How much time should I spend on post-game analysis?

For a club player who wants to progress, 15 to 30 minutes per game is a realistic and productive target. Analysis can be done in two phases: a short session right after the game (while impressions are fresh, 10-15 minutes), and a longer session later with the tools. You don't need to analyze every game: better to seriously analyze one game in three than to skim through all of them.

Which engine should I use: Stockfish, Leela Chess Zero, or something else?

Stockfish (free on Lichess and via download) is the standard for tactical analysis: it's extremely precise on concrete moves. Leela Chess Zero (Lc0) has a positional evaluation style closer to human play, which can be more instructive for long-term positional judgments. Chess.com and Lichess integrate engines directly into their analysis interfaces, which is often enough. The engine isn't the limiting factor: your way of using it is.

How do I identify my recurring mistakes?

By keeping an error journal over time. After each analysis, note the main mistake(s) of the game with a short description. After 10-15 games, re-read: patterns appear. 'I missed center counter-attacks 3 times this month.' 'I mishandled rook endgames in 2 games.' 'I always develop the knight before checking gambit counterplay.' Those patterns are your priority training program.

Should I also analyze games I won?

Yes, but with special attention. In a won game, look for moments where your opponent would have made your life harder if they had played differently: those reveal your vulnerabilities even in victories. Also look for moments where you played 'the best move by accident' without calculating the variation correctly: a reminder that winning doesn't validate the process.

How do I analyze when I'm a beginner and don't understand what the engine shows?

Start with engine-free analysis only. Replay the game move by move and identify: where you sensed things were going wrong, the moves you weren't sure about during the game, and the move where you think you lost the advantage. Discuss with a player slightly stronger than you: human analysis is more pedagogical than engine analysis when the two players are at comparable levels. The engine becomes useful when you have a hypothesis to verify ('was that really winning?'), not as an oracle to ask 'what was the right move?'

Analyzing Your Chess Games: Practical Method to Progress

You just lost a game. Your opponent played something unexpected at move 15, you panicked, and the position deteriorated quickly. You open a new game.

It's the most common decision, and the least productive.

Replaying a chess game without analyzing it is training to replay the same mistakes in slightly different positions. Post-game chess analysis is the most profitable task for chess improvement. It's also, by far, the most neglected.

Why Post-Game Analysis Is Irreplaceable

Chess progression relies on a simple loop: play, identify mistakes, understand why they occurred, work on identified gaps. Post-game analysis is the central step of the chess improvement process, without it, the loop is broken.

Playing many games without analysis develops fluidity and cognitive endurance, but not playing quality. You play fast and often, and you repeat the same error patterns in fast-forward.

Studies on club player progression (notably Gobet and Campitelli, 2007, on 104 players over several years) show that the factor most correlated with progression is not the number of games played, but the time devoted to deliberate practice, and post-game analysis is a central component.

Deliberate practice is defined precisely: it's not comfortable, it targets specific gaps, it includes performance feedback. Post-game analysis meets all three criteria. Replaying blitz games meets none.

The Method: Human First, Engine Second

The most common error in post-game analysis: opening Stockfish first.

If the engine analyzes before you, you lose the bulk of the exercise's value. You don't develop your capacity to evaluate positions yourself, you read evaluations. You don't reinforce your intuition, you become dependent on external confirmation.

Phase 1: Human analysis without engine (15-20 minutes)

Replay the game mentally or physically, move by move. At each move, ask three questions:

"Did I understand the position at that moment?" If not, this is a move to analyze in priority.

"Was there an alternative I considered but rejected?" If yes, why did you reject it? Was it correct?

"When did I sense the position changing?" This moment (the move or sequence where the advantage shifted) is the most instructive of the game.

Note your impressions. Identify the moves you want to verify. Formulate hypotheses about what happened.

Phase 2: Engine verification (10-15 minutes)

Enter your hypotheses into the engine, not the entire game. "At move 15, I played e5 rather than d5: was that an error?" The engine confirms or refutes your hypothesis. Look at the proposed variations, but seek to understand why they are better, not just to note them. The right use of the engine is to answer specific questions you have, not to let it tell you what you should have thought. If you see the engine's best move and immediately think "I would never have considered that," write that thought down, it marks an important gap in your chess thinking process.

