Fair Teams: Balancing 4 Padel Players of Different Levels

Quick Answer: Four players make exactly three possible pairings. For a single match, pair the strongest with the weakest — that is almost always the closest one. For a whole night, play all three pairings: everyone partners everyone once, and fairness stops being a decision you have to make.

There is a moment at the start of every crew night where four adults stand on a court and pretend not to be doing arithmetic. Everybody knows roughly who is best. Nobody wants to say it. So the pairs get chosen by whoever changes their shoes fastest, and then two of you spend forty minutes losing 6–1 and calling it a warm-up.

The fix takes ten seconds and it is not diplomatic. It is mathematical. It is also one stage of a larger system — the guide to weekly padel with friends covers the rest.

Four players make exactly three teams

Call your players A, B, C and D, strongest to weakest. There are only three ways to split them into two pairs:

That is the entire universe of options. Not "lots of combinations" — three. Which means you never have to guess: you can check all of them, every time, in your head.

And the pattern holds no matter who your four are. The strongest with the weakest is the balanced pairing. The intuition that feels fair — "let's split up the two best players" — is right, and the version that feels generous — "I'll take the weakest player, you two go together" — is the same thing, correctly executed.

The sum test, if you don't have ratings

Give everyone a number from 1 to 5. Not a Playtomic level, not a claim about your ability in the abstract — just where they sit inside this crew, decided in ten seconds by four people who play together every week.

Say you are 5, 4, 3 and 2. Add each pair:

• A + B = 9 vs C + D = 5 → a gap of 4
• A + C = 8 vs B + D = 6 → a gap of 2
• A + D = 7 vs B + C = 7 → a gap of 0

Pick the pairing whose sums are closest. That is it. The sum test takes five seconds, works with any numbers, and is right almost as often as anything more sophisticated — because at four players there are only three answers to choose between.

Where it stops being enough is when the spread is wide: a 5 and a 2 on the same team still means one player is being hidden and the other is covering the whole court. When the numbers are that far apart you are past pairing and into handicaps and expectation management.

What the arithmetic actually says

If your crew keeps ratings, you can put a number on how uneven a pairing is instead of arguing about it. Doubles rating systems do this by treating a team as the average of its two players — Playtomic's own explanation of its level system says it considers the level of all players and the combined level of each team — and once you have two team numbers, the expected result follows from the gap between them.

Take a real crew: Ana 1150, Marco 1050, Jonas 1000, Sofia 900.

Same four people, same court, same hour — and depending on which of three arrangements somebody picks casually, the night is either a coin flip or a 70/30. That is the whole argument for spending ten seconds on it.

Here is the conversion, so you can read any gap:

Anything under 100 is a real match. Between 100 and 200 the underdogs need something to go their way. Past 200 you are watching, not playing, and it is time to change something other than the pairs.

Two tools do this arithmetic for you: the fair pairs generator ranks all three splits by balance, and the padel win probability calculator turns two team ratings into a percentage. If you want to see how a result then moves those ratings — including why beating the pair you were expected to beat is worth almost nothing — run a night through the padel Elo calculator.

Rotating partners is the fairness mechanism

Here is the idea that makes all of the above mostly unnecessary.

Play all three pairings in one night and you have run a complete rotation. Each player partners each other player exactly once, and faces each other player exactly twice. Nobody had a lucky draw. Nobody spent the evening carried, and nobody spent it carrying.

Three matches, and the fairness is structural rather than negotiated:

Yes, match 1 is the lopsided one — that is fine, because match 3 is its mirror and match 2 sits between them. Over the night the individual record means something, which is exactly what a single match between fixed pairs can never tell you. If you win two of three while partnering three different people, you were the best player on court and there is no argument available.

This is why "we always play the same pairs" is the single most common way a crew's results stop being interesting. There is more on keeping the same four fresh across a whole season, including the schedule for when a fifth person shows up.

