Guide
Blunder vs Mistake vs Inaccuracy in Chess
Engine tags classify move quality by centipawn loss. Learn the difference between blunders, mistakes, and inaccuracies — and which to fix first.
Updated 2026-06-15 · Part of improvement series
How engines classify moves
After analysis, each move is compared to the engine's best line. The drop in evaluation (centipawn loss, or CPL) determines the label. Exact thresholds vary by platform; MAZChess uses standard bands capped at 300 cp for outliers.
- Best / good — minimal eval loss; often the top move or near it
- Inaccuracy — small eval drop; suboptimal but not losing
- Mistake — significant eval drop; often turns equal to worse
- Blunder — large eval drop; often loses material or the game outright
Which tag matters most for your rating
Inaccuracies add up at master level. Below 1800, blunders and mistakes dominate rating swings. Fixing one recurring blunder type (back rank, hanging queen, missed fork) can gain more Elo than polishing quiet moves.
Do not shame yourself over inaccuracies
Even grandmasters play inaccuracies. The training goal is trend: fewer blunders per game this month than last. MAZChess highlights mistakes in the review sidebar and lets you jump between them with prev/next navigation.
Frequently asked questions
- Is a blunder always losing?
- Not always immediately, but blunders typically cost a full pawn or more in evaluation — enough that winning requires your opponent to err back.
- Why does the same move get different tags on different sites?
- Thresholds and engine depth differ. Compare tags within one platform over time, not across sites move-by-move.