♟️ Chess Elo Rating Calculator

Enter both players' current ratings, the game result, and the K-factor to calculate the expected score and the winning/losing player's new rating using the standard Elo formula.

Your Rating

Opponent Rating

Result

K-Factor

Expected score first, then the adjustment

Elo asks what the rating gap predicts, then pays you the difference between that prediction and what happened. Ra is your rating, Rb the opponent's, S the result as 1, 0.5 or 0, and K the multiplier:

E = 1 / (1 + 10^((Rb - Ra) / 400))
new rating = Ra + K × (S - E)

The defaults, 1613 against 1609 with K=32, give E = 0.506, so a win is worth 32 × (1 - 0.506) = +15.8 points and a loss costs 16.2. A draw between two players this close moves almost nothing: -0.2. Every 400 points of gap shifts the expectation by a factor of ten in the odds, which is where 400 comes from.

Choosing a K-factor

K=40 is FIDE's setting for a player's first 30 rated games and for under-18s rated below 2300 - ratings that still need to move fast.
K=20 is the established-player value below 2400, and K=10 applies once a player has hit 2400, where a rating is expected to be close to correct already.
K=32 is the old USCF-style figure most club and online systems use, and it is the default here.

The tool rates one side of one game. The opponent only loses exactly what you gain when both of you carry the same K-factor, which is often not true in a mixed field.

Frequently asked questions

How many points do I gain for beating a player 200 rated above me?

24.3 at K=32. Enter 1400 against 1600: the expected score is 24.0%, so a win pays 32 × 0.76 while a loss costs only 7.7. Draws are worth +8.3.

Why did I only gain 8 points for winning?

Because you were favoured. At 1600 against 1400 the expected score is 76%, so a win pays 32 × 0.24 = 7.7 points. Beating weaker opponents cannot move a rating quickly - the risk is the 24.3 you drop if you lose.

What does an expected score of 90% mean?

Not that you win nine games in ten. It is the average points per game the gap predicts, so nine wins and a loss in ten, or eight wins and two draws, both come to the same 90%. A 400-point gap gives 90.9%.