Twenty-five of the released documents were rated contested. They are kept in the collection on purpose, so the rating can be checked. This is exactly why each one was set aside.
Every document in the corpus was read and rated the same way. Thirty-eight held up as standout cases. The great majority are circumstantial. And twenty-five, gathered here, were marked contested.
A collection that only showed its strongest evidence would be advocacy. This page does the opposite. It puts the weakest records in one place and says, plainly, what is wrong with each, a mundane explanation the witness offered themselves, a known piece of physics, a drawing instead of a photograph, a redaction that leaves nothing to read, a cultural artifact mistaken for a sighting, or an official explanation doing its job.
Keeping these visible is what lets the rest be trusted. Read alongside The Standout Cases, this is the other half of the same method.
Some of these are honest baselines, what orbital debris or a wind-driven balloon actually looks like on a sensor. A few are the very explanations a strong case has to survive. None of them carry the weight of the standout set, and that is the point: the line between them is drawn in the open.
The case is only as honest as the documents it leaves out, so this page leaves none of them out.