This order is never a ranking of “the worst”: it is what people are asking about. Counts are aggregated per case and without identifying who searches — no IP, no cookies, no profiles.
Note also that “insufficient evidence” is never painted green. Not having been able to verify something does not mean it is clean: it means we do not know, and saying otherwise would be the most damaging mistake we could make.
It is the Ponzi Signal Index: the share of the rubric’s weight whose indicators were confirmed by archived evidence. It is a reproducible measurement — you can recount it and reach the same number. It is not a statistical probability and not a legal determination.
Coverage always travels with the number. A low index at 20% coverage does not mean “this is clean”: it means “we don’t know”. Below 40% we publish no index at all.
Illustrative example. No real case has been published yet.
- Archive first, reason later — all evidence is captured and sealed with its hash and date before any conclusion. Schemes erase their tracks when they fall.
- The rubric is applied, indicator by indicator — each one requires its own cited evidence. Marking it "because it is obvious" is forbidden.
- A verifier tries to refute it — an adversarial pass tries to show the evidence does not support the claim. When in doubt, it refutes.
- A person approves it, or sends it back — nothing is published until a human opens the citations and checks they say what the dossier claims.
- The evidence stays public — screenshots, documents, transcripts and hashes, visible and downloadable. Verifiable without trusting us.
Cases and their evidence live in a versioned repository rather than a database: every copy someone downloads is a complete archive with its history. If this domain disappears, the work does not. Copying this project is not merely allowed — it is the plan.
git clone https://github.com/Rastipunk/isitponzi.git npm install npm run build
No keys and no services to sign up for: the site is static and builds from the same repository.