How the 30-Question Sorting Test Works — and What It Can’t Do

The Sorting Hat on a table in the Great Hall

Most online sorting quizzes ask you to pick the house you already like. Ours tries to do something harder: infer a result from how you answer, then show you how close the call was. This page explains the scoring, the design choices, and the limits — because a quiz that hides its method is asking for trust it hasn’t earned.

Why 30 questions instead of 50+

A longer quiz is not automatically a better one. After testing, we found that 20 scenario questions plus 10 statements gave us enough signal to separate the four houses without turning the experience into a survey. The full run takes about six minutes. If a question does not help distinguish houses, it is noise, no matter how magical it sounds.

Two question types

1. Twenty scenario questions. Each asks what you would actually do — facing a challenge, handling a secret, choosing between loyalty and rules. Every option awards points to all four houses, with one house usually leading. There is no single “correct” answer, and the strongest option for a house is not always obvious.

2. Ten 5-point statements. You rate agreement from strongly disagree to strongly agree. Each statement carries a vector of weights; your answer multiplies that vector by −2, −1, 0, 1 or 2. This captures intensity, which a four-option question cannot. It also lets disagreement count against a house instead of being ignored.

How the scores work

Every answer adds to four running totals, one per house. Scenario answers typically contribute 0–10 points; statements contribute up to ±8. At the end, the highest total is your house.

Two details matter more than the totals:

Margin. We keep the gap between first and second place. A 99-point lead and a 3-point lead should not read the same, so the result page shows the split and labels close calls.
Tie-break. If two houses finish level, we resolve the tie in a fixed order that does not favour Gryffindor. Earlier versions of this test silently defaulted to Gryffindor on ties, which is exactly the kind of bias a fan-made quiz should not ship.

What the result means

The result is an interpretation, not a diagnosis. It reflects the values your answers leaned toward on one day, under one set of questions. It is not:

an official Sorting Hat result, or affiliated with Warner Bros. or J.K. Rowling;
a validated psychometric instrument;
a permanent verdict on your personality.

The books are explicit that the Hat considers choice as well as aptitude. A quiz can only model a small part of that. Treat the result as a prompt for reflection — or as the start of an argument with your friends.

How to get a more honest result

Answer for who you are on an ordinary Tuesday, not who you would be in a crisis.
Do not choose the house you want. The quiz cannot stop you, but the result becomes less interesting.
Use the back button if a question genuinely did not fit. Skipping is allowed and recorded as neutral.
Retake it after a few months. People change; the quiz is designed to be retaken, not to lock you in.

What we are improving next

We are reviewing the ten statements with native speakers in German, Japanese and Korean, adding explanations to the scenario questions, and testing whether a shorter 24-question version keeps the same accuracy. If you find a question that feels unfair or ambiguous, tell us — this is a fan project and the feedback loop is the point.

References

Wizarding World, “The Sorting Hat” — the official lore page for the Hat and the four houses.
J.K. Rowling, Harry Potter and the Philosopher’s Stone, Chapter 7: “The Sorting Hat.”
Pottermore / Wizarding World, Sorting Ceremony archives and house welcome letters.

How the 30-Question Sorting Test Works — and What It Can’t Do