The Verg Credibility Matrix (VCM)

Grade the claims. Examine the explanation.

A UFO case is not adequately described by a binary label. The Verg Credibility Matrix proposes evaluating how credible a case’s reported claims may be, alongside a separate assessment of how convincingly a conventional explanation accounts for the observations.

The ambition is to generate new comparative research data—not simply another collection of verdicts. This page records the working concept as it develops. The criteria, weighting, score labels, and review process have not been finalized. Four curator-approved provisional assessments form the original pilot, now joined by requested source reviews of Cobden and McMinnville; these are not a validated scoring standard.

Two independent 0–100% assessments

Independent, not complementary. The two assessments do not have to total 100%. A low score for an explanation would not automatically raise confidence in a case’s extraordinary interpretation. Where several explanations compete, each would need separate consideration.

Evaluative scores, not statistical probabilities. The percentage format expresses an assessment within a developing framework. It does not mean that a case is scientifically proven to have a particular probability of being true. Missing evidence is uncertainty—not automatically a zero.

Has anyone done this before?

Before developing the VCM further, we scoured the public internet, historical ufological literature, and recent academic research for any existing equivalent. The finding: no existing framework combines two independent percentage scales — one for the case claims and one for the conventional explanation — into a comparative matrix. The Verg Credibility Matrix is, to the best of our research, unique.

What does exist are partial precedents, and the VCM openly builds on their spirit:

• Hynek’s Strangeness–Reliability curve (1972): J. Allen Hynek plotted cases on two axes — how strange the report was and how reliable the witness — but both axes measured the case itself. He never scored the competing explanation.
• Classification systems (Hynek, Vallée): These sort encounters by type and evidential quality, but produce categories, not graded confidence in rival interpretations.
• A 2025 peer-reviewed matrix (Acta Astronautica): A recent academic proposal evaluates UAP reports against potential explanations — the closest relative found — but it is a single-axis plausibility screen, not a dual independent percentage assessment, and it is not in public use as a research tool.
• Debunker scorecards and believer verdicts: Countless sites rate cases “real” or “debunked.” All collapse the question into one verdict; none grades the explanation on its own scale.

The VCM’s contribution is the separation: the claims and the explanation each stand trial on their own axis, and the relationship between the two scores — not either score alone — is where the insight lives.

The matrix in practice: pilot and source reviews

The original four values were proposed through AI-assisted inquiry and approved by Micha Verg on November 1, 2026. Source-based Cobden and McMinnville reviews were requested on November 2, 2026; their new numerical estimates have not been separately approved by the curator. McMinnville’s review reflects the curator’s high-confidence view while retaining the disputed suspended-model alternative. All remain provisional editorial judgments, not validated calculations or probabilities of alien origin. Every other unreviewed case stays pending.

1. Socorro / Lonnie Zamora (1964)

   Provisional editorial assessment

   Credibility Score — Case Claims

   80%

   80%

   Credibility of the case as reported

   Explanation Offered — Credibility Score

   25%

   25%

   College prank / experimental balloon

   * Independent evaluative scores on separate 0–100% scales (do not sum to 100% because they are separate scores).

   Case-claim credibility is not proof of an anomalous craft or alien origin. The explanation score applies only to the named candidate.

   The reported observation, prompt investigation and described ground traces motivate the higher claim score. Prank and balloon proposals remain candidates, not established identifications. Verify original witness statements, trace records and each candidate's physical fit before revising this starting value.

2. Roswell Incident (1947)

   Provisional editorial assessment

   Credibility Score — Case Claims

   55%

   55%

   Credibility of the case as reported

   Explanation Offered — Credibility Score

   70%

   70%

   Project Mogul flight #4

   * Independent evaluative scores on separate 0–100% scales (do not sum to 100% because they are separate scores).

   Case-claim credibility is not proof of an anomalous craft or alien origin. The explanation score applies only to the named candidate.

   Contemporaneous debris reporting and later extraordinary testimony have different evidential weight. The Mogul hypothesis provides a conventional candidate for the debris, but the specific flight attribution and retrospective claims require separate source checks. This score does not validate crash-retrieval or alien-body claims.

3. Rendlesham Forest (1980)

   Provisional editorial assessment

   Credibility Score — Case Claims

   75%

   75%

   Credibility of the case as reported

   Explanation Offered — Credibility Score

   35%

   35%

   Orford Ness lighthouse beacon

   * Independent evaluative scores on separate 0–100% scales (do not sum to 100% because they are separate scores).

