Your card carries a D&D alignment. Three playful questions put it there, the score barely notices it, and one tap from you overrules it completely. The whole mechanism, since we would rather explain the joke than hide it. 🎲
This is not a literature review. Nobody has validated an alignment chart, and dressing one up in research language is exactly the bluff every other app runs. So this page is about how we built it, what it is allowed to do to your score, and why it is on your card at all.
The interview asks three questions about other people, deliberately small ones. A claim that survives a late library book is worth more than one that survives a crisis nobody has had.
Keeps the itinerary, drives, names the van. That is the order axis, leaning lawful.
Makes sure nobody misses a meal. That is the moral axis, leaning good.
Reroutes on a whim and calls it an adventure. Naps in the back, happy anywhere. Both are real answers, and one of them is a refusal to be placed.
Wins on procedure and points of order. Orderly, and not in a warm way.
Defuses it, even when they were right. That is the good pole at work.
Makes it a highlight reel, reconciles by morning. Waits it out, baffled by the fuss.
Impossible, they are subscribed to the due-date reminder. About as lawful as a question gets.
Returns it personally and apologises to the librarian. Good, and a little embarrassed doing it.
The book is in another city, oops. Three weeks? Time is a suggestion.
Each answer leans one way on one of two axes: how the person relates to rules, and how they weigh other people. Two answered questions is the floor. Answer one and nothing is derived at all, because one vote is a guess. A tie on an axis lands on neutral for that axis, and the answer "happy anywhere" abstains instead of being forced.
| Good | Neutral | Evil | |
|---|---|---|---|
| Lawful | Lawful goodthe plan, and nobody gets left | Lawful neutralthe rule is the rule | Lawful evilthe fine print, weaponised |
| Neutral | Neutral goodkind without a rulebook | True neutralkeeps the balance, not the side | Neutral evilgets what they want, quietly |
| Chaotic | Chaotic goodbreaks the rules for you | Chaotic neutralfreedom first | Chaotic evilburns the table down |
The derived value always arrives flagged as needing your confirmation, with a confidence that tops out at 0.8. It is a proposal, worded as one, and it never gets to be more certain than that.
Here is the arithmetic in full, because this is the part an app normally keeps to itself.
Four percent, named in the open, reversible by you. That is the entire power the chart holds, and it is deliberately smaller than the effect of a single dealbreaker.
Because people answer it happily. It is the part of the interview that feels like a conversation instead of an assessment, and the answers are genuinely social: does this person keep the itinerary or reroute on a whim, do they defuse a dinner argument or win it. That information is real, and it is also not the kind of thing you want a scoring model making much of.
So it stays where a thing like that belongs. On the card, in the open, worth almost nothing, and useful the way a good opener is useful. The score is 50% Big Five similarity, 25% attachment fit and 25% dealbreakers. The chart is not in that list.
No research language attached to a game. No probe answers kept. No chart in the ranking beyond the 4% it declares, and no derived label that survives your own correction.
Want to change yours? The You page takes one tap. Ask our AI assistant if you cannot find it. 🍻
Create your profile