Is it isefull to learn (grand)master's games?

Hello,

I wonder if it's better to go through a lot of famous games whith annotation to understand every move (if possible) but without memorizing a lot or to memorize every single move of a few games (and maybe their variations) to acquire automatisms or do both?

How would you recommend me to study?

I precise that I am still a beginner and I focus a lot on tactics for the moment but I want to also learn strategy so I thought of this idea. I am currently reading "logical chess move by move" by Irving Chernev so I would use these games first as they are not so difficult.

Thanks,

Playing through GM games are an excellent way to improve.  

My method...and by no means the best or suggested.  I play through a game rather quickly.  Just trying to subconsciously pickup things like piece placement, pawn structure, middlegame ideas, weak pawns, weak squares.  I spend no more than 5 minutes on a game.  

Play Solitaire Chess.  Get a GM game, play throguh the first 6-7 moves.  Playing the side that lost, cover up the remaining moves, and decide on what moves to play.  Uncover the winning side move, and repeat.  

What is very important!!!!

Use a REAL BOARD AND PIECES.  

So these are the two types of work you do. 5 minutes per game and solitaire chess which may be much longer. That's a really good idea to try to guess the winning moves. I will do that, thanks :)

So these are the two types of work you do. 5 minutes per game and solitaire chess which may be much longer. That's a really good idea to try to guess the winning moves. I will do that, thanks :)

Those are the examples i use.  You can just study GM games if you choose.  What might be of even more use to you is to find games by lower rated players like Expert.  If you can find a database of games from players 1 rating section above you would also help.  Thats kinds like playing people 1 rating class above you.  

Ok 10 games is a good number to start! GodsPawn2016, you gave me an idea but I don't know if it's that good actually. It's to play against stronger players (+200) and then if I loose the game I can play solitary chess with your method by trying to guess or remember my opponent's moves or good variations with the help of the computer when I don't have anymore ideas.

urks suggestion is good as it forces you to retain what youre going over.  

Playing people 1 rating section above you is recommended.  

Play correspondence chess against someone 200 points above you is a good idea.  You can write down yomr thoughts, ideas, candidate moves, analysis, erc. during the game.  After the game you can review, and play solitaire chess.  

If you want to play an unrated game send me a game request.  

I've studied Fischer's 60 Games, Alekhine's, Keres' and Tal's games, but the BEST book of games I've ever seen is Dan Heisman's The World's Most Instructive Amateur Game Book. It talks about the things that I have to correct that barely get a mention in most GM books like "One of the sub-goals in chess is to use all your pieces all the time" and "If your opponent gives you room to advance your central pawns safely in the opening, you should usually do it." - while showing why or why not that's occurring in the game.

What I get out of the GM game books are ideas about how to transition from the opening to the middlegame.  I look to see where the player attacked and hope the game says something than, "This move was clearly better than Nxe5, Bxe5 etc."  Maddeningly, too often those books do NOT say what the GM planned for the attack.  There's seldom anything like, "I saw that he was overprotecting the base of his pawn chain so, instead of attacking the base as theory recommends, I attacked the other end."

I'd rather study a good middlegame book. Kotov's discussion of when it's right to castle on the opposite side in his excellent, 50-page chapter "Strategy and Tactics of Attacks on the King" in his and Keres' The Art of the Middlegame has won far more games for me than any game collection.