Studying Master Games

Do you guys study the games of masters?

If so, how do you go about studying the games? Is it better to just run through the game or to actually sit down and think?

Also, do you guys recommend memorizing master games? What's the best method for memorizing the games?

I have Kasparov's "My Great Predecessor's" Part V, and I must say it's too advanced for me. Yet I still go through the games and follow the annotations Kasparov (and others) give, but I only go through the variations mentally. The only pieces moved on the board are the pieces moved in the actual game.

Other books I have that don't give as many annotations, I often just run through. If I run through enough, I'll eventually have a better understanding of a certain pattern. 

Some people play through tactics, I play through GM games (and sometimes IM so they're at least closer to my level). 

I'd like to start playing solitaire chess! There's are books on it - I need to pick one up.

Both are good techniques. It's instructive to watch the flow of a game without getting distracted by alternate line analysis. It's also good to study the alternate lines to become more familiar with what's good and what's bad. I use both techniques depending on how much time I have and how much effort I feel like putting in at the moment.

I don't think memorizing entire games is an efficient use of time. You might be better off studying the principles behind certain key positions in those games.

I think it's a good idea to study a super-GM game in the following way:

-> Advance the position until it's just out of the opening.

-> Study the position and try to decide what move you would make and why.  Even a good idea to write it down as you go so that you can keep records of your ideas and see how they progress when you look at the same game in a few months.

-> Play the move actually played by the GM and read their reasoning for the move.

-> Consider the same position again with the GM move and see how your move/ideas differ.

I always tell people that it's important to have an idea/plan of why you are making a specific move.  I speak to so many people who play Chess OTB and they say that in most cases they can't think of a move so they 'just' play any decent move.  It's never going to make you a stronger player with this philosophy because stronger players will have a 'game-plan' and you will fall victim to it.  It would be like getting in your car to go somewhere and not planning your route.  You might get there eventually but in most cases you will simply become lost and run out of ideas, the equivalent in Chess is the opponent Checkmating your King.

At least if you know why you are making specific moves then if you still lose you can say that you were beaten by the better player in that specific situation.  You can then go home and work out why their plans were just a bit better than yours.  In many cases the difference between you winning or losing could be something very simple and you can rectify it for your next game.  It's always about learning from your mistakes, if you can do that then you are always improving.

A

i go over them while i'm bored at work and post them in a thread for everyone.

http://www.chess.com/forum/view/game-showcase/famous-games

I agree with him. If you first figure out where to move, and compare it to the GM move it will not only be instructive in general, you will also see your differences in thinking between the GM and you. Trying to understand GM games really helps your understanding of chess as a whole, and similarly trying to find GM combinations will make you tactically more like them over time.

There is something to be said about going over them quickly for patterns, but I prefer deep analysis because with just a quick playthrough you probably wouldn't fully understand what was going on which makes it harder to learn. If you note some patterns GM's play, you better know why they do them; one time a long time ago I see Kasparov play h4 in a bunch of 1 d4 games (but they were in specific circumstances), so I try it for myself in positions where it doesn't make sense!

So should I first review many games at first just to get pattern recognition down then sit down and analyse?

It's up to you. Try all of the methods for a while and see which one works best for you, because there are a lot of good ways to go over a master game.

There is also the possibility of doing both, I have a few books, such as Nottingham 1936, Zurich 1953 and so forth where I go over the game slowly. Then I have such books as Kasparov's Great predecessors where I go over the games pretty quickly, and usually only once.

Interesting topic. How many people on here memorise master games, and if so, has it been useful? Or do you think it is better to play though more games and just get the concepts?

After reading a lot of chess books and not really improving I'm now following Heisman's advice on how to improve.  For intermediate players he focuses on three things, mastering simple tactics, getting a good thought process and playing over a large volume of master games.

