Who does the dishes? Which restaurant? Which name goes first on the paper? Choosing fairly is an ancient problem — straws and dice were its first tools, and random pickers are the modern version.
What “fair” means here
A fair picker gives every option an equal, independent probability. Two properties matter:
- Equal: every entry has the same chance — no bias toward the first or last item.
- Independent: each pick is unaffected by previous picks (unless the tool supports “no repeats,” which changes the rules deliberately).
The failure modes
- Predictable “random”: a naive generator with a fixed seed picks the same sequence every time — anyone who knows the seed knows the result. Raffles have been rigged this way.
- Hidden bias: tools that weight items silently (or bugs that favor index 0) produce “random” results that aren’t.
- Human attempts at randomness: having a person “just pick” reintroduces bias — people favor middle options and avoid extremes.
Using a picker
- Paste your list of options into the random picker.
- Pick once, or spin repeatedly for a winner.
- For “everyone gets a turn,” use a no-repeat mode rather than re-rolling and re-rolling past winners.
The trust rule
For real stakes — raffles, seat assignments, tournament brackets — use a picker built on a cryptographically secure random source, and do the draw publicly. Fairness is not just about the algorithm; it’s about the process being verifiable by the people affected.
A picker turns “someone has to decide” into “nobody decided” — which is, oddly, the fairest decision of all.