Chess Rating Calculator

What is a chess rating calculator?

A chess rating calculator tells you two things about a single game: the expected score the Elo system predicts for you against a given opponent, and the rating change you earn after you win, draw, or lose. You enter your current rating, your opponent’s rating, and a K-factor, pick the result, and the tool returns your new rating. It uses the same Elo math behind FIDE, the USCF, Chess.com, and Lichess, so the numbers match what those systems do for one game.

Enter your numbers in the box above and choose your result. The math runs entirely in your browser, so nothing about your games is uploaded or stored.

How does the Elo rating formula work?

The calculator runs two steps. First it computes your expected score, then it applies that to your result to get the rating change.

expected = 1 / (1 + 10^((opponent − you)/400))

new = old + K × (result − expected)

Every symbol, with its meaning and unit:

• you — your current rating, a whole number of Elo points (for example 1500).
• opponent — your opponent’s current rating, in the same Elo points.
• expected (E) — the expected score: a number between 0 and 1 that is the points you should average against that opponent. 0.5 means an even match; closer to 1 means you are favoured; closer to 0 means you are the underdog.
• result (S) — the actual outcome, scored as 1 for a win, 0.5 for a draw, and 0 for a loss.
• K — the development coefficient (K-factor), which sets how much a single game can move your rating. It commonly runs from 10 to 40; many sites use 20 or 32.
• 400 — a fixed scaling constant built into Elo. A 400-point gap makes the stronger player about 10 times more likely to score, by design.
• new — your updated rating, rounded to a whole number of Elo points.

The logic is intuitive: result − expected is how much better (or worse) you did than predicted. Beat expectations and the term is positive, so you gain points; fall short and it is negative, so you lose points. Multiplying by K scales that surprise into actual rating points. Because the expected score already accounts for the gap, you gain little for beating a much weaker player and a lot for upsetting a much stronger one.

The expected score depends only on the rating gap (opponent − you), not on the absolute numbers. A 100-point underdog has the same 0.360 expected score whether the players are rated 1500 and 1600 or 2500 and 2600.

Examples

Every example uses the two formulas above exactly, so you can reproduce each number by typing the same inputs into the calculator. Expected scores are shown to three decimals and rating changes to one decimal before rounding the new rating.

Example 1 — you 1500 vs opponent 1600, K = 20

The gap is 1600 − 1500 = 100, so:

E = 1 / (1 + 10^(100/400)) = 1 / (1 + 10^0.25) = 0.360

Applying each result with K = 20:

• Win: 20 × (1 − 0.360) = +12.8 → new rating 1513
• Draw: 20 × (0.5 − 0.360) = +2.8 → new rating 1503
• Loss: 20 × (0 − 0.360) = −7.2 → new rating 1493

As the slight underdog you gain a healthy 12.8 for the win but shed only 7.2 for the loss, and even a draw nudges you up because you scored above expectation.

Example 2 — evenly matched, you 1800 vs opponent 1800, K = 32

With a zero gap the expected score is exactly even:

E = 1 / (1 + 10^(0/400)) = 1 / (1 + 1) = 0.500

With K = 32:

• Win: 32 × (1 − 0.500) = +16.0 → 1816
• Draw: 32 × (0.5 − 0.500) = 0.0 → 1800 (unchanged)
• Loss: 32 × (0 − 0.500) = −16.0 → 1784

When two equal players meet, a draw is the expected result and changes nothing, while a win and a loss are symmetric at ±16.

Example 3 — heavy favourite, you 2000 vs opponent 1700, K = 10

The gap is 1700 − 2000 = −300, so your expected score is high:

E = 1 / (1 + 10^(−300/400)) = 0.849

With a low master-level K = 10:

• Win: 10 × (1 − 0.849) = +1.5 → 2002
• Draw: 10 × (0.5 − 0.849) = −3.5 → 1997
• Loss: 10 × (0 − 0.849) = −8.5 → 1992

Because you were expected to win, beating a 300-point-weaker player earns just 1.5 points, but a draw actually costs you 3.5 and a loss is a painful 8.5 — the risk of playing down.

