How long should a beginner spend analyzing a chess game?

For a beginner, five to ten minutes per game is plenty. At that stage almost every game is decided by hanging material or a missed elementary tactic, so digging deeper is wasted effort. Focus on a single question — "on which move did I lose material?" — and write the answer in one sentence. Once that question stops producing an obvious answer in your games, that's the signal to stretch the budget to fifteen minutes.

Do I need to analyze every blitz game I play?

No — and trying to is the fastest way to abandon reviewing altogether. Blitz mostly produces speed and attention errors, which you don't fix by rereading them one by one. A sound practice is a five-minute triage on one or two games per day, ignoring the rest without guilt. Save serious review for time controls where you actually had time to think: those contain your genuine errors of understanding.

Is it better to analyze with or without an engine?

Both, but in that order: without first, with afterwards. Replaying the game engine-free forces you to form a personal judgment, and that judgment is what the engine then corrects or confirms. If you open computer analysis immediately, you're reading answers to questions you never asked yourself, and nothing sticks. A practical rule: at least three minutes of personal search before every machine check.

Should I review my wins or only my losses?

Reviewing only losses throws away half the available information. A win obtained through an opponent's blunder from an inferior position contains exactly the same errors as a loss — they simply went unpunished. Those errors will be punished the moment you face a stronger opponent, and you'll never have identified them. A simple rule: review any game whose result you can't clearly explain, won or lost.

How many games per week should I analyze to improve?

The ratio matters more than the absolute number: aim for roughly one third of your playing time spent on review. If you play three hours a week, one hour of analysis is a realistic and sufficient target. In practice that usually means five quick triages, one standard review, and one deep dive. The decisive factor isn't volume but consistency — a modest routine sustained for six months beats one intense month followed by burnout.

← Back to blog

How Long Should You Spend Analyzing a Chess Game? (And What to Look at First)

August 22, 2026 · ChessPivot · Guide

A 1000-rated player who plays four blitz games a day accumulates roughly 120 games per month. Spending thirty minutes on each one would mean sixty hours of monthly analysis — more time than they spend actually playing. That arithmetic explains why most club players end up analyzing nothing at all.

So the question of analysis time isn’t academic. It decides whether you keep a review habit for six months or drop it after three weeks. And the answer depends far less on total duration than on the order in which you examine things.

This guide gives numbers: how many minutes for which type of game, what to look at first, and how to know when to stop. The goal is turning a vague chore into a repeatable ten-minute routine.

The three-budget rule: 5, 15 or 40 minutes

Not every game deserves the same investment. A bullet game flagged in a winning position teaches nothing; a classical loss in a rook ending can feed two weeks of study.

Here is a workable split for players between 800 and 1400 ELO:

The 5-minute budget — quick triage. For blitz and bullet. You are looking for exactly one thing: the moment the game flipped. Note the position, move on.
The 15-minute budget — standard review. For rapid games (10+0, 15+10) and blitz games that stuck in your mind. You identify three critical moments and try to find the best move yourself before checking.
The 40-minute budget — deep analysis. For classical games, tournament games, and one deliberately chosen online game per week. You replay the opening, walk through the middlegame plan by plan, and study the ending if there was one.

The classic trap is trying to apply the 40-minute budget to every game. It collapses within a fortnight. A realistic weekly split across ten games: eight quick triages, one standard review, one deep analysis. Total: roughly ninety minutes a week.

The underlying principle is blunt: ten shallow reviews plus one deep one beats zero perfect reviews.

Step by step: the 15-minute routine

Standard review is the highest-yield format. Here is a timed walkthrough that works with nothing more than a board and the game’s PGN.

Minutes 0–3: replay the game with no engine. Click through at normal speed. Say out loud, or write down, every moment where you remember hesitating, panicking, or moving too fast. Those moments are your real candidates — not the ones the engine will flag.
Minutes 3–6: find the tipping point. There is almost always one move after which the evaluation shifts for good. Locate it. It isn’t necessarily the biggest centipawn drop, but the one after which the game changed character.
Minutes 6–11: search before you check. Sit on the critical position and give yourself three to five minutes to find something better. This is the step everyone skips, and it is the only one that produces genuine learning.
Minutes 11–14: verify with the engine. Compare your candidate to the machine’s. If they differ, try to work out why before reading the suggested line.
Minute 15: write one sentence. Just one. "I traded my active piece for a passive one." "I missed a check that won a piece." That sentence is the deliverable; everything before it was scaffolding.

That last step deserves a defense. Without a written trace, an analysis evaporates within forty-eight hours. Twenty one-line notes accumulated over a month reveal patterns that ten hours of scattered analysis never will.

