Chessy
Chess Statistics & Data Studies
Original chess statistics computed from Chessy's corpus of over 3,500 analyzed openings — real Lichess win rates and Stockfish evaluations, snapshotted July 2026. Every study below links to a downloadable CSV and JSON dataset, free to reuse under CC BY 4.0.
Chess Opening Win Rates: The Best (and Worst) Openings by Real Game Data | Chessy
Across 2,716 chess openings analyzed with at least 50,000 Lichess games each, the King's Pawn Game: Wayward Queen Attack (2.Qh5) scores 81.8% for White, the King's Gambit Declined: Falkbeer Countergambit scores 84.0% for Black, and the King's Pawn Game remains the single most-played opening at over 1.69 billion recorded games.
Read the study →Gambit ROI: Which Chess Gambits Outperform Their Stockfish Evaluation | Chessy
Across 587 chess gambits with at least 50,000 Lichess games each, the Queen's Gambit Accepted: Normal Variation: Nc6 has the best practical-vs-engine gap — 69.0% for White against a Stockfish evaluation of only +0.87 — while the King's Gambit Declined: Falkbeer Countergambit: fxe5 is the worst, scoring just 14.5% for White despite Stockfish only rating the position -4.49.
Read the study →The Most Common Chess Opening Mistakes, by the Numbers | Chessy
Across 6,572 catalogued mistakes spanning 3,504 analyzed chess openings, 60.2% are classified as inaccuracies, 32.6% as mistakes, and 7.2% as outright blunders — and the Richter-Veresov Attack: h6 has the highest average severity of any opening at 755.7 centipawns lost per error.
Read the study →The Chessy Chess Openings Dataset: 3,504 Openings, Free to Download | Chessy
Chessy's chess openings dataset covers 3,504 named openings, gambits, traps and anti-openings, of which 2,716 carry at least 50,000 real Lichess games and 3,462 carry Lichess statistics of any volume — every one paired with a Stockfish depth-16 evaluation and, where relevant, a catalogue of the most common human mistakes.
Read the study →