ELO Calculator

What the Elo calculator does

The Elo system, born in chess and now used everywhere from video games to football rankings, turns each result into a rating change for both players. This tool does that calculation. You enter your rating, your opponent's rating, and the result, and it shows the expected score, the rating change, and the new ratings for both sides.

The idea at its core is simple and rather elegant: how much your rating moves depends on who you played, not just whether you won.

How to use it

1. Enter your initial rating and your opponent's initial rating.
2. Choose the result, a win, a draw, or a loss for you.

Press Calculate to see both players' rating changes and new ratings, or Reset to clear the fields.

How Elo works: the expected score

Before a game is played, the rating gap between two players predicts how it should go, as an expected score between 0 and 1. The formula turns the difference in ratings into a win probability:

Expected score = 1 ÷ (10^((opponent rating − your rating) ÷ 400) + 1)

If both players have the same rating, the expected score is 0.5, an even contest. The 400 in the formula sets the scale: a gap of 400 points makes the higher-rated player roughly a 10-to-1 favorite. So the bigger the gap, the more lopsided the expectation.

How the rating changes

After the game, your rating moves by the gap between how you actually did and how you were expected to do:

Rating change = K × (actual score − expected score)

A win counts as an actual score of 1, a draw as 0.5, and a loss as 0. The K factor, set to 20 here, controls how big the swing is. Do better than expected and you gain points, do worse and you lose them. This is why beating a much stronger player is worth a lot of points, since you were expected to lose, while beating someone far weaker barely moves your rating, since you were supposed to win anyway.

An example with real numbers

Say you are rated 1500 and you beat an opponent rated 1700.

• Expected score = 1 ÷ (10^((1700 − 1500) ÷ 400) + 1) = 1 ÷ (10^0.5 + 1) = about 0.24
• Rating change = 20 × (1 − 0.24) = 20 × 0.76 = about +15
• Your new rating = 1515, and your opponent drops to 1685

So beating someone rated 200 points above you earns about 15 points. Had you been the favorite and won, you would have gained far fewer.

Why the system is self-correcting

Two things make Elo settle toward true skill on its own. First, it is zero-sum: whatever you gain, your opponent loses, so points only ever move between players, never appearing from nowhere. Second, the expected-score mechanism quietly does the bookkeeping. A genuinely strong player who keeps winning against weaker opponents gains almost nothing each time, because those wins are expected, so their rating stops climbing once it reflects their level. Meanwhile an upset, where the result defies the ratings, moves both players a lot, nudging the numbers back toward reality. The K factor is the dial for how fast this happens: a larger K reacts quickly but makes ratings jumpy, while a smaller K is steadier but slower to catch a rising or fading player.

Questions people ask

What does the expected score mean?

It is how the game should go based on the rating gap, between 0 and 1. Equal ratings give 0.5, an even match. A 400-point gap makes the stronger player about a 10-to-1 favorite.

Why do I gain more for beating a stronger player?

Because you were expected to lose, so your actual result is far above your expected score, and the rating change is proportional to that gap. Beating a weaker player, which was expected, barely moves your rating.

What is the K factor?

It controls how much ratings move per game. A bigger K reacts faster but makes ratings jump around, while a smaller K is more stable but slower to adjust. This tool uses a K of 20.

Where do the rating points come from?

From the other player. Elo is zero-sum, so the points you gain are exactly the points your opponent loses. The total stays the same.

References

1. Arpad E. Elo, The Rating of Chessplayers, Past and Present (Arco, 1978).