What is Would You Rather?
134 impossible-choice questions across eight moods — Classic, Funny, Chaos, Food, Gaming, Wholesome, Impossible and Relationship — split between genuinely tricky dilemmas and ones written purely to start an argument. It's built for the moments a conversation needs a spark: long car journeys, dinner with people who have run out of things to say, a party that needs an icebreaker with actual stakes. Each mood is shuffled when you open it, so two sessions don't run the same order, and a streak counter tracks how many you have worked through. There is no voting and no scoring: your answer is yours, nothing is sent anywhere, and the interesting part is the argument that follows rather than a percentage. Everything stays clean enough for mixed company and family settings — the tension comes from the choice itself, never from anything awkward to read out loud in a group.
The trolley problem is a would-you-rather
The most studied thought experiment in modern philosophy has exactly this shape. Philippa Foot set it out in 1967: a runaway tram is heading for five people, and you can divert it onto a track where it will kill one. Would you rather pull the lever or leave it alone?
Judith Jarvis Thomson later added the version that makes it interesting. Same five people, same runaway trolley, but now you are on a footbridge beside a large stranger, and the only way to stop it is to push him off. The arithmetic is identical — one life against five — and the answers are not. Something close to nine in ten people say they would pull the lever. Only a small minority say they would push.
Nobody has ever produced a tidy justification for that gap, which is precisely why the question is worth asking. The answer is not the point. The reason you give for the answer is the point, and you do not find out what yours is until someone makes you pick.
Why there is no “neither” button
Every group playing this eventually produces someone who says 'neither' or 'both', and the game stops dead for as long as they are allowed to.
Survey designers hit the same problem and solved it the same way. Offer a neutral middle option and a large share of people will take it — not because it reflects what they think, but because it is the cheapest response available. Removing the midpoint forces a position, and a forced position is something you can actually discuss.
That is the entire design of this format. Two options, both costly, no exit. The discomfort is load-bearing: a would-you-rather where one option is obviously better is not a question, it is a statement.
The fastest way to learn something about someone
In 1997 the psychologist Arthur Aron published a procedure for manufacturing closeness between strangers: 36 questions, asked in turn over about 45 minutes, escalating from the trivial to the genuinely personal. Pairs who worked through them reported feeling markedly closer than pairs who made small talk for the same length of time. It is the study that later travelled the internet as 'the 36 questions that lead to love'.
What makes it work is graduated self-disclosure — each answer makes the next, slightly more revealing one feel normal. Would-you-rather is the shallow end of exactly that staircase.
This is why it beats small talk in a car or at a dinner table. Asked directly what someone values, most people produce something guarded and dull. Asked whether they would rather be famous and broke or wealthy and unknown, they answer instantly and truthfully, because it is obviously just a silly hypothetical — and then they explain why, which is the part that was never silly at all.
How to actually run it in a group
Make everyone commit before anyone explains. Answers given after hearing someone else's reasoning are not answers, they are agreement. Count down from three and have everyone say it at once if the group is the sort that drifts.
Then ask why, and only then. The choice takes two seconds and the reason takes five minutes, and the reason is the game. The best questions here are the ones where two people pick the same side for completely incompatible reasons.
Ban 'neither', ban 'it depends', and be wary of the group's tendency to converge. If everyone lands on the same side, the question was not a dilemma — skip it rather than labouring it. Switching mood mid-session is a reliable reset: Food and Gaming produce loud disagreement, Wholesome and Relationship produce quieter and more revealing answers, and Impossible is for when the group has warmed up enough to enjoy being stuck.
What the tool does, and what it does not
There are 134 questions across eight moods — Classic, Funny, Chaos, Food, Gaming, Wholesome, Impossible and Relationship — with 16 to 18 in each and nothing shared between them. Switching mood shuffles that mood's questions, so no two sessions run the same order.
It does not collect votes and it never did the thing an older version of this page described: there are no percentages, no tally of how other people answered, and nothing you tap leaves your device. That is deliberate. A displayed majority changes the answers people give out loud, which is the one thing this format cannot survive.
The streak counter tracks how many questions you have worked through in the current session, and the saved list holds the ones you want to use again. Both reset when you reload the page.