FAQ · ChessStats

About ChessStats

ChessStats is an independent chess statistics project that ranks the world’s best players by the quality of their moves, not their rating. Here’s how it works — and answers to common questions.

What the metrics measure

Every metric is a 0–100 score relative to the field, color-coded from green (a strength) to red (a relative weakness). The headline score is built from the quality of what a player makes happen on the board — posing problems, holding difficult positions, converting advantages, pressing, and resolving sharp moments. The style signals just describe how a player plays, not how well — they never count toward the score.

EliteStrongSolidRelative weakness

Common questions

How is this different from a FIDE rating or Elo?

A rating (Elo) is built from game results — who won and lost. ChessStats is built from the quality of what a player makes happen on the board — the problems they pose, the positions they hold and convert — across different kinds of situations. The two usually point the same way, but they measure different things.

Which players are tracked?

The world’s top classical grandmasters — including Magnus Carlsen, Hikaru Nakamura, Gukesh Dommaraju, Nodirbek Abdusattorov, Javokhir Sindarov, Alireza Firouzja and Vincent Keymer — alongside dozens more elite players and the strongest junior prodigies. A player is included once they have crossed 2700 and have enough recent classical games to score every metric with confidence.

Why isn’t a certain player on the list?

If a player you’d expect is missing, it’s usually because they’re rated below 2700 in classical, play mostly rapid or blitz, or haven’t played enough recent classical games for a confident score.

What are the “style” stats?

Sharpness, Chaos, time-trouble frequency and decisive-game rate. They’re purely descriptive — how sharp or wild a player’s games tend to be, whether they run the clock low or tend to play for a result — and say nothing about quality, so they never affect the score or ranking.

Which games does it use?

Every classical game from roughly the last five years by a tracked player. Faster rapid and blitz games, and games from before that window, aren’t counted — so the scores reflect serious classical play. Each move is checked with the Stockfish engine, but the score is about practical play — the decisions a player actually faces over the board, not raw engine evaluation.

How often do the scores update?

Weekly — new games, ratings and scores all refresh every week.

A few caveats

No model captures everything. These scores track playing strength closely and capture most of what separates one elite player from another — but they remain an approximation. Read them as strong, consistent estimates of quality of play, not exact measurements — nothing here can be 100% precise, and it isn’t meant to be.

Contact

Questions, corrections or feedback? Email contact@chessstats.gg — we read everything.