The easiest way to divide teams for party games is to choose the method before anyone starts campaigning. Use a random draw when speed matters, count off after mixing the group for the fastest no-equipment split, or make one quiet host adjustment when experience, accessibility or an obvious mismatch matters. The goal is not to prove that two teams are perfectly equal. It is to create a split everyone can accept quickly enough to start playing.
“Fair teams” can mean different things. One group wants a visibly random process. Another wants both teams to have someone who knows the game. A third simply wants existing friend circles mixed up. No single method guarantees the same experience for every room, so decide what you are trying to balance before you divide anyone.
For mixed groups, begin with the low-pressure approach in Party Games for Groups Where Not Everyone Knows Each Other. For very large groups, pair this guide with How to Keep a Large-Group Game Moving.
The team-split wheel
Pick the segment that matches tonight's problem. Random assignment is quick and transparent, while intentional assignment asks the host to use limited, relevant information. Cornell's teaching guidance makes a similar distinction: random assignment is straightforward, but small adjustments may help when constraints or experience matter. Montclair's group guidance lists counting off as a simple in-person random-assignment method. This party version keeps the process much lighter.
Count off
Mix the standing order, then call 1, 2, 1, 2 around the circle.
Random draw
Use equal numbers of two colours in a bowl and let everyone draw once.
Band and alternate
Use broad, self-described experience bands, then alternate people across teams.
Pair and split
Pair people who arrived together, then place each person on a different side.
Rotate one role
Rotate a floater, clue reader, timekeeper or substitute between short rounds.
Five ways to divide teams without a public popularity contest
Random draw
Prepare the exact number of slips needed for each team. For two teams of five, use five red slips and five black slips. Fold them identically, mix them where everyone can see, and allow one draw per person.
This gives the process a clear random mechanism when the slips are balanced and mixed. It does not guarantee equal skill, confidence or familiarity with the game. Use it when the group values a neutral-looking split more than a carefully balanced one.
Mix, then count off
Ask everyone to change places first. Then move around the circle saying 1, 2, 1, 2. All the ones form one team and all the twos form the other.
The mixing step matters. If people count off while standing in couples or existing clusters, the pattern may produce a predictable split. This method needs no props and is useful when the game is low-stakes or the group has roughly similar experience.
Pair and split
Quietly identify pairs who arrived together or already know each other well. Place one person from each pair on either team. Continue until both sides have a mix of familiar and unfamiliar faces.
Do not announce private relationships, rank friendships or turn the process into commentary about who is close to whom. The host can simply say, “Let us mix the room a little.”
Band and alternate
Use this only when experience genuinely affects the game. Ask for a simple self-description such as first time, played a few times, or knows the game well. Shuffle within each band, then alternate people between teams.
Keep the bands broad and voluntary. Do not label people as good or bad players. This is a practical distribution method, not a public ranking system.
Random first, one adjustment second
Begin with a random draw. After the teams appear, allow the host one stated adjustment for an obvious constraint: two people need the same accessible seating area, both rule-knowers landed together, or one side has an extra person.
Name the reason without oversharing. Limit the repair to one swap or one rotation plan so the split does not become a negotiation.
How to split teams fairly, step by step
Count the active players
Include only people playing this round. Decide whether a host, clue reader or gamemaster is inside or outside the team count.
Name the thing you are balancing
Choose one priority: speed, randomness, experience, social mixing, accessibility or an odd number. Trying to optimise everything invites debate.
Choose the method out loud
Say, “We are drawing slips,” or, “I am spreading the people who know the rules.” A short explanation makes the process easier to follow.
Run it once
Do not redraw because someone dislikes the first result. Change only for a concrete constraint or an obvious counting error.
Give both teams the same start
Use the same explanation, practice turn, time limit and scoring rule. Cheatr's guide to explaining a new game offers a useful sequence.
Review after a short round
If one side is clearly struggling, change a role, adjust the next round or reshuffle later. Do not rewrite the result halfway through a point.
What to do with an odd number of players
An odd headcount does not always require one person to sit out. Choose a structure that fits the game:
- Rotate a floater: one person changes sides after each short round. Publish the order before play starts.
- Use a neutral role: the extra person can read prompts, keep time or track points when the game needs that role.
- Give the smaller team a turn advantage: where the rules allow it, the smaller side can start or receive the first choice. Agree before play.
- Change the team size: three small teams may fit a relay better than two uneven large teams.
Rotate the extra role so the same person is not repeatedly left outside play. If a late arrival changes the count, use the practical handover ideas in the late-arrival hosting guide.
Quiet fixes for common team problems
Swap one willing player
Ask for a volunteer from the more experienced side and explain that both teams need a rule anchor. Do not nominate the “weakest” person on the other team.
Change roles before people
Give each team a rotating spokesperson or clue-giver. A role change may address the game flow without treating personality as a defect.
Use random draw
Let the visible mechanism make the decision. If the activity is only for fun and the room strongly prefers self-selected teams, accept that perfect balance is not the only goal.
Prioritise access
Place people where they can see, hear, move or participate comfortably. Make the smallest necessary change and avoid explaining someone else's needs to the room.
The host's fairness test
Can you explain the method in one sentence? Did everyone have the same chance within that method? Did you avoid public ranking, private disclosures and endless redraws? If yes, the split is probably fair enough for a party game. Start the round and let the teams earn the rest.
What not to do
- Do not run an open schoolyard draft. The final pick becomes a public judgment, even when nobody intends it that way.
- Do not let captains quietly stack teams. If you use captains, form the teams first and let captains choose only a name, colour or starting side.
- Do not guarantee perfect fairness. A random split can be uneven; an intentional split depends on limited information.
- Do not use sensitive personal traits as balancing criteria. Keep the decision relevant to play, access and group comfort.
- Do not restart repeatedly. One transparent method and one limited repair are usually enough.
For ten-player formats and team-game ideas, see Party Games for 10 People.
Teams sorted. Now choose the game.
If you want to compare Cheatr's two current games after the split, the live bundle page lists Boom Machale, Khufia Mafia and the Cheatr dust bag together. Check the official player information there before choosing what fits the room.
View the Cheatr launch bundleFrequently asked questions
What is the fairest way to divide teams for a party game?
There is no single method that guarantees perfectly equal teams. Use a visible random draw when neutrality and speed matter. Use a limited intentional adjustment when relevant experience, accessibility or an odd headcount matters. Explain the method before using it.
How do you divide two teams without equipment?
Ask people to change places, then count around the group: 1, 2, 1, 2. All the ones form one team and all the twos form the other. Mixing the order first helps prevent existing pairs from creating a predictable pattern.
How can I make balanced teams when skill levels differ?
Invite people to choose a broad experience band such as first time, played before or knows the game well. Shuffle within each band and alternate people across teams. Keep the labels voluntary and avoid ranking players publicly.
What should I do when there is an odd number of players?
Rotate one floater between short rounds, use a neutral timekeeper or prompt-reader role, create three smaller teams, or agree on a small turn advantage for the smaller side where the rules permit it. Publish the rotation before play.
Should party-game teams choose themselves?
Self-selected teams are fast and may suit a very casual room, but they can preserve existing friend circles or create uneven experience. If social mixing matters, use pair-and-split or a random draw instead.
Team-formation methods were checked against current Cornell and Montclair teaching guidance and NIST's definition of randomisation on 9 August 2026. The advice is adapted for low-stakes party play and does not promise equal outcomes.
