Why should we analyse our own games?

Lewis Turner, 29 November 2024 One of the most fundamental and important pieces of advice any chess coach or strong player will give you is that to improve you must “analyse your own games.” Why is this? First, it develops your analytical skills and reasoning. Analysis is an important skill and mastering this will help you in developing other areas of your game, to name a few: intuition, calculation, schematic thinking, positional understanding etc. Second, it shows you where you are making mistakes, and this can be an important first step to improving your overall game as if you appreciate where you are making mistakes you can take steps to avoid this in future and even design a training program with the aim of minimising these mistakes! Players often do not look at their own games at all and this is sad for them. They are missing out on a great learning opportunity. I understand losses are painful though attributing them to some other factors is not going to cut it if you want to improve. There is always a reason why we lose a chess game and facing up to this and doing some honest reflection and analysis over time can really help your results to skyrocket.

The Post-Mortem

It often saddens me how often the post-mortem tradition is skipped with people preferring a quick flick on their phone engine. This wastes a great learning opportunity! Especially when you play a stronger player their insights can be most valuable. I always take the opportunity to analyse a game with my opponent regardless of the result if possible. It can also be interesting to record your thoughts and conclusions and compare this with an engine later. So, piece of advice number 1 is to always offer a post-mortem and resist the temptation to reach for your engine.

“Human first, engine later!”

Of course, engines will give you a very brutal appraisal of your performance almost every time you play and point out your mistakes in excruciating detail. What an engine will not do for you however is identify why you made a mistake. It will also not tell you when you made a good practical decision even though objectively the move you played is not best. Before resorting to the engine, go through the game and add comments on what you were thinking during the game, and any variations you calculated. If possible, seek the assistance of a player who is stronger than you (or a coach if you have one) and ask them to go through the game with you again (and this is important) without the help of an engine. I picked up a useful piece of advice when analysing a recent tournament game over dinner with 2 strong players that my calculation is “too superficial and often pessimistic” which means I give up too early when calculating a promising line and this can lead to me avoiding the line altogether and picking an inferior move. This comment really made the penny drop with me. It would have been useful to have known this 9 months before this conversation happened when I played the following game and the below position arose: Facing the threat of Nc2, I could not see a good way forwards as I would have to part with the exchange on a1 or my e3 bishop which is probably my best minor piece after a later d4-d5. My preparation had only covered Bb4 which is the engine’s suggestion for black instead of Nb4. I thought for 45 minutes in this position [45 minutes?! Have you read my article on time management? – Martin], I felt Nb5 was correct, but I could not make it work and instead opted for an inferior move and lost. I concluded after Nb5, Nc2 I had insufficient compensation for the exchange. Was I right? No! I recorded this in my notes to the game which helped me come back to this critical position. What was my mistake? Firstly, I trusted my 2250 rated opponent a bit too much and secondly, I did not evaluate the resulting exchange sacrifice correctly. Let’s think about what could have happened if I had trusted my intuition and calculation. After 14. Nb5! Nc2?! 15. Qxa5! the position is not that hard to evaluate, and the calculation to reach it is not too difficult either as we are only looking to answer the question: “can black take material with Nc2?” If black takes the rook or bishop, white has Qa7+ followed by either d5 or a5, eliminating the b6 knight when black must give up massive material to stave off the attack. So, you can see how this mistake in calculation and evaluation of the resulting position (“the what(s)”) were the result of a pessimistic thought process (“the why”) leading to this mistake and countless others.

Checking the opening

At some point you will have run out of or forgotten your opening theory, or your opponent will have played something you are not familiar with. You should go back to your source (openings book, YouTube video etc) to see what the recommended moves were. If you decide that the new move in the game was actually good then you will need to update your opening notes to reflect this.

How to use the Chess Engine, the “what” and the “why” of mistakes

The chess engine is useful for identifying “what” you did wrong but it will not tell you “Why” you went wrong, and this is far more valuable to any improving player. I think I have shown this with the example above though as this is important, I will offer some more thoughts here. I think a chess engine is most useful in two areas when analysing a game you have played: 1. Showing points where the evaluation moved in an unfavourable direction because of the move you made. These are your mistakes “what” you have done wrong Try to take 3-5 or so from every game that you play. 2. For more advanced players (and ambitious improvers!) if you have recorded your thoughts and conclusions in a particular position, it can be useful to analyse that position with an engine in more depth to see how your conclusions differ from the engine. In this way you can help to calibrate your intuition and understand where your evaluation differs from the engine and why and can lead to many important lessons being learned. Only by challenging your conclusions can you improve your understanding of the game.

Putting it all together

So, you have done a post-mortem and been through your games with a stronger player and an engine and recorded your thoughts and conclusions. What next? I believe it is important to keep a record of the games (I use a Lichess study though ChessBase is fine too) and record your annotations. Here is an example of what an annotated game should look like. Eventually, you will end up with a file that contains your own games analysed with annotations. After a set number of games (try to play at least 10) you can draw conclusions and look for patterns in the mistakes you are making. The file is also useful if you are preparing to play one of your opponents again.

Ask yourself:

• Are you making the same mistakes consistently? • Why did I make the mistake? • Can I categorise the mistake? (Endgame, calculation, positional understanding, lack of opening knowledge, time management etc). • When you have done this and drawn conclusions you will be more aware of the mistakes you make which alone can help you. • If you want to go further, you can structure your training with resources to help you avoid these mistakes! • For example, if you consistently make mistakes when exchanging pieces, you may wish to incorporate puzzles to help with this. I can recommend an excellent book “To exchange or Not – The Ultimate workbook” by Eduardas Rozentalis published by Thinkers Publishing. • Or if you often enter rook endgames equal but lose them a study of practical rook endgames may be of great benefit to you! Perhaps you could add the games of Akiba Rubinstein to your study plan!

Online blitz games

The comments in this article mainly relate to classical games where you have recorded the moves, but you should also do some analysis for online blitz games. Clearly you aren’t going to spend a lot of time on a game that only took a few minutes, but it is worth running the game review feature available in Chess.com and Lichess to see what tactics or key ideas you missed. Also you should do the same review of the opening moves as you would for a classical game. This is an excellent way to build up your opening knowledge.

In Conclusion

- Analyse all your games, especially the ones you lost. - Ask for post-mortems. - Record your thoughts and assessments BEFORE consulting an engine. - Try to find 3-5 mistakes from every game and draw a conclusion about each mistake. - If you have a coach or strong player to consult, use them! Even if you email games to a stronger player for their input this is better than nothing though it is best to sit down and explain things to them face to face. - Ask “why” you make a mistake as well as “what” the mistake was. -Understand why your assessment of the position was different to the engine. This can help pinpoint mistakes in chess understanding and evaluation. - Keep records of your game analysis and look for patterns and draw conclusions. - Become aware of the mistakes you make and change your thinking process / training to try and avoid these in future!