Elo
The original chess rating. Predict the result from the ratings gap, then correct by how wrong the prediction was — beat someone above you and yours jumps.
R' = R + K × (actual - expected)
Best at
Pure duels with a stable roster: chess, darts legs, head-to-head anything. Cross-group play knits ratings together, so it travels to a global board.
Where it misleads
Two sides only — a free-for-all needs Multiplayer Elo. Early ratings swing wildly until a player has some history.
How it works
Elo is the most famous rating system in the world, created by Arpad Elo for chess in the 1960s. It estimates player skill using pairwise win/loss outcomes and predicts future match results.
How it works: Each player starts with a rating (default 1500). After a match, the winner gains points and the loser loses points. The amount exchanged depends on the rating difference: beating a higher-rated opponent earns more points, while losing to a lower-rated one costs more. The K-factor controls how responsive ratings are.
When to use it: The gold standard for 1v1 games with established player pools. Its predictions improve with more matches, and it naturally creates an intuitive skill ladder.
Watch out for: Only supports 1v1 or team-vs-team (not FFA). Doesn't model uncertainty like Glicko-2 or TrueSkill. Ratings can be slow to converge for new players without a provisional K-factor period.