Chess.com's game review is particularly good for this phase. The chess.com tool highlights the critical moment in the game where the evaluation changed most sharply, which helps you find the right moment to analyze in depth. Using chess.com analysis after your own human review gives you the best of both worlds: your own thought process documented first, then engine verification to confirm or correct it.

Identifying Your Error Family

After 10-15 analyses, a pattern begins to emerge. Chess errors group into recognizable families.

Punctual tactical errors: losing moves caused by incomplete calculation or a "blunder" of surveillance. "I forgot that his rook covered that square."

Strategic plan errors: playing individually reasonable moves without overall coherence. "I improved my pieces without a clearly defined plan, and my opponent took control of the center meanwhile."

Opening structure errors: leaving known theory in positions you don't master. "At move 12, I deviated from theory and found myself in an unknown position where my opponent was comfortable."

Time management errors: taking too much time on easy moves and lacking time on critical moves.

Endgame errors: losing technically winnable positions, or drawing lost positions through lack of technique.

Identifying your dominant family allows targeting your work.

The Critical Moment: The Most Instructive Move

In almost all games, there's a precise moment where the advantage changes sides, or at least where a significant lead is established. It's the critical moment of the game.

This moment is often identifiable without engine: it's the move after which you sensed "something is wrong," or your opponent's move that surprised you.

Understanding why the position changed there, what idea you missed, what defensive resource you didn't see, what offensive plan you didn't anticipate, is the center of gravity of the analysis.

A player who deeply understands 3 critical moments in 3 different games progresses more than a player who superficially skims 30 games with the engine.

Error Taxonomy at the Critical Moment

Calculation error: you saw the correct variation but didn't calculate far enough. Treatment: long variation calculation exercises.

Surveillance error: you didn't see the opponent's threat. Treatment: develop the routine of "verifying opponent threats before playing."

Evaluation error: you saw the right move but underestimated the resulting position. Treatment: work on static position evaluation.

Plan error: you played "reasonable" moves without a plan, and your opponent had one. Treatment: habit of formulating a concrete plan before each move.

Using Available Tools to Analyze Chess Games

The chess improvement ecosystem now offers outstanding free tools to help you analyze chess games and identify your recurring mistakes. Here is how to best use each one.

Chess.com game analysis

Chess.com is the largest chess platform in the world and provides detailed game analysis for all games played on the site. The Chess.com game analysis tool calculates accuracy percentages, flags blunders, mistakes, and inaccuracies, and shows the best move for each moment. The chess.com review process is fast and easy to follow: after each game, a single click gives you the full analysis.

Chess.com also provides opening statistics comparing your chess game moves to top players' choices. This helps you understand where your chess opening preparation diverges from the best practice. Over time, chess.com analysis helps you build a map of your recurring chess mistakes, game by game.

One tip to better use chess.com analysis: don't just look at the accuracy percentage. Focus on the moments where the chess evaluation dropped sharply, those are the critical chess mistakes that cost you the most.

Lichess game analysis

Lichess offers completely free chess analysis for all games on the platform. The Lichess analysis board is excellent and uses Stockfish, currently the strongest chess engine available. Learning how to read a Lichess chess game analysis is straightforward and the tool is very good at flagging the key chess mistakes in each game.

Lichess also has a "Learn from your mistakes" feature that turns your chess game mistakes into interactive puzzles. This is one of the best ways to learn from chess analysis: after reviewing a chess game, you immediately practice the correct moves in a game-like context. The connection between chess analysis and chess puzzles accelerates improvement significantly.

Chess coaches and human review

Beyond engines, a chess coach is the best tool for improving through game analysis. A chess coach can help you understand why you made certain mistakes in a way that an engine cannot explain. The best chess lessons come from a coach who reviews your chess games and identifies patterns that recur across multiple games, not just isolated chess mistakes.

If a chess coach is not available, analyzing chess games with a stronger player (a chess club friend, an online partner) also gives you the benefit of human perspective. The process of explaining your chess game decisions to another person helps you understand your own thinking better.

ChessBase and SCID

ChessBase or SCID (free) allow deeper chess analysis with manual annotation and database search. These tools are better suited for serious chess study where you want to compare your chess game to reference games from master players.

The best tool is the one you use regularly with discipline. A paper notebook with 10 minutes of honest chess analysis per game is worth more than ChessBase opened twice a year.

The Three-Game Rule

For players short on time: only analyze games meeting at least one of these criteria.

