Tool
Would You Rather Generator
Draw a dilemma, tally the room's split, and see where the group actually lands.
Would You Rather Generator
20 dilemmas across 12 themes.
Draw a dilemma, let everyone tap the side they would choose, and watch the split appear. Twenty dilemmas across ten themes, and none of them has a correct answer.
The tally is the actual feature. A would-you-rather question with no record of what the room said is a question that evaporates; a 9-to-1 split is a fact about the group that people will argue about for the rest of the evening.
The generator does not repeat a dilemma within a session, so a long game keeps producing new material rather than looping back to the first three.
How the Would You Rather Generator works — and why it's fair
Dilemmas are drawn from a no-repeat cycle over the full set: each draw takes one from the pool of dilemmas not yet used in this session, and the pool refills only when it empties. Over a complete cycle every dilemma appears exactly once, which is a coverage guarantee rather than an equal-chance one — the last draw of a cycle is a certainty, not a one-in-twenty.
The two sides of each dilemma are stored in a fixed order and displayed in that order. This is worth flagging because it is a small, real, deliberate asymmetry: there is a well-documented tendency for people to pick the first option they read, and the tool does not randomise the side order to counteract it. Randomising would make the tally cleaner and the game worse, because the same dilemma reading differently on two occasions is confusing in a room where people are shouting at each other.
The tally itself does nothing to the draw. Voting does not make a dilemma more or less likely to reappear, does not feed into a model of your group, and is discarded when you draw the next one. It exists purely so the room can see the split.
Nothing balances the two sides. Several dilemmas here are lopsided by design — most groups split heavily on "never wait in a queue again" versus "never have a bad night's sleep again" — and a 9-to-1 result is information about your group rather than a fault in the question. A generator that tried to produce evenly-split dilemmas would be selecting for blandness.
The theme labels are descriptive rather than functional. They are not filters, and the draw does not balance across them; you will sometimes get two Memory dilemmas in a row, because a no-repeat cycle over twenty items makes no promises about spacing.
When the Would You Rather Generator is fair — and when it is not
What it does guarantee
- Every dilemma appears exactly once per cycle, so nothing is left out over a long game.
- The vote tally is display-only. It cannot influence which dilemma comes next.
- Nothing is stored beyond the current session, and nothing about how your group voted leaves the page.
What it does not
- Side order is fixed, and people demonstrably favour the first option they read. The tool does not correct for this.
- The draw does not spread themes out. Two dilemmas on the same theme can land back to back.
- The dilemmas were chosen to be interesting, not to split evenly. A lopsided result is the expected outcome for several of them.
Two worked examples
A group of eight, first draw of the evening
- The pool is all 20 dilemmas, so the first draw is 1-in-20 across them.
- Eight taps produce a split such as 6 to 2 — displayed as 75% and 25%.
- The next draw is 1-in-19 from the remaining pool, not 1-in-20 again.
Late in a long session
- After 19 draws, exactly one dilemma remains unused.
- The twentieth draw is a certainty, not a one-in-twenty — which is what a coverage guarantee costs.
- The twenty-first draw refills the pool and the cycle restarts from all 20.
Where a tally beats a question
Long car journeys and pub tables are the obvious home, and the tally is what makes it survive past three questions. Without a record, a would-you-rather round is a series of unconnected opinions; with one, it becomes a running argument about who in the group is strange.
It works surprisingly well as a low-stakes team activity precisely because none of the dilemmas is about work. A room that will not answer "what is going badly" will happily fight about whether unlimited time beats unlimited money, and the shape of the disagreement is often informative anyway.
Teachers use it as a discussion warm-up where the point is justifying a position rather than holding the right one. The visible split gives the quieter half of the room something concrete to respond to.
For questions that get answered rather than voted on, the conversation starter picker climbs a depth ladder instead. For a two-option decision you actually need settled, the this-or-that picker will just pick one.
Frequently asked questions
Why does the tool not balance the two options?
Because a lopsided split is a real result about your group, and selecting for evenly-split dilemmas would mean selecting for the dullest ones.
Does the order the two sides appear in matter?
It probably does — people tend to favour the first thing they read — and the tool deliberately does not randomise it, because the same dilemma reading differently on two occasions is confusing mid-game.
Can the same dilemma come up twice?
Not until every other dilemma has been used. The cycle then refills and starts again, which the counter shows.
Does the tally affect which dilemma comes next?
No. Votes are display-only and are discarded when you draw again.
Are the theme labels filters?
No, they are descriptive. The draw ignores them entirely, so two dilemmas on the same theme can appear consecutively.
Is the voting anonymous?
It is a shared screen and shared taps, so it is exactly as anonymous as your room is. Nothing is recorded anywhere else.
Can I add my own dilemmas?
No. The twenty here were chosen to split a room rather than to be balanced, which is a property an open list loses immediately.
Does the session survive a page reload?
The no-repeat cycle is kept in this browser, so a reload usually continues where you left off. Vote tallies are not kept.
How many people can vote?
As many as you like — each tap increments the count. It is a tally, not a per-person poll, so someone tapping twice counts twice.