What are the best ways to analyze games?

Was wondering if people could give advice on things they use to analyze their own game. I am currently using Fritz blunder check after I finish my games, by I am really not sure how much I am getting out of it. I feel most of the work is done by Fritz, and I just scan the results to see where I messed up. I can't really see myself improving by just doing that

Was wondering what other people do when they analyze their own games.

Originally posted by meyekal
Was wondering if people could give advice on things they use to analyze their own game. I am currently using Fritz blunder check after I finish my games, by I am really not sure how much I am getting out of it. I feel most of the work is done by Fritz, and I just scan the results to see where I messed up. I can't really see myself improving by just doing that

Was wondering what other people do when they analyze their own games.

good advice

Originally posted by chesskid001
good advice

Fritz' blundercheck is rubbish.Last time I used it I gave checkmate with a pawn and it annotated it with a ? cause according to fritzie mating with the rook was way better 😕🙄
Since it was mate it's only laughable but imagine this happening in a complex middlegame.You could waste some serious time trying to figure out why your move is so wrong while it's perfectly good.

My advise to analyse your game: first go through it yourself,annotate your moves,try to figure out where you went wrong and try to find improvements.Then use an openingbook to check how well you played the opening and if you had the correct middlegameplan(s).Then check how well you played the endgame,either with tablebases or an endgame manual.Then,if you must,run it through fritz.

I never analyze my games. When my games are finished I archive them never to see them again. My advice would be to play through your games move by move. Study each turn for a plan and play that move. Maybe with each move you may learn how to play a stronger game. Try not to know which move you played in the game while you are studying the game for a plan. Be sure to study your oppenent's side of play for threats that you may need to prevent. When you have finished your study you may now consult a opening database for a better opening play and allow a chess program to analyze the game for you. If you can you may hire a Master or GM to analyze your games as well. I did see a GM offering to study your games for you on some chess server one time. "You" would be any person who has seen the add. For a GM to study your games I suggest that you give a good sample of wins and loses and that the games would not have a data base used in them.

when In a seroius mood, I do this.


create a notepad document, And as the game progresses, I put all my analysis, thoughts, etc, etc. and save it.

do this right till the end of the game/ or till the win is VERY DAMN EASY.

then If you lost, go over the game with an engine(and yourself) and what improvements it found.

then, most importantly go throught your notepad document, and check to see how valid and "sound" your variations and thoughts were.

the point to this, is to avoid the bias of hindsight -- if you know exactly what you were thinking at the time you can judge your positional evalutations very critically.

I never use any engine analysis, but I'm sure it works for some. What I do when I'm actually in the mood to analyze a game is to set it up on a board and play through it until I spy something about the position that irks me, something that I think should have been avoided. Then I work very slowly backwards until I find the move at the source of the flaw. Usually at this point I'm done with the game, and I don't analyze it anymore, so perhaps it's not a good system. Yet, that one flaw I do discover is a doozey as in the end I feel that I understand that flaw and thereby add to my chess understanding.

Not saying this is a good thing to do, just reporting what I do.

I haven't bought Fritz and only use GNUChess, so my engine is much worse than yours, but I still occasionally use it to analyse games. As someone else said, I don't rely on it much to tell me my move is right, as to tell me -- show me -- my move is wrong. For the most part, I find it only useful for showing me tactical shots I may have missed, but I think thats better than nothing.

Analyzing your own games is useless in my opinion since if you are unaware where you went wrong during a game it is unlikely you will find it later. Even if you found where you thought you made a mistake how would u know u were right? (unless the mistake was glaring).

I learnt all my chess playing through 1000s of annotated GM games in books THAT is the way to analyse in my opinion.

If you are really lucky you might find a high rated player who could look through finished games for you.

I am happy to help you (my rating is 2130) if u are serious about improving.

The problem with annotated GM games books that there are too much analysis without enough explanation. For example, in the book "Barraging the Barricades" there are so much analysis that I gave up doing all of them on my chessboard and just followed the mainlines instead. Without the complex variations, I understood the main point of the chapter clearly. However I am worried that It won't get me too far.

Any ideas or help?

Yes i agree not to follow all the variations but concentrate on the mainlines. The huge help is that they tell you which moves are good and which are bad. As i said if anyone REALLY wants to improve you have to walk through hundreds of games and not many people will do that. If anyone is interested I recommend Capablancas best 100 games as the best book for positional play, and Fishers best 60 games as best for all round play. Alekhine was incredible as was Tal, but you wont learn much from Tal because his play was so exciting (and dangerous!). If you play through hundreds of games you will find an amazing difference in your play because it will all have seeped into your mind. You cannot improve at chess quickly.

What I do, is thoroughly (at least as thoroughly as a weak player like mysel f can do) analyze my game using the method described in The Chess Exam and Training Guide. I do this by myself, then I run it through blunder check on ChessMaster 10th. I don't analyze all my games, only the ones which I consider to be worth analyzing. A good idea mentioned in this thread is to write down all your ideas on a notepad while playing. You've partially analyzed it, and you can see your thought process when you make an error.

Set Fritz to analylse the game but at a high level and looking for moves that are .5 pawn or more better than your moves.

This helps identify the key swings.

Then go through it carefully and at each of the swing points look to see why your move was bad and if you can improve upon Fritz. Use infinite analysis to help you play through and assess alternate lines.

Don't accept everything Fritz says. Computers can be notoriously bad with gambits and sacrifices seeking to regain material too quickly so if your gut tells you a sac is good play it through. 3 or 4 moves in Fritz may suddenly change its assessment.

However where Fritz is good is in those tactical errors that win or lose quickly. I analyse very few of my games with Fritz but when I do, I find that most games swing quite wildly in Fritzs assessment with many moves only being Fritzs second or third choice except when it becomes obviouly best or the only defense or winning method when given time I will find them myself and time is something we have plenty of here.