say something like this "I grew up studying so and so's games". How exactly do they study the game is the question? Firstly I presume it is annotated in a book. So how does one study annotated games? I have several books that have classic games annotated. My impression is that there are several ways to study and I am curious to know if there is one good way people recommend that is most beneficial in improving one's own game. Below I list some of the methods that I have tried -- I am hoping the chess teachers/players out there might have some ideas about this.
  1. Have a board and play the moves and all variations (and sometimes beyond the text). This seems to be the most obvious method to study. But I often get lost in the crazy lines given and more often I am looking from book to board -- back and forth to just make the move and often even lose track of where in the book's page I was just reading. Also it gets difficult to retrace the moves to revert to the main line. Sometimes a single annotated game takes hours with this method and at the end of it, I am not sure if I actually understood everything the annotator said.
  2. Dont use a board, but instead use the computer and do same as above. The advantage here is that you can retrace your moves very quickly. Here again looking back and forth from the book to the computer is tedious. I am not not sure if looking at the computer is as good as looking otb. With this and the one above, you will have to repeat the whole process if you want to take a look at the same game again.
  3. Another method that I sometimes do is blindly enter all variation with annotations into your computer without breaking your brain to understand it. Once it is done, just go over what you enter on the computer. This is like converting the book to a pgn. Well this is also very time consuming to enter all annotations. I wish they sell CD with the pgn along with every chess book
  4. Yet another way would be to enter the main line that was played into the computer and analyze the game yourself (without the book's annotation) -- entering variations and evaluations and finally comparing your analysis to the notes in the book. This sometimes seems fast, but quite often I enter variations not listed at all in the book and also miss several variations that are listed -- so I end up only comparing my evaluations on the very few positions in the book that overlapped with my variations.
Your comments/suggestions are welcome.


  1. Thank you, Harish.

    What I do sometimes is:

    1. I try to find a game in question in the databese, so I do not need to rewrite it. I make a copy and add variations that are in the book I am reading and go over a few times. If I find something different I will write my annotation and that's all.

    2. I am sure it would be good to study on the big board and use smaller boards for variations, but that seems to take much time.

  2. Harish, interesting insight. I read somewhere that there are 3 phases of analysis (i believe this is according to timman):

    How do you analyse a chess game?

    If it's someone else's:
    (1) Play through the whole game fairly quickly, seeing what ebb and flow the game had.
    (2) Divide the game up into episodes, even if they are only opening middlegame and endgame, and try to describe what happened and who if anyone had the advantage.
    (3) Try and find the critical steps, where one or other side made an important choice (even if it was only to overlook something!).
    (4) Put your account aside for a few days and have a fresh look later. What other ideas now occur to you? Did you make a too-smooth moral fable out of a more fluctuating scrap?

    If it's your own:
    (1) write down as much as you can remember about what you thought about at the time, and
    (2) any thought your opponent shared during a Post Mortem.
    (3) Leave it for a few days and come back fresh, then analyse it as you would somebody else's. Be particularly wary of self-serving interpretations, especially of the "I was winning all the way through" variety. What did you miss?

    For me however, there are a few steps...

    a. find an opening i want to learn.
    b. find a player who plays that opening well.
    c. look over the game quickly without stopping
    d. this is the step thats optional for me... look at the game again in case i think that there was something interesting there in the first run-through.

    so typically this leaves me with karpov, petrosian, miles, and occasionally tal...This is what i did when i was weaker. Now i mainly read books right now im reading 4 or 5...

    Finally, (and i know this is getting long-winded) i almost NEVER use the computer for analysis. I find more often than not, unless its a tactical situation, it teaches you the wrong things about chess, and breaks fundamental principles...granted, the computer is much stronger than myself. Recently i have been using an analog clock to time how quickly i can do tactics puzzles, to make sure that i am taking longer than necessary...(i've been playing to quickly as of late :))

  3. The study of annotated games is an art in itself, 25 years ago I would play through a game twice, first to get a feel for the strategical & tactical struggles that developed and then the second time only to subject the game to a closer subjective analysis. The review of any annotated game with another player also tends to give you a close idea of the truth hidden in any position that way if you have any prejudice it should be nullfied! Barry D