Fair Pairs Generator — balanced 2v2 in one tap

Four players can be split three ways and no more. Put the names in, rate each other if you want to, and see all three options ranked by how even they are — or let the random pick settle it before anybody starts negotiating on court.

The four
Player 1
Player 2
Player 3
Player 4

Levels run 1 to 5 — 1 is the newest, 5 is the one everybody wants to partner. Skip them if you do not rate each other: four equal levels make all three splits even, and the random pick is the honest way to choose.

Method

How the splits are ranked

Nothing is searched or optimised here, because there is nothing to search. Four players can be split into two pairs in exactly three ways and no more: 1 and 2 against 3 and 4, 1 and 3 against 2 and 4, and 1 and 4 against 2 and 3. The tool enumerates all three, every time.

The ranking is one subtraction. Each player carries a level from 1 to 5 — 1 the newest, 5 the one everybody wants to partner. For each of the three splits, add the two levels on each side and take the absolute difference. That difference is the gap, and the splits are sorted by it, smallest first; a tie keeps the canonical order above, so the same four levels always produce the same ranking in the same order.

The 1-to-5 scale is ours and it is deliberately coarse. A five-step scale is about as fine as people can honestly rate each other in a car park, and anything finer invites an argument about whether someone is a 6.4 or a 6.6 — which is the argument the tool exists to end.

One case is handled separately. If all four levels are equal, all three gaps are zero, there is genuinely nothing to rank, and pretending otherwise would be dishonest — so the random pick becomes the primary action and the page says all three splits are even. That is also the honest answer for a group who do not rate each other at all: leave the sliders alone and let chance choose, then rotate through all three splits across the night, which gives everybody every partner exactly once.

Rivals is not affiliated with any padel federation and nothing here is coached, reviewed or official. Where a number is our own judgement rather than a published figure, the note above says so and cites nothing. Method last checked 21 August 2026.

Why there are only three ways to split four players

With four players there are exactly three ways to split into two pairs: 1 and 2 against 3 and 4, 1 and 3 against 2 and 4, or 1 and 4 against 2 and 3. Nothing else exists. Knowing the number is three is what turns an argument into a choice — you are not searching for the fairest teams, you are ranking three options.

Once you rate each other from 1 to 5, balance is just the two sums. A split where both sides add up to the same number is even; a one-point gap is close enough that the serve matters more than the pairing; a four-point gap means one pair is carrying. Splitting the two strongest players is usually the fairest call, because it puts the biggest single difference on both sides of the net instead of stacking it on one.

When the four of you are genuinely level, all three splits come out even and the honest answer is to pick at random. Playing all three across a night is the same thing as a full rotation: everyone partners everyone once, which is what the rotation planner lays out round by round.

Questions

How do you make fair teams in padel?

Rate the four players roughly, then compare the two sums. There are only three possible splits, so the fairest one is whichever pairing leaves the smallest gap between the sides — usually the one that puts the strongest and the weakest player together. This generator ranks all three and shows the gap for each, so nobody has to negotiate on court.

What if we do not know our levels?

Leave the sliders alone. With four equal ratings all three splits come out even, so the fairest thing you can do is take one at random and play — which is exactly what the random button is for. If you want a number to argue from, a padel level test turns a dozen honest questions into a 0 to 7 self-assessment you can put in the sliders next week.

Should the two best players be on the same team?

Usually not. Pairing the two strongest players together makes one side heavy favorites, and a lopsided match is less fun for all four, winners included. Splitting them puts the biggest single difference on both sides of the net, which is what makes a close scoreline likely. The exception is a deliberate handicap match, where the whole point is finding out whether two can hold off the other two.

How do you keep teams fair across weeks, not just tonight?

Write down who won. One night's fair split is guesswork about levels; a season of results is evidence. Rotate through all three splits so everyone partners everyone, and keep the scores somewhere that lasts — that is the difference between a page that forgets and a crew board that turns fifty matches into a rating nobody can argue with.

More tools

Read next

Fair pairs from real ratings, not guesses

A slider is an opinion about who is good. Rivals rates your crew from the matches you actually played, generates the pairings itself, and keeps every result — so next week’s teams come from evidence.

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