Lost game with advantageous position: you were clearly better, and you lost. What gave way?

Game where you didn't understand what was happening: you felt lost at some point, without knowing why.

Game with a missed beautiful move: you saw afterward that there was a combination or beautiful resource you didn't play.

These three criteria filter the most instructive chess games and make chess game analysis manageable even with little available time.

How Chess Analysis Connects to the Rest of Your Chess Training

Chess analysis does not exist in isolation. It connects directly to chess puzzles, chess opening study, and chess endgame training. Here is how the chess improvement process works best when all pieces fit together:

Analyze your chess games to find your recurring chess mistakes.
Train chess puzzles focused on the type of mistakes you found, tactical chess mistakes call for tactical puzzles; strategic chess mistakes call for positional study.
Study chess openings relevant to the positions where you made chess mistakes in the opening phase.
Apply the lessons in future chess games and check whether the same chess mistakes recur.

This chess improvement cycle is how serious chess players improve at every level. The key is that the chess analysis step must come first, it gives direction to all other chess study. Without chess game analysis, you are doing chess puzzles and chess opening study without knowing whether they address your actual weaknesses.

Chess improvement is a process, not a single fix. Each time you analyze a chess game seriously, you learn something that will help you in future games. Over time, the chess mistakes you used to make repeatedly become recognizable patterns, and patterns you recognize are mistakes you can avoid.

The best chess players in the world analyze their chess games throughout their careers. Not because they are perfect chess players, but because chess game analysis is the most direct feedback loop the game offers. Every chess game you play is a lesson. Whether you learn from it depends on what you do after the game ends.

Final Thoughts on Analyzing Your Chess Games

This article has covered the core method for analyzing your chess games and the main tools available. Let me close with a few thoughts on the mindset behind good chess analysis, because the right approach makes an enormous difference.

Remember that the goal is understanding, not justification. The most common failure mode in analyzing your chess games is looking for reasons why you were actually right. The engine is your friend here, it is objective and does not care about your feelings. Remember also that a chess player who genuinely wants to improve must be willing to look at their own games honestly, including the games they won.

View every game as data. A game you lost is not a simple failure, it is a lesson about where your chess thinking broke down. A game you won may hide important moments where your opponent let you off the hook. Based on this view, analyzing your chess games becomes something you look forward to rather than something you avoid.

Consider the whole arc, not just individual games. One game tells you little. A month of analyzing your chess games tells you a lot. It is only by looking at patterns across many games that you discover what you actually need to improve. Students who commit to this process, even for 15-20 minutes per game, almost always discover something surprising about their chess thinking, something that was hiding in plain sight.

A chess player who consistently analyzes games and acts on what they find will reach their potential far more efficiently than a player who simply plays more. It is not a complex or difficult process, but it requires a shift in mindset: from "I want to play" to "I want to understand." That simple shift is what separates chess players who keep improving from those who stay on a plateau.

Discovering your improvement path starts with a single game. Pick a recent chess game you found interesting or instructive, set aside 20 minutes, and work through it move by move before looking at any engine line. Write down your thoughts at each key moment. Then check your analysis against the engine. That process, repeated consistently over months, is the most direct path to chess improvement available to any chess player at any level.

Sources

Gobet, F., & Campitelli, G. (2007). The role of domain-specific practice, handedness, and starting age in chess. Developmental Psychology, 43(1), 159-172.
Ericsson, K. A. (2006). The Cambridge Handbook of Expertise and Expert Performance. Cambridge University Press.

Key Takeaways

Post-game analysis without engine first is essential: if Stockfish looks at your game first, you only learn to read evaluations, not to improve your chess thinking
Chess errors group into families (tactical, strategic, opening, time, endgame): identifying your dominant family is the right foundation for targeted work
The critical moment of the game (where advantage shifted) is often identifiable without engine: understanding why matters more than finding the optimal move, and it will help you improve faster
Analyzing lost games is more instructive than analyzing won games: in won games, both sides' errors offset. But also look at the right moves you made by instinct, those confirm the chess skills you have already built.
20-30 minutes of serious chess game analysis per game produces more improve-ment than an extra hour of unanalyzed play
Use chess.com or Lichess to verify your hypotheses, they are the right tools for quick, accessible chess analysis at any level

Analyzing Your Chess Games: The Practical Guide to Real Progress