Guides / How to analyse your own chess games

How to analyse your own chess games

Almost every strong player gives the same advice: the fastest way to improve is to study your own games, especially the ones you lost. Playing more games mostly repeats the mistakes you already make. Looking at them, carefully and soon after, is what stops them. This guide is a routine you can follow after any game, from a three-minute blitz game to a long tournament game, in ten to twenty minutes.

Step 1: think before the engine does

The most common mistake in analysis is to open the engine first. The engine tells you what was best, but not why you missed it, and the why is the part that helps you next time.

Before you start a review, spend two or three minutes on your own:

• Where did you feel the game turn? Most players know the moment: a move after which things felt different.
• What were you worried about there? Write down the threat you saw, or the plan you were following.
• Where did you spend a lot of time? Long thinks often mark a hard decision, and hard decisions are where mistakes cluster.

Even a sentence or two is enough. Later, you will compare your notes with what the engine says.

Step 2: run the review

In GChess Review, search your username on Lichess or Chess.com, or paste the game's PGN. The review grades every move and finds the critical moments: the positions where the result was decided. It opens at the moment the game turned, not at move one, so you can start where it matters.

For a casual game, the default Quick strength is enough. For a game you want to study properly, choose Deep in the analysis settings: it takes longer and searches the critical moments more thoroughly.

Step 3: go to the critical moments first

Do not step through all forty moves. Go through the critical moments, one by one, and for each ask three questions:

1. What did I play, and what did the engine want? The engine's line is drawn on the board.
2. Why is the engine's move better? Play its line forward a few moves. Often the reason becomes obvious: a piece left undefended, a check you did not consider, a pawn break that opens your king.
3. Could I have found it? Be honest. If the best move needed a ten-move calculation, the lesson is different from a one-move blunder.

Then try your own ideas: you can play moves from any position on the board and watch the evaluation change. Checking "what if I had played this instead?" is how analysis turns into understanding.

Step 4: name the mistake

A mistake you can name is a mistake you can stop making. Most errors fall into a handful of types:

• Hanging a piece: leaving something undefended, or moving a defender away.
• Missing a tactic: a fork, pin, skewer or discovered attack, for you or against you.
• Ignoring the opponent's threat: playing your own plan without asking what their last move wanted.
• King safety: delaying castling, or pushing the pawns in front of your king.
• Time trouble: a good position thrown away in the last minute.
• A wrong plan: no single bad move, but a slow drift into a worse position.

Write the type next to the game. After ten games you will see a pattern, and the pattern tells you what to train.

Step 5: compare with your notes

Now look back at what you wrote in step 1. Did the game turn where you felt it did? Often it turned earlier, on a move that felt harmless. Were your worries real? Players often fear threats that are not there and miss the ones that are. This comparison is the most valuable part of the routine: it trains your judgement, not just your memory of one game.

Step 6: turn it into training

• Missed tactics? Train the theme with the tactics puzzles, sorted by pattern.
• Lost in the opening? Read about common opening traps and try the opening puzzles for your opening.
• Could not convert a winning position? The endgame book trains technique with few pieces.
• Collapsed under pressure? The defensive puzzles are positions where only one move holds.

Reading the labels without panic

A review full of red labels can feel harsh. Two things help. First, read the labels in context: an Inaccuracy in a position that was already lost changes nothing, and a Blunder that the opponent did not punish still cost you only potential. Second, look at the good labels too. A Great move means you found the only move that held; a Brilliant move means a sacrifice worked. The move classifications guide explains each label, and how accuracy is calculated explains the number at the top.

A routine that fits in a busy week

• After every serious game: steps 1 to 4, about ten minutes.
• Once a week: look at the mistake types you noted, and pick one theme to train.
• After a bad run: review the losses together. Patterns are easier to see across several games than in one.

Your wins deserve a look too. A won game often contains a moment where the opponent could have turned it round, and you will want to recognise that moment before the next opponent finds it.

Analyse your last game