Elo Rating Calculator

Calculate win probability and new ratings after a match using the Elo rating system.

FIDE: 40 (new), 20 (established), 10 (2400+). Online: 32 typical.
Player A Win %
64.01%
Player B Win %
35.99%
If Player A Wins
A: 1512  |  B: 1388
If Player B Wins
A: 1480  |  B: 1420
If Draw
A: 1496  |  B: 1404
Rating for 90% Win Probability vs Opponent
A needs 1782 vs B  |  B needs 1882 vs A

How to Use the Elo Rating Calculator

  1. Enter Player A's rating — their current Elo rating (e.g., 1500 for an average chess.com player).
  2. Enter Player B's rating — their current Elo rating.
  3. Set the K-factor — this controls how much ratings change per game. Higher K-factors create more volatile ratings.
  4. Read the results — see win probabilities, new ratings for each possible outcome, and the rating needed for a 90% win chance.

Understanding the Elo Rating System

The Elo rating system, devised by Hungarian-American physicist Arpad Elo in 1960, is the most widely used skill measurement system in competitive games. Originally designed for chess, it has been adopted across esports, sports, and even non-gaming contexts like evaluating AI model quality. The system works on a simple principle: after each match, rating points flow from the loser to the winner, with the amount depending on the expected outcome.

The Expected Score Formula

The probability of Player A winning is calculated using a logistic function:

EA = 1 / (1 + 10(RB - RA) / 400)

The constant 400 is chosen so that a 200-point advantage yields approximately a 76% expected win rate. This logistic curve ensures that even massive rating differences never produce a 100% or 0% expected outcome — there is always a chance for an upset.

Rating Update After a Match

After a game, ratings are updated using: R'A = RA + K(S - EA), where S is the actual score (1 for win, 0 for loss, 0.5 for draw) and K is the K-factor. If a player wins an expected game, they gain fewer points. If they pull off an upset, they gain significantly more.

What the K-Factor Controls

The K-factor is the maximum number of rating points that can change after a single game. FIDE chess uses K=40 for new players (first 30 games), K=20 for most players, and K=10 for those rated above 2400. Online platforms typically use K=20 to K=32. A higher K-factor makes ratings more responsive to recent results but also more volatile.

Elo Beyond Chess

The Elo system's mathematical elegance has led to adoption far beyond chess. League of Legends, Dota 2, and Overwatch use Elo-derived matchmaking systems. FIFA publishes World Football Elo Ratings. Researchers use Elo to rank large language models (the LMMSYS Chatbot Arena). The same logistic curve that predicts chess outcomes works anywhere two competitors face off.

Frequently Asked Questions

The Elo system assigns each player a numerical rating that increases when they win and decreases when they lose. The amount of change depends on the expected outcome: beating a much stronger opponent gains more points than beating a weaker one. A 400-point advantage translates to roughly a 91% expected win rate.
The K-factor determines the maximum number of rating points that can change after a single game. FIDE chess uses K=40 for new players, K=20 for established players, and K=10 for those rated above 2400. Higher K-factors make ratings more volatile and responsive to recent results.
A 100-point Elo advantage gives the higher-rated player approximately a 64% expected win rate. A 200-point gap means about 76%, and a 400-point difference translates to roughly 91%. The relationship follows a logistic (S-shaped) curve, not a linear one.
Yes. The Elo system is used in esports (League of Legends, Dota 2, Overwatch), competitive gaming platforms, sports (FIFA World Football Elo Ratings), academic research, and even dating apps. The mathematical framework applies to any competitive scenario with binary win/loss outcomes.
Mathematically, the Elo system allows ratings below zero, but most implementations set a floor (e.g., 100 in FIDE chess). In practice, a player rated near 0 would need to lose consistently against very low-rated opponents.