how do you study master games?

i try to study master games but 

i find I dont know enough to really get much out of them. 

how do you study master games. 

what advice have you been given?

I heard 

yasser Seirwan say that you should pick a hero and play as his partner 

guessing his moves and watching him solve the problems the opponent presents. 

is this what you guys do?

I just replay through them, picking up ideas from how they moved their pieces with different structures and typical ideas of different positions.

it is tough to match the moves. with the hero concept. but I am doing it with some Fischer Games.

so out of 28 moves I may guess three or four right.

Its a baseline anyway.

Focus on the games of a player that plays the same opening as you. Maybe you use a different opening for black than they do, then choose a second player for that. It might also help if you get books of annotated games, that explain why the moves are being made. There are many game collection books of this type.

Choose a player that plays the same opening as you. If you play a different opening as black, choose a second player for that. It's also helpful to get a few game collection books of annotated games where they explain why the moves are made. There are many books like this.

A much needed source of master games is the predessesor series by Kasporov, not for the games so much but the elevation of chess evolution through the stages of development. Understanding how chess has evolved throughout history not only improves your game, but garners a deep respect for the evolution of this game and how it evolved.

Might I suggest a subscription to chessgames.com? They have a wonderful feature (for which you pay about $25 per annum) called "Guess the Move." You are given the first few moves of a master game, and then you have to guess the move from then on, playing as the winner. You can choose from over 14,000 games now, and if you want to focus on 19th century, or one particular player, etc., it's easy.

When you guess the move correctly, you get 3 points. If you guess an equally good move, but not the one the master played, you also get 3 points. You get 2 or 1 points for "OK" guesses, but you can actually lose points for blunders.

Your goal is to at least beat the par score for the game. I find it to be fascinating, relaxing, challenging, and instructive. You can try a few for free before deciding to plunk down $.

You can also download Fred Mellender's free "Guess The Move" here:

https://sites.google.com/site/fredm/

Go to the bottom of the page and click the corresponding down-arrow on the right.

With Fred's GTM you can play "Guess the Move" with any game you have  in PGN notation. A chess engine will grade your moves.

great suggestion because what I did is study the master games with GTM in mind. There is a lot of quality games to study in master games. My suggestion is to emulate the player you want to study. There are countless opening theory moves I was able to gain from these games. I also enjoyed a lot of the players reliving their immortality in these famous games. Good luck and enjoy your endeavor! :)

Ziryab's got a good point about looking at older master games where the logic is more clear.

I stopped playing chess in the seventies. When I returned a few years ago I was surprised at how much more subtle 21st century chess had become.

There are books where instructive master games are given and explained move by move. Also  Dan Heisman made some great videos for ICC where he explains games in great detail. 

When people tell you to study master games, here's the many things they don't tell you:

1. The lower your rating, the earlier the master you should study.  A 1000 player should not be studying the games of Kasparov, Karpov, Kramnik, Anand, and Carlsen.  A 1000 player should be studying players like Steinitz, Lasker, and Capablanca.  Mid-tier players, like 1300 to 1700, should be studying Alekhine, Botvinnik, and Tal.  Experts and above should be hitting up on Karpov, Kasparov, Kramnik, Anand, and Carlsen.
2. Don't just pull out your Fritz database and go thru a bunch of moves in unannotated games.  Find books on these players with annotated games.  Then, go thru the games twice!  First, just play the main moves in the game, trying to figure out and understand why each move is played.  Then, after you finish, go back and play thru the game again, going thru the annotations.  A great source is many of the Move by Move books by Everyman, like Capablanca Move by Move, or Nimzovich Move by Move.  If one game takes you less than an hour and a half to study, you are going too fast!  The first time thru should take you about 20 to 30 minutes, and the second time thru, going thru the notes, should take an hour minimum for a short game, and most of them should take an hour and a half to two hours to do.
3. DO NOT use a computer screen.  Pull out a board and pieces, and play through the moves like you were at a real board playing against a human sitting across from you.  What this does is when you get through a long variation, you have to replay quickly through the previous moves to get to the point you were at.  So if there is a long variation after Black's 8th move, then after that, you reset the position, and have to play thru moves 1 thru 8 quickly to get to move 9.  Then there is a long deviation at move 12.  Guess what, you have to reset and play through moves 1 thru 12 again (not going thru the notes of course in this case).  What this does is because you had to play through the earlier moves 5 to 10 times, the opening moves and ideas get engrained in your head without actually doing deep opening study.  This will get you through your opening moves in a real game without having to study a single opening book until you are about 1800.  Then you need to start thinking about understanding theory rather than concepts.

Hope this helps.

great advice thrillerfan! +1 :)

just remember Zurich 1953. great quality post modern ideas. In fact this is the golden age of chess!

You can play "solitaire chess" where you cover up the moves and try and figure out what was played.  

What i do is, go through a game pretty quickly (2-3 minutes) to pick up general ideas, patterns, etc.  Then i will go back over it the next day (30 minutes) taking notes.  Then i compare my notes with the analysis in the game.  

Something else to consider is finding games by players at a level 200 points, or a rating class above yours.  I can remember the guys name, but he wrote a book called "How to beat an A player" where he gives insight on how class players think.  

Find a database of 500 games by a player you like, stick it on autoplay at 1s per move, grab a brew and find a comfy armchair, sit back, relax and watch the games. Should help your pattern recognition a lot!

One problem a lot of people have is that they don't actually know what high-level ("real") chess looks like.

you're right. What I do is focus on the opening ideas and emulate the excitement of each move. Reliving the past for a historical reference of a new and evolved plan of piece devolopment. I like it. +1 :)

Other big ones are New York 1924 and Carlsbad 1929

chessgames.com is now $29/year.

http://www.chessgames.com/perl/tour