Chess Accuracy Hub: Every Rating Benchmark Explained

Chess.com accuracy is the metric every player checks after a game -- yet most misread it. A 72% score looks discouraging until you realize it is normal at 1400 in rapid. A 96% game looks suspicious until you understand a GM in a sharp tactical line routinely produces it. This is your single reference for every accuracy band, every rating tier, and every variable that moves the number.

What Chess Accuracy Measures

Chess.com calculates accuracy by comparing each move against the engine's preferred candidate, then converts the average evaluation loss into a 0-100 percentage using a nonlinear formula that weights blunders more heavily than minor inaccuracies. A single 0.5-pawn oversight in a sharp position can pull a game from 88% down to 69%.

The core principle: accuracy measures engine agreement, not chess quality in context. A correct strategic sacrifice the engine rates 0.3 centipawns below its first choice still costs accuracy. This deflates scores in long positional games and inflates them in short forcing sequences.

Lichess uses Average Centipawn Loss (ACPL) instead, where lower is better. A 25 ACPL game on Lichess corresponds roughly to 88-92% on Chess.com, but the platforms use different Stockfish depths and weighting formulas. Track trends within each platform separately.

Accuracy Benchmarks by Rating Level

Observed ranges across rapid and classical on Chess.com. Blitz runs 5-10 points lower at every tier.

These are per-game medians, not ceilings. Short tactical finishes inflate scores; long maneuvering endgames deflate them. Evaluate your 20-game rolling average, not individual results.

What Each Band Signals

Under 70%: Real Oversights

Games in this range contain genuine errors -- undefended pieces, missed one-movers, material-losing tactics. Normal below 1300 in rapid and below 1600 in blitz. The priority is not raising the number but finding the specific blunder patterns producing large evaluation swings. See Is 70% Accuracy Good in Chess? for the full breakdown.

70-80%: Competent Club Play

Principled chess with tactical basics in place, interrupted by positional miscalculations. Typical for 1300-1600 in rapid. Blitz for 1700-2000 players also lands here -- time-control deflation is not the same as skill regression.

80-90%: Intermediate Strength

Errors are inaccuracies rather than blunders -- inferior plans, endgame lapses, evaluation misreads under time pressure. Players averaging 83-88% in rapid are operating in the 1600-2000 range. If your accuracy systematically drops in moves 15-30, you have a middlegame calculation problem, not a general deficit. See Why Your Chess Accuracy Drops After Move 15.

90-95%: Expert Level

Sustained precision across all phases, typical for 1900-2300. Averaging 90%+ across 20+ standard-length rapid games is a legitimate expert-level indicator -- a single 91% game proves nothing since a short forcing finish can produce it.

95%+: Near-Engine Territory

Rare outside titled players in classical. A single 95% game is not suspicious and is achievable by any strong club player in a sharp tactical finish. Sustained high accuracy across many games combined with unusual timing patterns is what fair play review actually looks at. See What Chess Accuracy Is Suspicious? for the exact thresholds.

The Four Variables That Shift Accuracy

Time control dominates: Classical beats Rapid beats Blitz beats Bullet by 5-10 points per step. A 2100-rated player averaging 91% in rapid may average 82% in blitz.

Game length: Short forcing finishes inflate scores (few decision points). Long balanced endgames deflate them (40+ moves with marginal engine preferences).

Opening theory gives free accuracy points. Your first 12 theory moves approach 100% because you are pattern-matching rather than calculating. This can mask poor middlegame play behind a high headline number.

Position type: Tactical sharp lines have clear engine preferences. Maneuvering positions with many near-equivalent moves deflate accuracy even when your plan is correct.

Frequently Asked Questions

What is a good accuracy for a 1500-rated player?

In rapid, 76-84% is the healthy range. Consistently hitting 80%+ reflects sound tactical discipline. Track your 20-game rolling average rather than reacting to individual results.

Is 80% accuracy good in chess?

Depends entirely on time control and rating context. In rapid at 1300-1600, 80% is above average. In classical at 2100+, it would indicate significant errors. Compare within your rating and time-control tier.

Why does my accuracy fluctuate 15-20 points between games?

Normal variance, especially in blitz and rapid. Short games and deep opening theory inflate individual scores; long balanced endgames deflate them. A 20-game rolling average filters the noise.

What accuracy do grandmasters typically score?

In classical, 94-98% on Chess.com analysis. In rapid, 90-96%. In blitz, even super-GMs score 85-92%. The absolute number tracks time control more than relative skill -- what stays consistent at GM level is the near-absence of tactical blunders.