Guide · 2026-11-03 · 7 min read
How to analyse your own chess games: a simple method that works for kids and adults
Reviewing your games is the fastest way to improve, but only if you do it in the right order: your own thoughts first, then the turning point, then the engine, then one lesson. A step-by-step method with a worked example.
Most players finish a game, glance at the result and click "new game". The players who improve quickly do one more thing: they spend ten minutes finding out why the game went the way it did. Strong coaches agree on this more than on anything else. The method below is the one we built Chess Castles’ review tools around, and it works with nothing more than a board and a notebook.
Why review at all?
Playing more games mostly repeats what you already do, mistakes included. A review turns one game into a lesson: you find the moment the game was decided, understand why, and fix that one thing. One game reviewed properly teaches more than ten games played in a row.
The method, step by step
- Right after the game, write three lines: where did I think I was better or worse, which move was I unsure about, and what did my opponent do that surprised me? Do this before any engine, while the thoughts are fresh.
- Play the game through on a board, without an engine. At each move where you spent a long time, or where the position changed (a capture, a check, a pawn break), stop and write what you were thinking.
- Find the turning point: the single move after which the result was no longer in doubt. In beginner and junior games it is almost always one of three things: a piece left unprotected, a tactic missed (yours or your opponent’s), or a king left unsafe.
- Now check with the engine. Look first at the moments you marked yourself. Where the engine agrees, good; where it shows a big jump in the evaluation that you did not notice, that is the real lesson.
- Ask "why?" in words. Not "I should have played Nf5", but "I did not see that his bishop was attacking my rook" or "I moved my queen away from defending f7".
- Write one lesson in your notebook, one sentence. Then do ten puzzles on that theme.
The order matters. If you open the engine first, it tells you the answers before you have asked the questions, and you remember almost nothing.
A worked example
Here is a short game of the kind every coach sees: 1.e4 e5 2.Nf3 d6 3.d4 Nd7 4.Bc4 Ngf6 5.dxe5 Nxe5 6.Nxe5 dxe5.
Step 2, without an engine: Black recaptured twice on e5 and the position looks equal in material. But look at what the recaptures changed. The d-pawn has left d6, so the queens now face each other on the open d-file. And the pawn on f7 is protected only by the king.
White plays 7.Bxf7+! Black has two replies. If 7...Kxf7, then 8.Qxd8 wins the black queen, because the king has been pulled away from the d-file and nothing else covers d8. If 7...Ke7, then 8.Bb3: White has won a pawn and the black king can no longer castle.
Step 3: the turning point is the moment Black allowed this. Step 5, the "why?" in words: "When I recaptured, I did not ask what the recapture opened. The d-file opened between the queens, and f7 was only guarded by my king." Step 6, the lesson: "Before every capture, ask what lines it opens." Then ten puzzles on discovered attacks and attacking f7.
What to look for in every game
- Hanging pieces: did either side leave a piece unprotected? Check every capture you or your opponent missed.
- Your opponent’s threats: on the move where you lost material, what did their previous move threaten? (Our 10-second blunder check is the cure.)
- The opening: at which move did you leave the moves you know? Compare with the games database to see what strong players do there.
- The clock: did you spend twenty minutes on move 8 and then blunder in a hurry on move 30?
- The endgame: if you were winning and drew, was it stalemate, a fortress, or a rook pawn? See king and pawn endgames.
How Chess Castles helps
- Game Report checks every move with Stockfish and marks the turning point in words ("25% of the winning chances gone in one move"), with accuracy and move badges. Use it for step 4, after your own review.
- Chess Guru can tell the story of a game in English, Hindi or Marathi, and explain the turning point in plain language.
- Game Insights reads your games from Lichess or chess.com and finds the patterns across many games: the openings where you lose most, the kind of mistake you make most often, with a weekly plan.
- Puzzles by theme turn each lesson into practice.
For parents: reviewing with a child
- Let the child talk first. Ask "where did you feel the game change?" before showing anything.
- Praise the questions, not just the moves: "good that you noticed you were unsure here".
- Keep it short: ten to fifteen minutes is enough for a young child. One lesson per game.
- Review wins too. Children often win games where the opponent missed a winning move; finding it teaches as much as a loss. (See also when your child loses and cries.)
Questions players ask
Should I analyse every game?
Every slow game, yes (rapid and classical). Blitz games are too fast to teach much; review one blitz game a week at most, or only the ones you lost in the same way twice.
How long should a review take?
Ten to twenty minutes for a club game. The goal is one clear lesson, not a perfect analysis of every move.
Is it cheating to use an engine to analyse?
No. Using an engine after the game to learn is exactly what strong players do. Using one during a game is cheating, on any site and over the board.
अपने chess games का विश्लेषण कैसे करें: बच्चों और बड़ों के लिए काम करने वाला आसान तरीका
अपने games का review सुधार का सबसे तेज़ तरीका है, पर तभी जब सही क्रम में किया जाए: पहले अपने विचार, फिर turning point, फिर engine, फिर एक सबक। एक हल किए उदाहरण के साथ step-by-step तरीका।
स्वतःच्या chess games चं विश्लेषण कसं करायचं: मुलं आणि मोठ्यांसाठी चालणारी सोपी पद्धत
स्वतःच्या games चा review हा सुधारण्याचा सर्वात जलद मार्ग, पण तेव्हाच जेव्हा तो योग्य क्रमाने केला जातो: आधी स्वतःचे विचार, मग turning point, मग engine, मग एक धडा. सोडवलेल्या उदाहरणासह step-by-step पद्धत.