I'm a Class B, but I would like to reach Master Class one day. I been playing a lot of OTB games recently, and I would like to spend time studying my weaknesses. My coach noted some major weaknesses of my play. Unfortunately, I currently can't afford any more chess lessons or tournaments, for a two month period. I would like to uncover more weaknesses on my play and become significantly stronger. How do I seriously analyze my games?

If it helps, my chess engine is Fritz 15. I'm not sure how do use the engine. Any advice on that would be greatly appreciated.

  • Try to play the games that you lost with a computer.
    – SmallChess
    Commented Dec 19, 2015 at 11:15
  • @StudentT What mode do you put your chess engine in while your playing your games with the computer?
    – user122965
    Commented Dec 19, 2015 at 11:19
  • Highest difficulty and no analysis until the end of the game. If you want to do this, make sure you only try it from a critical position.
    – SmallChess
    Commented Dec 19, 2015 at 11:21
  • @StudentT After going over the game, do you use infinite analyze or some other mode to find the correct better moves for both sides?
    – user122965
    Commented Dec 19, 2015 at 11:24
  • I do use infinite analysis to find the better moves, but it's usually not enough because you won't understand exactly why. The moves given by the computer assumed best-play (over 2500+ level), so you shouldn't expect your opponent would go for it. You have to understand the position yourself to handle any response your opponent would play. Set yourself some critical positions and play against a computer would help.
    – SmallChess
    Commented Dec 19, 2015 at 11:27

3 Answers 3


I think engines only help for tactical improvements, not strategical. I saw a lot of moments where engine suggestions was far from perfect, strategically.

Also, in endgames, you can't depend on engines even tactically.

I used to use Fritz GUI with strongest engine(for that day) for blunder checks and tactical misses, in Fritz GUI you can do it easily with one click, give the game, leave the computer for 30 mins and when you return you get an tactically annotated game. With this, you can improve your tactics ability.

But remember, use engine suggestions only for finding answers for tactics, if you can't understand why the answer is correct, then you can't benefit of using engine.

For strategy, best way is to find someone who is at least 300 elo higher than you, and analyze games together. Also read every strategy classic, if you didn't read yet.

And for endgames, you need to be a real geek in reading endgame books. Best way is to read quality endgame books, and try to solve endgame studies.

Studying endgame is really beneficial for elo improvement, because people under 2000 elo don't care much about endgame. If you get a good endgame knowledge, then you can go for exchanges happily and solve your problems in endgame. Also, it's said that a good endgame knowledge requires an excellent grasp of pieces and their abilities, which also naturally boost your tactical ability.

My favorite endgame book: http://www.amazon.com/Secrets-Pawn-Endings-Karsten-Muller/dp/1904600883 which is an excellent and complete source for pawns.

Lastly, about openings, don't memorize, but understand. Don't memorize and use popular strong lines just because they are popular and strong, if you don't understand the harmony in the moves, then the opening won't be strong in your hands.

When you use an opening which you understand the strategy within, deeply, you almost always arrive a better middlegame/endgame comparing your opponent, which makes you win the game.

I won a lot of matches against strong players with openings which are quite unpopular, but I understand very well. And I saw these strong players fail to find correct continuations :)

To give an example, I used to play Dutch Defense with blacks successfully, later I wanted to play it with whites, which found silly by others from time to time. But because I knew Dutch very well, it went good with whites too, and I got quite good results.

A bit long but, these are my advices. Hope it helps.


I used to think that computer analysis was very useful for lower level players; now I am not so sure. It isn't helpful if you don't know why some move was made or if to reap the benefit you have to be able to see 8 or 10 moves ahead. We B-players can't do that.

That being said, when I analyze a game, I first let the machine analyze the entire game. I do this at 3 or 5 minutes per move (both sides) while I am at work.

I then examine the results, focusing on the most significant score deviations. These will be where one side missed a shot, or where one side blundered.

Now the hard part. Look at the machine's variation against what really happened. See how the machine punishes the bad move. There is no shortcut here - keep at it until you understand why Fritz chose the move.

Fritz is going to make moves worthy of a savant grandmaster. Some of them will be inscrutable. Don't worry about it.

What is psychologically comforting is that you'll probably find that you had real chances to win every game, even when the game was "lost". The moves are there if you can find them.


1. Start small, be consistent. Keep analyzing.

Imagine analysis like a muscle you can develop. You only have to start, and keep improving it your analysis skill through exercise. Make it a habit to analyze at least two of your game everyday -- one win and one loss.

2. Analysis style.

At first you can analyse games using text and a few variations here and there. By text, I mean -- adding your thoughts about the move, your plans, the back story about the game, your opponent, etc. I'll give you an example -- Observe the style of analysis in Zurich international chess tournament, 1953 by David Bronstein.

The book is famous for its easy to understand style of analysis.

You can also check out how others are analyzing their games.

3. Feedback:

So far you're only supposed to use your mind to analyze the game, NO COMPUTER. In step 3, get feedback of your analysis from stronger players, friends or online communities -- like Reddit's Chess community

This will give you a lot of ideas and insight about how you can improve the process of analysis and about your moves.

4. Computer analysis:

Use computers to assess the accuracy of your analysis and what you have missed. Run full analysis using any chess playing software to see what you have missed, and what the computer thinks about your variation. This will help you improve the depth and accuracy of analysis.

Hope you find this useful. All the best.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.