For a serious review, spend five to ten minutes on your own thinking before revealing evaluations and best moves. Then use the engine as a verifier, not an answer sheet. The useful lesson comes from the gap between what you believed and what the position required.
Why self-analysis should usually come first
The engine can show that a move failed, but it cannot automatically tell you why you chose it. If you reveal the answer immediately, you lose evidence about your candidate selection, calculation depth, evaluation, time use, and emotional response.
The four-pass analysis workflow
- Pass 1: reconstruct.Replay the game without assistance. Mark uncertainty, surprise, long thinks, quick automatic moves, and changes of plan.
- Pass 2: analyze.At each marked position, write two or three candidates, calculate the opponent's strongest reply, and evaluate the resulting position.
- Pass 3: verify.Turn on the engine. Check tactics first, then compare the resulting structures and plans—not only the first recommended move.
- Pass 4: translate.Name the cause of the gap and create one specific exercise or rule for future games.
If you have not selected the positions yet, use How to Find Critical Moments in Your Chess Games before starting this workflow.
A position is more useful when your thoughts are attached
Review example · Central tension
Record the decision before checking the answer
Moves: 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.e3 O-O 5.Bd3 d5 6.Nf3 c5 7.O-O Nc6
White has several serious ways to handle the centre. Before using an engine, write what you expected after maintaining the tension, exchanging on d5, or changing the queenside structure. Your note might be: “I kept the tension because I feared releasing Black's bishop, but I did not calculate the next capture.”
FEN: r1bq1rk1/pp3ppp/2n1pn2/2pp4/1bPP4/2NBPN2/PP3PPP/R1BQ1RK1 w - - 2 8
When the engine is turned on, do not ask only “What is best?” Ask which candidate was missing, which reply changed the evaluation, and whether your original positional judgment was correct. Sometimes your move is acceptable but your reason was wrong; that is still valuable training material.
Five questions to ask the engine
- What is the opponent's strongest reply to my move? This often reveals the first calculation failure.
- Which candidate did I fail to consider? Candidate selection is a trainable skill.
- Where does the recommended line become understandable? Stop following moves when the strategic reason is clear.
- Is the difference tactical or positional? A numerical change does not explain its cause.
- Would I find this move in a real game? Prefer a lesson you can recognize and apply under time pressure.
For help interpreting the numerical side, read Chess Accuracy, Evaluation and Move Grades Explained.
When it is reasonable to use the engine sooner
Self-analysis first is a strong default, not a ritual. Shorten it when:
- the game was bullet or very fast blitz and you remember little calculation;
- the game ended with an obvious one-move tactical oversight;
- you are a new player and cannot yet identify legal candidate moves in a complex position;
- you only have a few minutes and the alternative is skipping review entirely;
- you are checking a concrete opening line rather than reviewing your decision process.
How much time should each stage take?
| Game | Self-analysis | Engine verification |
|---|---|---|
| Bullet or blitz | 2–5 minutes; mark obvious decisions and time trouble. | 5 minutes; check decisive tactics only. |
| Rapid | 5–10 minutes; reconstruct three to five positions. | 10–15 minutes; compare candidates and plans. |
| Classical | 20 minutes or more; preserve calculations and clock use. | As needed; verify important lines and technical conclusions. |
The goal is not equal time. Spend enough time without assistance to capture your thinking, then enough time with assistance to correct it.
Common engine-analysis mistakes
- Reading the label and moving on
- “Mistake” is a classification, not an explanation. Open the position and identify the missed idea.
- Following a long principal variation
- Stop when you understand the tactical point or positional change. Memorizing ten moves rarely improves the original decision.
- Checking only your blunders
- Review uncertain choices, successful plans, and positions where you were right for the wrong reason.
- Collecting too many corrections
- Finish with one priority that can change your next game.
Frequently asked questions
Can an engine replace a coach?
An engine is excellent at tactical verification and objective comparison. A coach can interpret patterns, adapt explanations, and design training around your recurring decisions. They solve different parts of the problem.
Should beginners analyze without an engine?
Yes, but briefly. Beginners should record what they saw and then verify quickly so that missed tactics do not become false conclusions.
Should I hide the evaluation bar?
During the first two passes, yes. Knowing the evaluation can influence your candidate search and make the correct answer feel obvious after the fact.