What to look at first (the order matters)

Since time is finite, the order of examination determines the return. Here is the hierarchy that produces the most rating points per minute invested, from highest yield to lowest.

Priority 1: pieces lost in a single move. Between 800 and 1400 ELO, the large majority of games are decided by material left hanging or a simple tactic missed. Count material at the end of each phase: if you went from level to two pawns down, find the move responsible.

Priority 2: tactics you missed or allowed. A queen fork you never saw, a pin your opponent exploited, a skewer along the long diagonal. These motifs repeat from game to game with almost comic regularity.

Here is what a position looks like when the side to move has an immediate motif available:

Black to move. Fork: Qa5+ — the queen on a5 attacks the king (e1) and a bishop (a6) at once.
Real line: Qa5+ Nbd2 Qxa6 c4 — Black wins a piece.
Takeaway: a loose piece or an exposed king invites a fork — keep your pieces defended and your king safe.
This kind of position isn’t about deep calculation but about visual recognition. The question to ask in review isn’t "should I have calculated further?" but "did I check every square from which my queen attacks two things at once?"

Priority 3: how you exited the opening. Not theory — the position at move 12 to 15. King safe? Pieces developed? Rooks connected? If the answer is no three games running with the same opening, the problem lies in your understanding of the structure, not in your memorization.

Priority 4: middlegame plans. Where were the weak pawns? Which file was open? This layer demands more maturity and naturally belongs to the 40-minute budget.

Priority 5: the endgame. Important, but it only appears in a minority of games below 1400. Address it when it happens rather than building an obsession around it.

The 5-minute triage: how it works

For blitz games, a stripped-down method is enough. It rests on three questions asked in order, each answered in under two minutes.

Question 1: on which move did I lose material? Scroll the evaluation graph or the material counter. The drop is impossible to miss.
Question 2: was it avoidable with three seconds of thought? If yes, it’s an attention problem, not a knowledge problem. Note it and move on.
Question 3: was a tactical pattern involved? If yes, add that tactical motif to tomorrow’s puzzle set.

Five minutes, three questions, one notebook line. The format is deliberately poor: its main virtue is that you can do it every single day.

Black to move. Fork: Qc2 — the queen on c2 attacks a rook (d1) and a bishop (b2) at once.
Real line: Qc2 Re1 Qxb2 Qc6 — Black wins a piece.
Takeaway: a loose piece or an exposed king invites a fork — keep your pieces defended and your king safe.
A position like this one shows why triage works. The configuration is simple, the motif becomes visible within seconds once you know to look for it, and the opposing mistake comes down to a piece placement that leaves two targets lined up on the same lines.

Five minutes of daily review beats one weekly hour, because repetition builds pattern recognition and occasional depth does not.

When does a game deserve the 40-minute treatment?

The long budget shouldn’t be handed out randomly. Three situations genuinely justify it.

A hard-fought classical game. You thought at length, so your mistakes are errors of understanding rather than speed. That’s the richest learning material you can generate yourself.

A loss that surprised you. If you don’t understand how you lost, you’ve found a blind spot. Those games deserve real time, including a second pass a week later with fresh eyes.

A dubious win. Winning on an opponent’s blunder from an inferior position validates nothing. Many players review only their losses and therefore discard half the available information.

The deep routine follows the 15-minute one with three extra layers:

Replay the opening to move 15, comparing against the Lichess database (millions of games) to see exactly where you leave known paths.
Formulate an alternative plan at each turning point, even where the engine sees nothing special.
Replay the ending against the engine from the critical position, to confirm you can actually convert.

White to move. Fork: Qd5+ — the queen on d5 attacks the king (g8), a knight (c6) and a bishop (c5) at once.
Real line: Qd5+ Be6 Qxe6+ Qf7 — White wins a piece.
Takeaway: a loose piece or an exposed king invites a fork — keep your pieces defended and your king safe.
In a deep review, a position of this type gets worked in two passes: first find the move yourself, then rewind two or three moves to understand which opposing decision made the motif possible in the first place. That rewind is what separates deep analysis from a simple observation.

Common mistakes in game review

Five anti-patterns show up constantly among players who analyze without improving.

Mistake 1: scrolling engine lines passively. Clicking "best move" and nodding produces nothing. The brain retains what it searched for, not what it was shown. Impose three minutes of personal search before any verification, every time.

Mistake 2: reviewing twenty moves instead of three. A forty-move game rarely contains more than three decisive moments. Going through every move dilutes attention and burns motivation. Hunt for turning points, not completeness.

Mistake 3: reviewing only losses. Wins contain just as many errors — merely unpunished ones. A position won by luck is a time bomb against a stronger opponent.

Mistake 4: mistaking engine depth for understanding. A twenty-move line evaluated at +2.3 teaches you nothing if you can’t replay it from memory. Come down to a depth you can explain in words.

