RIVALS Get the app

Padel Elo Calculator — doubles ratings that show their work

Put four ratings, a K-factor and the games score in; get every player's new number out, with the win probability the maths expected and the reason it moved that far. Built for a group that plays each other every week — this is crew-scale arithmetic, not a federation ranking.

The match
Team A · Player 1
Team A · Player 2
Team B · Player 1
Team B · Player 2

Names are optional. Ratings are Elo points — if you are starting from scratch, give everyone 1000.

How hard one match hits. Higher K, faster ratings move.

Who won
Games won — Team A
Games won — Team B
Why this matters

Club players start tracking Elo for one reason: the argument on the drive home never ends. Everyone remembers the nights they won and forgets the ones they were carried through, so “who is actually best” gets settled by whoever tells the story loudest. A rating replaces the story with arithmetic — one scale, every result on it, no selective memory.

What it fixes is comparison. Win rate flatters whoever draws the easiest partners; a rating asks who you beat and by how much, and prices the result accordingly. Beat the strongest pair in your group and you climb hard. Lose to the weakest and you slide hard. That is the whole point: a number that costs something is a number people trust.

Questions
How is Elo calculated for padel doubles?
Each side gets a team rating: the average of its two players. From the gap between those two averages the calculator works out how likely each side was to win, compares that with what actually happened, and multiplies the difference by the K-factor and a margin multiplier taken from the games score. The result is one number of points, and both players on a side move by that same amount — a shared delta, because in doubles you win and lose together. Worked example: 1100 and 1000 against 950 and 1050 (teams of 1050 and 1000) with K=24, the first pair winning 6–3, moves 14 points — 1114 and 1014 against 936 and 1036.
What K-factor should I use?
K sets how hard one match hits. Use 32 for casual nights and new players, where ratings still need to find their level fast. Use 24 for an established crew that plays most weeks — big enough to reward a real result, small enough that one bad night is not a verdict. Use 16 once you have a long history together and the numbers are close to settled.
Why did the winners gain so few points?
Because they were expected to win. Elo only pays you for the part of the result nobody saw coming. Beat a pair you were 85% likely to beat and there is barely any surprise left to pay for, so the move is tiny. Beat a pair that was 85% likely to beat you and the same K-factor hands you several times as much. Expectation, not effort, sets the price.
Does the games margin matter?
Yes. A 6–0 counts for more than a 6–5, through a margin multiplier built from the difference in games. The multiplier is damped when the rating gap is wide, so a strong pair thrashing a weak one does not get paid twice for the same mismatch. It is the same idea FiveThirtyEight used for sports Elo ratings: reward a convincing win without letting a scoreline run away with the rating.
How many matches until ratings are accurate?
Expect roughly 10 to 15 sets each before a rating stops swinging and starts describing you. Before that the numbers are a guess being corrected in public, and they can move a long way in a week — which is normal, not a bug. Ratings are always an estimate with uncertainty around it; more matches narrow the range, they never remove it.

Rivals keeps this rating for your whole crew — automatically

Log the match once and every rating in your crew moves on its own: margin-aware, private to the four of you, and able to show you exactly why each number changed.

Get Rivals — it’s free

Free for a crew of four. Nothing inside is metered.