He says there's basically three ways of studying master games, blitzing through many games in a very short period of time, playing guess-the-move and studying the games in depth or a halfway approach where you play over the game thinking about the moves but don't spend too much time on any one game maybe 15-30 min per game.  He says all are good but the last approach is the most important for building up a context of how masters usually play different positions.

So that's what I do, playing through games, reading all the comments but skipping long complicated variations that would take me too long to understand if i ever did.  I like this approach.  Heisman (and other coaches like Silman I believe) say that you should play over literally thousands of games if your goal is make it to master.

My favorite games collection is Tartakower's 500 Master Games of Chess which has a lot of games by the old masters.  It's really neat seeing how 'standard' opening traps were actually novelties 100 years ago, like the Tarrasch trap in the Ruy Lopez from Tarrasch Marco 1892.

ps, so basically what Estragon said 

I memorized Morphy's opera house game.  You don't have to remember the whole game at once if you chain the moves so each move reminds you of the next.  It's actually helped a lot when playing against the philidor if black plays 3...Bg4.

You may read the following on the topic, but I would not disregard the advice given in the other posts.  Like Elubas said, it's up to you what works best for you.

http://www.chess.com/article/view/the-point-of-studying-master-games-part-one

http://www.chess.com/article/view/the-point-of-studying-master-games-part-two

http://www.chess.com/article/view/the-point-of-studying-master-games-pt-3

I have memorized 4 or 5 key games for the openings I play, so about 20 games in all. I'm trying to get it up to 100, but that'll take some time. It's a way of helping assimilate the ideas of the opening by having concrete examples to think about when ever I get stuck on a move. 

Here's the way I've approached it. (I've rocketed all the way up to 1595 so take it for what it's worth ) 

As others have mentioned, I think it is useful to review games specific to either an opening (which is easiest) or a theme. Example, I've a list of 10 games with the Benoni. I then review the game pretty quickly maybe 5 minutes and try to note the following: general opening lines, where pieces seem to be located, any major pawn thrusts or exchanges, and how the end game ends up (R vs R, B vs N, etc). 

I then flip the board and play thru it with the other color, also noting the same things as above. I've found it is really important to see it from the other side.

I then take 5 to 10 minutes on the key points I've noted above.

I'm a big fan of Neil McDonald's book "The Giants of Strategy" where he explores key points such as the 7th Rank, Outposts, Pawn Thrusts, etc so I try and note these themes.

Now (here's a key point) I wait for about 30 minutes to an hour and then replay it again, trying to anticipate key points. There is a lot of cognitive science that supports the idea that a review some short time later enhances recall by a significant amount.

Finally I might review the game again the following day.

Total time with the game review should be 30 - 45 minutes.

 Thanks for all the suggestions. I shall give this approach a try as it makes sense.

Playing though master games help to develop a flow in your own games.

Bronstein recommended something like this:

1) First play over the game quickly in a few minutes. Only the game moves, you read the annotations but don't play through the sidelines.

2) then if you find the game interesting play through it again slower this time (perhaps 20 min.) looking closer at the key points and annotations. (If you didn't find the game interesting then skip it and go to the next game)

3) play over it again, this time you should be able to do it almost without looking in the book.

I use this method and find it works well. The method is about learning patterns and not deep study. Sometimes I play over a game a fourth and perhaps a fifth time until I have a feel for the game.

The suggestion by jwalexander to play over the game an hour later and again the next day is a good one; I think I will add it to my method.

When you play through the games do you set up a board and work through them that way or use an online board/database?

 if looking through alot games or got lot of variations use online board, just looking at a game and taking your time use the board.

Thanks this make sense. I only ask because it seem to take me quite a long time to set up the board and work through games properly. It seems to take me 45-1 hour to go through an annotated game on the board even trying to do it quickly. I don't know why this is. It seems that I'm wasting time that I could have been doing tactics puzzles instead. I have started to use the online board for shorter demonstrations anyway. It helps a great deal when using a database then all you have to do is to flick the forward arrow.