Methodology · consolidated v0.5
We rate the evidence, not the phenomenon.
Every entry in the Disclosure Register is scored by a published formula applied to stated facts — not by anyone's opinion. This page is that formula. If you think an entry is mis-scored, this is the page that tells you exactly which fact to challenge.
What we measure
Two questions, kept completely separate.
Almost every source on this topic blends "how good is the evidence" with "how strange is the object" into a single impression. That blend is what produces noise. We split them into two independent axes and score each on its own.
Evidence Strength asks how solid the record is — who released it, how it's corroborated, whether you can trace it to a primary document.
Anomaly Level asks only whether officials have explained it — nothing about what we think the object is.
A weather balloon, fully documented in a Pentagon report, scores high on Evidence and low on Anomaly. That is the system working exactly as designed.
Axis one
Evidence Strength — how solid is the record?
Five factors, each scored against fixed criteria, then weighted into a 0–100 score. No factor is a judgment call about credibility; each is a checkable property of the release itself.
Heaviest weight goes to who put it on the record, because for an officially-disclosed catalog that's the strongest single signal.
Axis two
Anomaly Level — has it been explained?
Three factors, equally weighted. This axis reads only what officials said about resolution — it never characterizes the object itself.
Each factor reflects the official record's own words, not our reading of how strange something looks.
The one rule we never bend
A high Anomaly Level means officials have not explained it — it does not mean the object is extraterrestrial. "Unexplained" is a statement about the absence of an official explanation, nothing more. We never assert cause.
What gets in
Officially disclosed only — and once in, never removed.
An entry is eligible only if it traces to an official government release: a report, a declassified document, sworn testimony, a formal acknowledgement, or an agency data release. If it can't be traced to a primary source, it isn't here.
Two decisions we handle in opposite ways. Inclusion is total and mechanical — everything officially released is logged, with no quality gate, because what gets left out shapes the picture more than how we score what's in. Prominence is ranked by the rubric — stronger cases surface first, thinner ones sit deeper, but nothing is ever hidden. The test we hold ourselves to: you can find every single thing the government released by searching this catalog. Ranking is a spotlight, never an eraser.
The principles we hold
Why you can audit us instead of trusting us.
01
No human assigns a score
Editors record checkable facts from the primary release. A published formula turns those into the score. Nobody types a number.
02
Never omit, only rank
Everything disclosed is logged and findable. Prominence is decided by the rubric, never by our taste for what's interesting.
03
"Unexplained" ≠ "extraterrestrial"
We rate how solid the record is and whether officials resolved it. We never assert what something is.
04
Every score shows its work
Each entry opens to the factor-by-factor breakdown, so you argue with an input, not a verdict.
05
Versioned and logged
The method is dated and versioned. When it changes, affected entries are re-scored and the change is recorded in the open.
06
We don't claim zero bias
Any instrument reflects its builder's choices. We claim something checkable instead: no per-entry influence, and every upstream choice made public.
A note on the moment
We're building the instrument that finds the baseline — not applying one.
Sustained, official, public UAP disclosure at this scale has not happened before. There is no established record of what "normal" looks like, because the pattern is forming in real time as we measure it.
So our parameters are provisional defaults, openly flagged and revised as the record grows — not settled findings dressed up as certainty. We think that's the more credible posture for a project like this: a careful observer of something unfolding, showing its work as it goes.
Current version: consolidated v0.5 · June 2026 · changes are logged and dated.