The wire produces a byproduct: a systematic record of how strange claims move through the press. This is where that gets measured, under rules set in advance.

Every analysis below was written down — hypothesis, method, confounds and the earliest date we are allowed to look — before the data to test it existed. Results are published whatever they say. A register where most entries read “no effect” is the point, not an embarrassment: it is the only thing that makes the occasional positive worth anything.

Two standing limits on everything here. This data measures reporting, not phenomena — report counts track newsrooms and population, not objects in the sky. And corroboration cannot be fully separated from syndication, so every rate below is an upper bound.

Everything above is available as JSON at /api/v1 under CC BY 4.0 — the gradings, cluster structure, corroboration timings, daily metric series and the register itself. Article excerpt text is deliberately excluded; it belongs to the publishers.

If you use it, cite it. A permanent archived release with a DOI follows once there is a year of data behind it.