When ratings are wrong, and what to do about it

A number cannot see that Marco slept four hours, that Jonas has a taped wrist, or that Ana and Sofia have never once managed to decide who takes the middle.

So use ratings as the default and allow exactly one human override per night, stated out loud. "Ratings say Ana with Sofia, but Sofia's been on the right side all month and Marco can't play right at all — so Ana with Marco." Said openly, that is a tactical adjustment. Made silently, it is favoritism, and everybody notices.

Three factors that beat rating arithmetic more often than people expect:

• Sides. Two players who both only cover the right side make a bad pair regardless of level. Sides are a hard constraint; levels are a soft one.
• Chemistry. Some duos are worse than their ratings and some are better. It is real, it is measurable over a season, and it is not something you should guess at from two matches.
• Recent form. Somebody who has lost five in a row is playing below their number, and putting them with the weakest player is not kindness.

How to announce the pairs without insulting anyone

The arithmetic is fine. The wording is what gets you.

Never explain the pairing in terms of who is being helped. "I'll take Sofia so it's fair" is technically true and socially awful — you have just told one adult that they are the handicap in a game they paid to play.

Say the pairs. Nothing else. "Ana with Sofia, Marco with Jonas. First to six, golden point." If somebody asks why, "closest split" is a complete answer.

And frame the sides as jobs rather than rankings. "You take everything down your side of the middle, I'll cover the lobs" tells a weaker partner what they are for. "Just get it back over" tells them what they are not.

The one night not to balance

Balance is for regular Tuesdays. Twice a season, deliberately break it:

• The series night. Fixed pairs, best of three, same two against the same two, and the result goes on the record with names attached. Rivalries need a fixed matchup to grow in, and rotating pairs every night means nothing ever gets the chance.
• The dream-duo test. Put your two strongest together on purpose, once, and let the other two find out what defending actually is. It is instructive, it takes twenty minutes, and everybody remembers it.

Both are better if somebody writes the result down. Keeping a record of who beat whom is what turns "we split up the good players" into an actual competition.

Where Rivals fits

Rivals picks the pairs from your crew's real results rather than from a guess. The rating is crew-local — it exists inside your four or your six and nowhere else, so there is no global number to protect and no reason to duck a match — it is margin-aware, so a 7–5 and a 6–0 do not count the same, and it shows its work: tap any change and it tells you what it expected and what happened.

Free covers one crew of four players, nothing inside it metered, no ads. The pairing arithmetic above is free everywhere, including on paper.

FAQ

How do you make fair padel teams with 4 players of different levels?

Pair the strongest player with the weakest. With four players there are only three possible pairings, and strongest-plus-weakest against the middle two is almost always the closest of the three. If you have numbers for everyone — ratings or a simple 1-to-5 scale — pick the pairing whose two sums are nearest to equal.

Should the best two players ever play together?

Only on purpose, and rarely. Putting your two strongest together produces a lopsided match: in a typical crew that is a 70/30 on paper, which is not a competitive game for anyone on court. Save it for a deliberate one-off, and use the balanced pairing for normal nights.

Is it better to rotate partners or keep fixed pairs?

Rotate for regular play. Four players make three pairings, and playing all three in a night means everybody has partnered everybody once, so the individual record actually means something. Keep fixed pairs for occasional series nights where the same two play the same two and the result is supposed to matter.

How big a level gap can a padel pair absorb?

Roughly a 100-point rating gap between the two teams still gives the underdogs a real chance — about 64/36. At 200 points the stronger side wins about three times in four, and past that the match stops being a contest. When pairing cannot get the gap under 200, change the format instead: shorter sets, a handicap, or a shot restriction on the strongest player.

What if two players in the crew both want the same side of the court?

Treat sides as a hard constraint and levels as a soft one. Two right-side players make a poor team however well matched their ratings are, because one of them spends the match playing a side they cannot cover. Solve sides first, then balance the levels among whatever pairings remain.