It isn't the Elo.
It's personal.

Rating predicts how often you go wrong — and almost nothing about what kind of wrong you are.

Interactive lesson — The hanging piece: the question that saves games: Play it on the board below, and make the key move yourself. White to move. Run the cheapest check in chess before you touch anything, which black pieces have a defender, and which have none? Rxd5. Nothing guards the knight, so the rook simply takes it. No calculation, just counting. Now look for the recapture. The pawn on b6 covers a5 and c5, not d5, and the king is on the other side of the board. A whole piece, won by asking one question. Ask it every move, and ask it for your own pieces too.

Blunders, by rating

Blunders per 100 moves, rapid. From beginner to 1500+ the rate halves — error frequency is the thing Elo actually measures.

Every clock, same staircase

Blunders per 100 moves halve from beginner to expert — on every time control. A 2000 bullet player still blunders like a 1200 rapid player.

Same Elo, different DNA

Two players at the same rating share a #1 weakness less than 1 in 3 times. Hanging pieces rules until 1800+ — then the crown passes to missed skewers.

Weaknesses change shape

Endgame errors rise with rating — stronger players survive long enough to reach them. Opening drift never goes away.

You, on a clock

Same player, two time controls: median gap 167 rating points from yourself — one player spans 936.

Every player, one dot

All 68 rapid players. The trend line explains 58% of the differences — and not one 1500+ player blunders above the median sub-900 player.

4.8 → 4.5

Small mistakes per 100 moves. Almost the same at every level — everyone makes them.

−32%

Your worst move only shrinks a third from beginner to 1500+. It stays huge — strong players just recover from it.

+13%

Faster chess, more blunders — for the exact same player.

Method. 6,288 games from 103 Chess DNA players (130 player×clock samples: rapid 68, blitz 41, bullet 21), each analyzed by Stockfish 17 at depth 18. A blunder loses ≥20% win chance; errors are counted per 100 moves because higher-rated games run longer (61 → 75 moves). Gates: ≥15 analyzed games per player per time control, ≥8 players per rating band (smaller bands merged), opponents within ±250 Elo. Weakness profiles span each player's recent games across time controls. R² = share of player-to-player differences rating explains; per-theme miss rates score R² 0.00–0.09.

Keep going

If you're in the 900–1199 band — where hanging pieces still decide most games — the step-by-step 800-to-1200 improvement roadmap walks through exactly what to fix first. Already past that? See the 1200-to-1600 roadmap.

Chess DNA — see what you actually struggle with.

Get your Chess DNA →