Sports Rankings

Elo Rating System — How It Ranks Competitors

The Elo rating system ranks competitors by exchanging points after every game — you gain more for beating a strong opponent than a weak one.

Reviewed 2026 Updated only when the core methodology changes

What the Elo Rating System Is

The Elo rating system is a method for calculating the relative skill levels of competitors in head-to-head games. It was developed by the Hungarian-American physics professor Arpad Elo for the United States Chess Federation, which adopted it in 1960; FIDE, the international chess federation, followed in 1970.

The system replaced earlier chess rating methods with a statistically grounded approach: every player carries a single number, and that number rises or falls after each rated game depending on the result and the strength of the opponent.

What the Elo Rating System Ranks

The Elo system ranks individual competitors in two-player, win-or-lose contests — originally chess players, but the model applies to any paired competition.

Each player holds a rating representing their estimated strength. Ratings are meaningful only relative to other players in the same pool: a 1500 rating in one community is not directly comparable to a 1500 in another.

Core Inputs Used by the Elo Rating System

The basic Elo calculation uses a small set of inputs:

  • Current ratings — of both competitors before the game
  • Game result — win, loss, or draw
  • K-factor — a coefficient controlling how quickly ratings change

The system does not consider:

  • margin of victory
  • game quality or style
  • anything outside rated results

How Elo Ratings Are Calculated (High-Level)

The Elo system works by comparing what happened to what was expected.

In simplified terms:

  1. Before a game, each player’s expected score is calculated from the rating difference. A player rated 400 points higher is expected to win roughly 9 times out of 10.
  2. After the game, the actual result (1 for a win, 0.5 for a draw, 0 for a loss) is compared with the expectation.
  3. Ratings adjust in proportion to the surprise: beating a stronger opponent gains many points; beating a much weaker one gains few.
  4. In its basic form, the system is zero-sum — the points one player gains are the points the other loses.

The expected score follows a logistic curve, so rating differences translate smoothly into win probabilities rather than hard cutoffs.

Conceptual model: Ratings chase results — they drift toward the number that would have predicted past games correctly.

Key Parameters or Factors

  • K-factor — the maximum number of points at stake in a single game. Higher values make ratings respond faster but more volatile. Chess federations typically use lower K-factors for established players and higher ones for newcomers.
  • Rating scale — Elo numbers have no fixed ceiling or universal meaning; they are anchored only by the pool of rated players.

Update Frequency

Elo ratings recalculate after every rated game. Published ranking lists, however, appear on their own schedules — FIDE, for example, releases official rating lists monthly.

Known Limitations and Criticisms

The Elo system has several well-documented limitations:

  • No measure of uncertainty — the system treats a rating based on 10 games the same as one based on 10,000, which the Glicko system was designed to address
  • Rating inflation and deflation — the pool’s average rating can drift over time as players enter and leave
  • Two-player assumption — multiplayer or team competitions require adapted variants
  • Relative, not absolute — ratings cannot be compared across different pools or eras with precision

Where the Elo Rating System Is Used

Elo-style ratings are used in:

  • Chess and other board games (FIDE, national federations, Go associations)
  • Online gaming and esports matchmaking
  • Sports rankings, including the FIFA World Rankings and independent football ratings
  • Academic modeling of paired comparisons

Few ranking ideas have spread as widely beyond their original domain.

Summary

The Elo rating system introduced a simple, durable idea: skill can be tracked by rewarding results that defy expectations. Its elegance — one number per player, updated after every game — made it the default model for competitive ranking, and most modern rating systems are refinements of its core logic.

References and Sources

  • Elo, A. E. The Rating of Chessplayers, Past and Present. (1978)
  • Wikipedia. Elo rating system.
  • FIDE. FIDE Rating Regulations (official handbook).
  • Glickman, M. E. The Glicko System (describing limitations Elo does not address).