Example 4 — big underdog, you 1200 vs opponent 1400, K = 40

The gap is 1400 − 1200 = 200, so:

E = 1 / (1 + 10^(200/400)) = 0.240

With a beginner K = 40:

• Win: 40 × (1 − 0.240) = +30.4 → 1230
• Draw: 40 × (0.5 − 0.240) = +10.4 → 1210
• Loss: 40 × (0 − 0.240) = −9.6 → 1190

A large K and a low expected score make an upset very rewarding — over 30 points for the win — while the loss you were predicted to take costs less than 10.

Expected score by rating gap

This table works the expected-score formula across common rating gaps, where the gap is opponent − you. It shows the points you should average against that opponent; the rating change for any result is then K × (result − expected).

Opponent − you	Meaning	Expected score (E)	Win term (result 1)	Loss term (result 0)
−400	You far stronger	0.909	1 − 0.909 = 0.091	0 − 0.909 = −0.909
−200	You stronger	0.760	0.240	−0.760
−100	You slight favourite	0.640	0.360	−0.640
0	Even match	0.500	0.500	−0.500
+100	You slight underdog	0.360	0.640	−0.360
+200	You underdog	0.240	0.760	−0.240
+400	You far weaker	0.091	0.909	−0.091

Multiply any “term” by your K-factor to get the rating points. With K = 20 and a +400 gap, an upset win is worth 20 × 0.909 = 18.2 points, while the expected loss costs only 20 × 0.091 = 1.8.

Common uses

A chess rating calculator is useful whenever you want to know what a game is worth:

• Predicting a rating change before or after a tournament game, so you know the stakes against a specific opponent.
• Deciding whether to play up or down, by comparing the small gain against a weaker player with the larger swing against a stronger one.
• Understanding your club or online rating when a platform shows a change and you want to confirm the math.
• Estimating expected results for a match or a simul, since the expected score is the average points you should take per game.
• Teaching the Elo system, by letting students see how the rating gap and K-factor drive every point of change.

Tips and common mistakes

• The gap is opponent minus you. In the exponent it is (opponent − you)/400. If your opponent is higher rated, the exponent is positive and your expected score drops below 0.5.
• Score the result as 1, 0.5 or 0. A win is 1, a draw is 0.5, and a loss is 0 — not the pin or point totals from other games. Using the wrong value breaks the change.
• Match the K-factor to the player. A 40-point K for a beginner moves ratings fast; a 10-point K for a master barely moves them. Using a too-large K makes ratings jumpy and unreliable.
• Beating weaker players gains little. Against a much lower-rated opponent your expected score is near 1, so a win earns only a fraction of K — and a loss is expensive. The system rewards facing stronger opposition.
• Expected score is per game, not a percentage to win. An E of 0.640 does not mean a 64% win chance with no draws; it is the average points (wins plus half-draws) you should score.
• Rounding happens at the end. The change can be a decimal like +12.8, but the new rating is rounded to a whole number, so tiny changes may round to the same rating.

Limitations and notes

This tool computes the Elo expected score and the rating change for one game using the standard formula new = old + K × (result − expected). It does not reproduce a full official update on its own: real FIDE and USCF calculations sum the change across every game in an event, can apply a different K to each player, and add rules such as rating floors, provisional formulas for new players, and the 400-point cap on rating differences for unrated or mismatched fields. It also does not model Glicko or Glicko-2 (used by Lichess and Chess.com), which track a rating deviation alongside the rating, so those sites’ changes will differ from plain Elo. Pick the K-factor that matches your federation or site, treat the result as the accurate single-game Elo change it is, and remember that everything runs privately in your browser — your ratings are never uploaded or saved, so you can model as many games as you like.

For more scorekeeping and rating math, try the bowling score calculator or the golf handicap calculator, use the percentage calculator to turn an expected score into a percentage, and browse the full sports category.