Random teams are not automatically balanced teams

Equal size answers just one question

With 21 people and four teams, the closest sizes are 6, 5, 5 and 5. If nobody should work in a group larger than five, four teams cannot meet that requirement. Five teams give 5, 4, 4, 4 and 4.

For collaborative classroom tasks, the Education Endowment Foundation highlights promising approaches using small groups and a shared outcome. It also rates the evidence as limited. That does not make one group size right for every class, sport or workshop.

Source: Education Endowment Foundation · Collaborative learning

Separate requirements from nice-to-haves

A task may need one equipment-trained person in each group, an accessible station, or two participants kept apart. Picknado's random team generator does not infer these constraints from names or enforce them automatically.

For example, if four groups each require a trained facilitator, assign those four facilitators first. Shuffle the remaining participants and place them into the available slots. That is a constrained plan with a random step, not a fully random division of everybody. Tell the group which rule you used.

Do not redraw until the result looks convenient

Repeatedly rerolling until friends separate or a strong player moves changes the selection process. If a requirement is important, state it before the draw and apply it consistently. If a group really needs changing afterward, explain the change rather than presenting the final arrangement as untouched randomness.

For a learning task, define the output and how everyone will contribute. An organiser can assign a reader, recorder or timekeeper and rotate those roles later. The group draw does not itself make quieter participants heard.

Use rotation for a different goal

If the aim is to meet different people, permanent teams are not the right structure. Partner rotation produces successive pairs without repeats within the generated schedule. An odd-sized group has a resting participant each round.

Choose the Team Generator for one grouping, and its rotation mode for successive conversations. The separate workshop guide explains timings and complete cycles. Neither mode assesses ability, guarantees social comfort or replaces an organiser's judgment about the activity.