Mistake 5: writing nothing down. The costliest of all. Without a notebook you will re-analyze the same error twelve times without ever recognizing it as recurring. One sentence per game is enough.

A sixth habit deserves mention: reviewing immediately after a frustrating loss. Emotion skews judgment and turns analysis into self-punishment. Wait at least an hour, ideally until the next day.

How to read an analysis board without drowning in it

Modern analysis interfaces display a lot: move classifications, evaluation curves, accuracy percentages. Here’s how to rank them.

The evaluation curve exists to spot drops, nothing else. A vertical fall means a moment worth examining. Micro-fluctuations are meaningless at your level.
Move classification (inaccuracy, mistake, blunder) is useful for counting, not for understanding. Three blunders in one game points to a time or attention problem rather than a knowledge gap.
Accuracy percentage measures comfort in the position, not playing strength. It rises mechanically in simple positions and collapses in complex ones.
Suggested variations should only be read after you’ve searched yourself. Otherwise they replace your thinking instead of correcting it.

A typical analysis board looks like this: a list of color-coded moves, a jagged graph, a few numbers. The value doesn’t come from the volume displayed but from your ability to ignore 90 % of it and focus on the two or three positions that mattered.

One good habit is hiding the evaluation on the first pass. Replay the game "blind," write down your impressions, and only then reveal the numbers. That artificial constraint forces personal judgment and makes it comparable to the machine’s.

Building a weekly routine you can actually sustain

The real goal isn’t reviewing one game better — it’s holding a rhythm for six months. Here’s a template running at roughly 90 minutes.

Monday to Friday — 5 minutes daily. Quick triage of the day’s game. One notebook line.
Saturday — 15 minutes. Standard review of the week’s most interesting game.
Sunday — 40 minutes. Deep analysis of one long game, plus a reread of the week’s five notebook lines.
End of month — 20 minutes. Reread every line from the month. Count repetitions: if "missed an opponent’s fork" appears six times, next month’s training program has written itself.

That final step is the most neglected and the most powerful. Individual analysis fixes a move; the monthly reread fixes a habit.

Adapt the volumes to your actual schedule. Someone playing three games a week doesn’t need five triage sessions. The ratio worth remembering: roughly one third of your playing time devoted to review — never more, never zero.

Conclusion

The right answer to "how long should I analyze a game" isn’t a single number but a distribution: five minutes for most fast games, fifteen minutes once or twice a week, forty minutes for one carefully chosen game. What matters more than duration is the order of examination — lost material first, tactics second, opening exit third, plans and endings last.

Three habits make the difference over six months: searching before consulting the engine, writing one sentence per game, and rereading your notebook at month’s end. None of them requires special software; all of them require consistency.

If sorting the critical moments by hand eats too much of your budget, a tool such as ChessPivot can hand you back those first few minutes by flagging the turning points — while the personal search that follows remains something no engine can do for you.

Frequently asked questions

How long should a beginner spend analyzing a chess game?: For a beginner, five to ten minutes per game is plenty. At that stage almost every game is decided by hanging material or a missed elementary tactic, so digging deeper is wasted effort. Focus on a single question — "on which move did I lose material?" — and write the answer in one sentence. Once that question stops producing an obvious answer in your games, that's the signal to stretch the budget to fifteen minutes.
Do I need to analyze every blitz game I play?: No — and trying to is the fastest way to abandon reviewing altogether. Blitz mostly produces speed and attention errors, which you don't fix by rereading them one by one. A sound practice is a five-minute triage on one or two games per day, ignoring the rest without guilt. Save serious review for time controls where you actually had time to think: those contain your genuine errors of understanding.
Is it better to analyze with or without an engine?: Both, but in that order: without first, with afterwards. Replaying the game engine-free forces you to form a personal judgment, and that judgment is what the engine then corrects or confirms. If you open computer analysis immediately, you're reading answers to questions you never asked yourself, and nothing sticks. A practical rule: at least three minutes of personal search before every machine check.
Should I review my wins or only my losses?: Reviewing only losses throws away half the available information. A win obtained through an opponent's blunder from an inferior position contains exactly the same errors as a loss — they simply went unpunished. Those errors will be punished the moment you face a stronger opponent, and you'll never have identified them. A simple rule: review any game whose result you can't clearly explain, won or lost.
How many games per week should I analyze to improve?: The ratio matters more than the absolute number: aim for roughly one third of your playing time spent on review. If you play three hours a week, one hour of analysis is a realistic and sufficient target. In practice that usually means five quick triages, one standard review, and one deep dive. The decisive factor isn't volume but consistency — a modest routine sustained for six months beats one intense month followed by burnout.