STATE OF
THE EVIDENCE
Our first audit of the Fractured Realities archive — which claims actually stand out from the noise, and which don't.
We take the twelve biggest themes in the archive and ask a simple question of each: does searching for it turn up more than a plain, vague "UFO" search already does? Most don't — they're no more distinctive than generic chatter. One theme clearly stands apart, three more edge just past the line, and exactly one is backed by published science.
Of everything in the field, what actually holds up? We test every theme the way a skeptic would. Search the archive for a specific theme, then compare that to a plain, vague "UFO" search. If the specific theme turns up meaningfully more, it's a real, distinctive pattern — not just people saying "UFO." We score it as a multiple of that plain search: 1× means no more distinctive than generic chatter; above 1× it starts to stand out; above 2× it's clearly its own thing. Most themes never beat 1×. Only one clears 2× — Navy radar-visual encounters; three more sit just above the line; and one is backed by published science.
One caveat worth stating plainly: this test measures how distinctive a theme's wording is — not whether it's true. A theme can score low here and still matter. Nuclear is the clearest example: talk of UFOs and nukes is so common that it blends into generic chatter and barely beats the plain search — yet the cases physically cluster near nuclear-weapons sites, 4.56× closer than chance in our proximity study. This report leads with what's weak; nothing here is asserted as fact.
What stands out from generic UFO talk
Each bar shows how much a theme stands out from a plain "UFO" search. The line marked 1× is that plain search: bars to the left are no more distinctive than generic talk; bars to the right genuinely stand out. Only themes past 2× count as clear standouts.
Three famous threads, strong for three different reasons
Navy radar-visual encounters
The only theme that clearly beats generic UFO talk — 7,021 different people describe it, and no single source accounts for more than 2% of them. And unlike most of the archive, it was caught on instruments, not just eyes: the Nimitz Tic-Tac (2004) is one of the best-corroborated cases we have (told in 24 videos and 19 documents), with radar, cockpit video and infrared all agreeing — plus the Navy's own debrief forms (2019-2020) now in the record.
Pre-satellite transients (VASCO)
It barely stands out as a search term — but that misses the point, because this is the one line in the whole archive that is actual published science, not eyewitness testimony. Villarroel et al. (Scientific Reports, 2021): nine star-like dots appeared and vanished within about half an hour on a 1950 photographic plate — not there before, not there after, not in modern surveys. Honest status: one peer-reviewed paper is out; the follow-ups are still preprints, not yet finalised.
UFOs & nuclear weapons
As a search term, nuclear barely beats a plain "UFO" search — because talk of UFOs and nukes is everywhere, so it blends into the background. But the real evidence isn't in the words, it's in the locations: UFO cases physically cluster near nuclear-weapons sites, 4.56× closer than you'd expect by chance in our proximity study. 3,215 people describe it, backed by declassified government documents spanning 70+ years.
The stories hang together on their own
When people describe these experiences, certain details keep showing up together — far more often than you'd get by chance. Missing time comes with hypnotic regression; paralysis comes with a night-time bedroom visit. Whatever the cause, the accounts are internally consistent across witnesses who never met. The big number is how many times more often each pair appears together than chance alone would produce.
The details that keep coming up
A handful of details show up again and again, at steady rates, across people with no connection to each other. That's a consistent description — not a claim about what causes it. Each bar shows how often a detail appears; the whisker is our margin of uncertainty.
"If there's one consistent picture, it's a silent hovering disc that interferes with electronics, seen by several people at once, and sometimes photographed."
Not by hype
Published science
The VASCO 1950 photographic plate — a peer-reviewed paper with real data behind it, not testimony. The single strongest line in the archive, however it scores as a search term.
Caught on instruments
Nimitz (radar, cockpit video and infrared) and radar-confirmed intrusions over nuclear sites — Pantex (2015), Malmstrom (1967). Machines saw it, not just people.
A pattern that survives a fairness check
The nuclear result: cases sit near weapons sites 4.56× more often than chance — and it holds up when we reshuffle the data to make sure it isn't a fluke. A pattern, not a single anecdote.
Testimony plus a government document
An eyewitness account with a matching official record on the same event — the best an eyewitness-based case can do without instruments or science.
The cases the most separate sources describe
Leading with the weak
Directed-energy harm to witnesses
0.1×The weakest thread here — only 351 people, 47 cases, well below a plain UFO search. Often claimed, rarely backed up by anyone independent.
Miraculous healing & transformation
0.1×Barely shows up as its own thing (451 people), and mostly from a handful of the same researchers — indistinguishable from generic contact talk.
Human-hybrid breeding program
0.2×Thin, and shaped by its promoters: 535 people across 127 cases, with Budd Hopkins the dominant voice. Culturally loud, evidentially quiet.
Anti-gravity / exotic propulsion
0.2×Weak, and it leans on one person — Bob Lazar accounts for 6% of a small base. Famous in the culture, thin in the data.
Weak in the data isn't the same as false
One last thing, and it matters most. Everything in this report is built from what's public — recordings, interviews, declassified files, published papers, and leaks. That is a hard ceiling. We have no access to what governments still hold classified, to sealed or closed-door testimony, or to what a whistleblower has seen but cannot show.
So when a theme scores low here, read it carefully: it means that claim doesn't stand out in the public record — not that it is false, and not that it is impossible. A pattern can be entirely real and still be quiet in our data, simply because the evidence for it isn't out in the open. Absence of proof in a public archive is not proof of absence.
We measure the discourse, not reality. This is the clearest read we can give of what's out in the open — no more, and no less.