Our networkdiscount5Fresh deals. Five at a time.Price drops and coupon codes, ending soonest first.See today’s deals
Fun · Randomizer

Random Team Generator

Enter a list of players or participants and split them into random, balanced teams in one click. Choose the number of teams or the team size. Perfect for sports, games, class projects, and work groups.

Last updated

By Team Count or Size
Balanced Teams
Shuffle & Regenerate
Copy All Teams
Random Team Generator
Free · Instant · No signup

How Fair Team Selection Works

The list is shuffled first, then split, and with Spread (the default) team sizes never differ by more than one: 10 players in 3 teams gives 4, 3 and 3. Paste one name per line, choose a number of teams or a team size, pick how leftovers are handled, and generate. Every possible split of the group is equally likely.

Participants are first shuffled using the Fisher-Yates algorithm, the mathematically proven method for generating a uniformly random permutation. They are then split into teams in that shuffled order. With Spread, the default, leftover players go one each to the first teams, so team sizes never differ by more than one.

Fisher-Yates Shuffle

Every possible arrangement of participants is equally likely. No position bias, no alphabetical clustering.

Balanced Sizes

With Spread, the default, teams differ by at most 1 person. Add to last team puts every leftover player on the last team, and Alternates keeps them apart as subs.

Team Sizes by Leftover Setting

When the group does not divide evenly, the leftover setting decides where the extra players go.

PeopleTeamsSpread (default)Add to last team
724, 33, 4
1034, 3, 33, 3, 4
1143, 3, 3, 22, 2, 2, 5
2346, 6, 6, 55, 5, 5, 8
3048, 8, 7, 77, 7, 7, 9

In Team size mode the tool makes as many full teams as fit. With 10 people and a size of 4 that is 2 teams: Spread gives 5 and 5, and Alternates gives 4 and 4 plus 2 subs.

How Many Ways to Split a Group?

Even small groups have a lot of possible line-ups, which is why the same teams rarely come up twice.

GroupPossible splitsChance two friends are together
8 people, 2 teams of 43542.9%
10 people, 2 teams of 512644.4%
12 people, 2 teams of 646245.5%
12 people, 3 teams of 45,77527.3%
20 people, 4 teams of 5488,864,37621.1%

The chance that two named people end up on the same team is (team size minus 1) divided by (group size minus 1): once your friend is placed, only the remaining seats on that team count.

Random Teams vs Skill-Balanced Teams

A random split is fair in process, not in strength: now and then all the strongest players land together. For a closer game, rank the players and use a snake draft. With 2 teams the picking order is A, B, B, A, A, B, B, A. Ranked 1 (best) to 8, team A gets players 1, 4, 5 and 8 and team B gets 2, 3, 6 and 7, and both rank totals come to 18. With 3 teams the order runs A, B, C, C, B, A and repeats.

A good mix: snake-draft the top players as captains or seeds, then use this generator for everyone else.

How the shuffle works

The page uses the Fisher-Yates shuffle with Math.random, the browser's standard pseudorandom generator. Every order of the list is equally likely, so it makes no difference whose name you type first. Everything runs in your browser.

Method and sources. Team sizes are read from the page code for each leftover setting. Number of splits into k unlabeled teams of equal size s: n! ÷ ((s!)k × k!). Chance two people share a team: (s − 1) ÷ (n − 1). All values computed in node. Shuffle: Fisher-Yates (Durstenfeld) algorithm.

Frequently Asked Questions

Participants are shuffled, then split into teams in that order. With 10 people and 3 teams you get 4, 3 and 3. Spread (the default) gives extra players to Team 1, then Team 2 and so on, so sizes never differ by more than one. Add to last team puts every extra player on the last team, which can make it several players bigger.

Yes. Switch the "Split By" dropdown to "Team size". Enter the desired number per team and the calculator determines how many teams are needed automatically.

There is no hard limit. In practice, hundreds of names work fine. The tool runs entirely in your browser.

Yes. The Fisher-Yates shuffle ensures every possible grouping is equally likely. There is no bias toward alphabetical order, order of entry, or any other pattern.

Yes. Click Generate Teams again for a completely new random assignment of the same participants.

If participants do not divide evenly, the remaining player(s) are placed in a separate "Alternates" team rather than added to an existing team. Useful for sports where you want equal-sized playing teams with subs.

Not currently: teams are labeled Team 1, Team 2, etc. You can copy the results and rename them in any document editor.

Yes. Random team assignment removes social bias in group formation, forces people to work with different colleagues, and is seen as fair by all participants since no one was deliberately included or excluded.

Yes for quick pickup games and round-robin seeding. For competitive leagues you may want balanced teams by skill level rather than random: in that case, pre-rank players and use a snake draft order instead.

Click "Copy All Teams" to copy the complete team list to your clipboard. Paste into any messaging app, email, or document.

It is (team size minus 1) divided by (group size minus 1). With 10 people in 2 teams of 5 that is 4 in 9, about 44.4%. With 12 people in 3 teams of 4 it drops to 3 in 11, about 27.3%.

No. The whole list is shuffled with the Fisher-Yates method before any team is formed, so every order is equally likely and the first name typed has no better or worse chance of any team than the last.