How to split a list into balanced teams
- Paste the participants in the text box, one name per line. You can use names, initials or anonymous numbers. Empty lines are ignored, while two identical lines count as two separate entries, since two people can share a name.
- Choose whether you know the number of teams or the target people per team. For example, 23 people in four teams produces team sizes 6, 6, 6 and 5.
- Press Create teams. The page shuffles the entries securely and distributes them evenly. Press Shuffle again to draw a different arrangement and Copy teams to paste the result into an email or chat.
All names are kept in browser memory while the tool is open; they are not submitted to the server or embedded in a shareable URL. Close the page or clear the text box to discard the roster. Be mindful of your own clipboard history if the names are private.
How fair are the groups?
The tool uses Fisher–Yates shuffling. At each step, it chooses an eligible position with browser-provided cryptographic random values. Sampling uses rejection rather than simply taking a remainder, eliminating modulo bias. After shuffling, names are distributed in a round-robin sequence among the teams. This ensures team sizes differ by no more than one person.
For instance, ten students split into three groups form groups of 4, 3 and 3. A list of twenty students split into five groups produces five groups of four. When the requested target size is four for twenty-three participants, the page makes six teams, most with four people and one with three. It will never hide a participant or silently create an empty team.
Random assignment versus skill balancing
A random draw gives each person an equal opportunity to appear in any group, but it does not guarantee equal skill levels, geographic distribution or social compatibility. A sports coach who must balance experienced and beginner players should collect suitable information and apply explicit constraints rather than treating random groups as skill-balanced groups. Likewise, an organizer arranging classroom support should review accessibility and safeguarding needs before assigning the final groups.
This version does not try to infer people's abilities from names, and it does not silently remove duplicate strings. If two different people both enter as “Alex”, they each receive their own place. To distinguish them in copied results, add an initial or a participant number before shuffling.
Where can you use this tool?
- Schools: split students into laboratory groups, debate teams or reading partners, while a teacher checks any required accommodations.
- Workshops: create breakout groups for a meeting with a fixed number of rooms.
- Community sports: set a fixed team size and make a fresh draw for each round.
- Board games: assign players without taking turns choosing favorites.
If you only need one random participant instead of several teams, try Spin the Wheel. To shuffle numerical assignments, use the number generator. Both are distinct from evenly balanced group assignment.
Frequently asked questions
Are the teams evenly sized?
Yes. Ten people in three teams yields group sizes four, three and three.
What about two people with the same name?
Both lines are kept as two separate participants. Add an initial if they need to be distinguished.
Is the list saved or uploaded?
No. The roster stays in this page while it is open and is neither sent to a server nor included in a share URL.
Does random mean skill-balanced?
No. The groups balance size, not skill or accessibility considerations. A human organizer should review those needs.
Last updated: October 17, 2026