Score only claims a source can confirm or deny. Values, slogans, jokes, and guesses about the future stay out of the 0–100.
- Accuracy = passes ÷ (passes + fails), shown as n/100.
- Coverage = checkable claims scored, not clips watched.
- Board slot only when coverage ≥ 10.
- Tiers follow school grades: A 90–100%, B 80–89%, C 70–79%, D 60–69%, F 0–59%. There is no E row.
- Every failed claim also gets a harm score from 1/100 to 100/100. Count how many independent assumptions sit between the false claim and real-world damage. Belief alone does not count as a step unless acting on that belief is already dangerous. One short step to death or irreversible injury is 100. A color mix-up like green vs teal needs a long chain before anyone could be hurt, so it scores about 1. The letter grade on the board still uses pass/fail count only; harm is a second label so a bleach-cure lie is not treated like a teal-vs-green slip.
- If they mix a fact with an opinion, score the fact only.
- Pass if the source matches. Fail if it does not, or if they made the source up.
- Check the full video, post, or article, not just a short clip. If they fix it there, count the fix.
- Same rule for every name. Party, audience size, and how much we agree with them are not scoring inputs.
How rules change
Propose a rule, an update, or a repeal on the home page. Current rules have keep / drop votes there too.
- Add or replace: 1,000 votes and 90% up.
- Drop or repeal: 1,000 votes and over 50% drop (or up on a repeal proposal).
- Then a 24-hour wait. If the ratio falls during the wait, nothing changes.
One vote per item per browser. A live host should use one vote per item per IP or account.
Photos
Default is initials. Add a photo only with a clear reuse right (public domain or a license). Credit the source. No random stock heads.