Chess strategy
How to Analyse Your Own Chess Games
Published Updated
A useful game review is not just an engine verdict. It compares what you thought during the game, what the position required, and what later analysis supports. Review in three passes: reconstruct your decisions without an engine, investigate the critical moments, then use Stockfish to test rather than replace your thinking.
On this page
Pass 1 — replay without an engine
Step through the game and record where you felt uncertain, surprised, confident, short of time, or unable to find a plan. Write the candidates you considered and the reason for the move you played. These notes preserve the human problem. If you switch the engine on first, its top line can erase the question you actually need to answer.
Find the critical moments
Do not assign equal review time to every move. Spend the deeper analysis on decisions where you were unsure, where the evaluation changed sharply, or where the position changed character. The remaining moves still deserve a quick scan for missed alternatives and recurring habits.
- The first move where you left known opening ideas and had to think for yourself.
- Every irreversible decision: pawn break, exchange, sacrifice, or king commitment.
- A large change in material, initiative or king safety.
- A moment where your plan failed or the opponent surprised you.
- The final practical chance before the result became unavoidable.
Pass 2 — analyse alternatives yourself
For each critical moment, identify the threat, generate candidate moves, and calculate the opponent’s best reply. Use the strategic checklist: king safety, material, activity and pawn structure. If you rejected a move during the game, test whether the reason was correct.
Classify the error before asking for a computer answer. Was it a missed tactic, a poor plan, an opening knowledge gap, time trouble, or an endgame technique problem? Different causes need different training.
Pass 3 — use the engine as a verifier
Now run the game review. Start at the critical moments, compare the engine’s candidates with yours, and follow enough of the line to understand the idea. An evaluation number alone is not a lesson. Translate it into chess language: “this loses because the back rank is weak,” “this exchange leaves a lost pawn ending,” or “the quiet move stops the only pawn break.”
Create one lesson, not twenty
End the review with one sentence and one action. Examples: “I opened the centre before castling — practise the king-safety checklist,” or “I missed a backward knight move — work through a set of defensive positions slowly.” Save the critical position, add it to a mistake set, or link it to the relevant lesson. A small correction revisited in future games is more usable than a long list of unrelated verdicts.
A useful review template
- Game plan
- What was I trying to achieve?
- Critical moment
- Where did the position or decision change?
- Root cause
- Tactic, plan, knowledge, time, or technique?
- Better rule
- What should I notice next time?
- Practice action
- One position, drill or lesson to revisit