Elo Rating Calculator

See your expected score against any opponent and exactly how many rating points a win, draw, or loss is worth.

How it works

Every rated player has an Elo number, and the gap between two ratings tells you how often the higher-rated player should win. The calculator above turns that gap into a percentage called your expected score, then works out how many points change hands once the game is actually played.

Expected score comes from one formula: divide 1 by (1 plus 10 to the power of the rating difference divided by 400). Plug in a 1500 facing a 1700, and the gap of 200 points works out to an expected score of about 0.24. In plain terms, the 1500 is expected to score roughly a quarter of a point on average against that opponent, well below the 0.5 a game against an equal would produce.

Once the game ends, your rating change is K times the difference between what actually happened (1 for a win, 0.5 for a draw, 0 for a loss) and what was expected. Take that same 1500 beating the 1700 with K = 32: the change is 32 times (1 minus 0.24), which rounds to 24 points, so the win takes them to 1524. Lose that same game instead and the change is 32 times (0 minus 0.24), which rounds to −8, dropping them to 1492. The upset win is worth far more than the expected loss costs, because beating a stronger player carries more information about your real strength.

K-factor sets how aggressively the system reacts to a single result. A higher K (32 or 40) swings your rating faster and is standard for newer or provisional players. A lower K (10 or 20) is used once a rating has settled, so one fluke result doesn't move the number much.

FAQ

Why do provisional ratings move faster than established ones?

A brand-new rating is a guess. The system hasn't seen enough of your games to know where you really belong, so it uses a high K-factor and lets each result swing the number hard until it converges on your true level. Once you've played enough rated games, most platforms and federations drop you to a lower K so a single bad night or a lucky streak can't distort a number that's already reasonably accurate.

Why does beating a much lower-rated player earn so few points?

Because you were already expected to win. If your expected score against that opponent is 0.95, a win only closes the remaining 0.05 gap, and even with K = 32 that's roughly 2 points. The formula only pays out for results that update the estimate of your strength, and beating someone far below you tells the system almost nothing new. Losing that same game, on the other hand, costs you far more, since it's the surprising result.

Is a FIDE rating the same as a Chess.com or Lichess rating?

Not directly. All three use the same Elo math in spirit, but each calibrates against its own pool of players, starts new accounts at different baselines, and may use different K-factors or rating floors. A 1500 on Chess.com, a 1500 on Lichess, and a FIDE 1500 are not interchangeable numbers, even though the underlying logic that moves them is identical.

Does a draw ever change my rating?

Yes, unless you're playing an equally rated opponent. A draw counts as a score of 0.5, so if your expected score wasn't exactly 0.5, the gap still gets paid out. Draw against a stronger opponent and you gain a few points; draw against a weaker one and you lose a few, because a draw was a worse-than-expected result for the higher-rated side.

For the full background on where this system came from and what a "good" rating actually looks like on different platforms, see How Chess Ratings Work: Elo Explained for Beginners. If you're trying to move your own number up, pair the calculator with our beginner improvement plan and a routine for analyzing your own games so the points you're calculating here actually start trending upward.