Anomaly Evidence Triangulation
Judge unusual claims across independent observations, not one dramatic case
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 94%
When Steven asks whether the Tic Tac incident could have another explanation, Dan argues that any isolated event permits many alternatives and that the stronger case comes from the collective: recurring reports, different observers, multiple collection systems, and observations under varied conditions. Hal makes a similar point about repeated underwater reports. The reusable method is to define the anomaly neutrally, inventory genuinely independent evidence streams, test mundane explanations, and compare patterns across conditions. Its key safeguard is to keep observation and cause separate. Multiple reports may increase confidence that something unusual occurred, but they do not by themselves establish who or what caused it. Where source data remain classified or unavailable, the conclusion must remain correspondingly limited and attributed.
Origin
Extracted from The Diary of a CEO
Core principles
- 01An isolated anomaly can support many explanations
- 02Independent sensors and observers reduce some forms of error
- 03Repeated observations across different conditions matter more than repetition of one story
- 04A residual anomaly is not automatic proof of a preferred explanation
- 05Confidence should track the evidence that can actually be inspected
How to run it
- 1
Describe before explaining
Write down what was observed in neutral terms, including time, location, motion, and instrument readings. Avoid embedding the suspected cause in the description.
Pro tip Replace labels with measurable features wherever the record allows.
Watch out A dramatic label can make later analysis circular.
- 2
Map independence
List each observer, sensor, document, and data path, then identify which ones could share the same source or error.
Pro tip Treat two retellings of one report as one evidence stream.
Watch out Apparent corroboration can collapse when every account traces back to one source.
- 3
Test ordinary causes
Check instrument artifacts, perception errors, known technology, environmental effects, and incomplete context before escalating the explanation.
Pro tip Record which explanation each evidence stream supports or contradicts.
Watch out Failure to identify an ordinary cause does not prove an extraordinary one.
- 4
Compare across conditions
Look for the same observable pattern across different people, instruments, environments, and times.
Pro tip Give more weight to convergence that survives materially different conditions.
Watch out Repeated institutional procedure can reproduce the same error across cases.
- 5
Calibrate the conclusion
State separately what likely happened, what remains unexplained, and which proposed causes are still speculative.
Pro tip Name the next inspectable evidence that would raise or lower confidence.
Watch out Do not turn unexplained into confirmed.
In the wild
Dan cites Commander David Fravor's reported visual observation together with multiple collection systems and then places that incident alongside other military reports. Steven continues to ask whether an isolated unusual event could have another explanation.
→ The discussion illustrates why converging records may strengthen confidence that an anomaly occurred while leaving the proposed non-human cause unproven in the public evidence discussed.
Common mistakes
Counting echoes as corroboration
Many accounts do not create independent support when they all repeat the same original report.
Equating unexplained with confirmed
Ruling out several explanations does not establish whichever explanation remains most interesting.
Ignoring inaccessible evidence
Claims about classified records may be relevant testimony, but confidence must reflect that outsiders cannot inspect those records.
Is it for you?
Best for
It is best for safety incidents, intelligence questions, scientific anomalies, and other cases with multiple partial evidence streams.
Not ideal for
It is not ideal when the observations share one unrecognized source of error or when the underlying records are unavailable for independent review.
From the transcript
“Any any isolated event like that you could do the whole could it be this, could it be that thing”
“you got to take a step back and look at this the collective”
“it's seen enough times under enough different conditions that we just have to accept it as real”
From the episode
UFO Roundtable: CIA Physicist Proves Aliens Exist