Chargement...
2 quick questions before we start
100% anonymous — just for the stats
Red or Green puts two behaviors side by side and asks you to pick the worse one. Here’s what the game actually measures, how the votes build the ranking, and what the numbers don’t tell you.
Two behaviors appear. You pick the one that seems more problematic to you. There’s no right answer: the question is about your judgment, not an established truth.
Once you’ve chosen, the game shows the share of players who made the same call on that exact duel. It measures how much you agree with the community — it’s not a score for getting it right.
Asking people to “rate this behavior from 1 to 10” produces answers that are hard to use. Everyone sets their own scale, two people who answer 7 don’t necessarily mean the same thing, and ratings tend to bunch up in the middle.
A comparison doesn’t have that problem. Choosing between two options forces a decision, and the result reads the same way from one player to the next. It’s the same principle used to rank chess players: you don’t measure a level in the abstract, you look at who beats whom.
Every behavior starts with a score of 1,000 points. After a duel, the winner takes points from the loser. How many depends on the gap between them: beating a behavior already judged very problematic earns a lot, beating a harmless one earns little.
A newly added behavior moves fast, then settles as votes pile up — past about a hundred duels, it only moves with many consistent votes. The methodology page explains the math in detail.
The ranking is something to talk about, not a personal assessment tool. To take stock of a real situation, the self-assessment tools in our resources — the violence meter, the consent meter — are built for that and rely on scales published by public institutions.
If you are in immediate danger in France: 17 (police emergency), 3919 (helpline for women facing violence, free and anonymous), 119 (children at risk). Anywhere in the EU, 112 also works.
No. No account is needed and no email address is asked for. The game runs in your browser.
Only to feed the group rankings, which measure the gaps in judgment between men and women or between age brackets. These two answers aren’t tied to any identity and aren’t used for anything else.
No. Pairs are drawn each round using several strategies: matching two behaviors with similar scores for a close duel, mixing two categories, or picking at random so that rarely seen prompts get votes. A duel you’ve already played in a session won’t come back.
Prompts are written and reviewed before they go into the game: they aren’t generated automatically. Suggestions can be sent to chadizarkal25@gmail.com.
The “Love & Sex” category is not shown to players who select the 16–18 age bracket, and they can’t choose it either. The rest of the content is suitable for all audiences.