   Case-claim credibility is not proof of an anomalous craft or alien origin. The explanation score applies only to the named candidate.

   The reported military accounts and contemporary memorandum motivate the claim score, without endorsing every later detail. The lighthouse is a candidate for some reported lights, not an established explanation of the complete narrative. Compare sightlines, timings, the recording and independently documented observations.

4. Tulli Papyrus (Attributed c. 1440 BCE)

   Provisional editorial assessment

   Credibility Score — Case Claims

   30%

   30%

   Credibility of the case as reported

   Explanation Offered — Credibility Score

   60%

   60%

   Metaphorical celestial omen / modern transcription artifact

   * Independent evaluative scores on separate 0–100% scales (do not sum to 100% because they are separate scores).

   Case-claim credibility is not proof of an anomalous craft or alien origin. The explanation score applies only to the named candidate.

   The absence of an authenticated original limits confidence that the alleged ancient event occurred. Literary interpretation and modern textual fabrication or transmission error are distinct candidates grouped only for this pilot. Neither the attributed ancient date nor this score authenticates the papyrus; separating those candidates is a priority for a later review.

5. Cobden UFO Sighting (1931)

   Provisional editorial assessment

   Credibility Score — Case Claims

   40%

   40%

   Credibility of the case as reported

   Explanation Offered — Credibility Score

   45%

   45%

   Aircraft with forward illumination and flashing lights

   * Independent evaluative scores on separate 0–100% scales (do not sum to 100% because they are separate scores).

   Case-claim credibility is not proof of an anomalous craft or alien origin. The explanation score applies only to the named candidate.

   Claim assessed: an unusual lighted object was reported near Cobden, not that it was an alien craft. NICAP's 1964 chronology preserves a dated summary, which gives the report a traceable archival basis. However, this review has not located a 1931 originating record, independently preserved witness statements, duration, measured trajectory, weather, radar, or an authenticated photograph. Those gaps keep confidence limited. An aircraft is a plausible candidate for the forward light, rear flashing lights and curved departure; apparent acceleration or climb cannot establish performance without distance and viewing geometry. No identified flight or reconstruction verifies that candidate, so its fit remains tentative. The AI recreation contributes no evidential weight.

6. McMinnville / Trent Photographs (1950)

   Provisional editorial assessment

   Credibility Score — Case Claims

   80%

   80%

   Credibility of the case as reported

   Explanation Offered — Credibility Score

   45%

   45%

   Small model suspended beneath an overhead wire

   * Independent evaluative scores on separate 0–100% scales (do not sum to 100% because they are separate scores).

   Case-claim credibility is not proof of an anomalous craft or alien origin. The explanation score applies only to the named candidate.

   Claim assessed: the Trents credibly reported an unusual disc and supplied two genuine in-camera photographs, not that an alien craft is established. The early newspaper record, two related film images, Hartmann’s finding of no negative tampering and supportive investigations justify comparatively high editorial confidence, reflecting Micha Verg’s stated view. Maccabee reported no payment for the photographs; Hartmann found prior motivation difficult to identify but noted a subsequent New York trip. Motive is contextual, not proof. The two witnesses belong to one household; the reported evening time, distance and motion are not independently measured. Hartmann explicitly retained a suspended model, and later geometry, lighting and claimed thread-detection analyses remain disputed. That candidate has a substantive but unconfirmed fit, reflected in its independent 45% estimate. Neither genuine negatives nor these scores establish extraordinary flight or extraterrestrial origin.

Upper-left positions indicate higher claim confidence and weaker fit for the named candidate—not proof of an anomaly or of alien origin. Upper-right positions can indicate credible reporting and a plausible conventional account. Lower positions reflect limited confidence in the claims. Scores compare only the specified claim and candidate, not every possible explanation.

What the next review needs

For each pilot case, identify the precise claim, link the earliest available records, separate contemporary evidence from later retellings, and record which observations support or contradict the named explanation. Where the pilot groups alternatives, assess them separately. Record any revised scores, reasons and review date; missing documents stay explicit uncertainties rather than invented evidence.

Questions shaping the method

• Source quality: Can the original record be traced, dated, and checked against later retellings?
• Testimony and corroboration: Which observations are independently supported, and which depend on a single account?
• Evidence integrity: What is known about the provenance and limitations of photographs, recordings, sensor records, and physical material?
• Explanatory fit: Does a proposed explanation address the full observation, or only selected details?
• Revision: What new documentation would change an assessment, and how should that change be recorded?

These are development questions, not an approved checklist or weighted formula. A consistent method must be established before numerical comparisons can be meaningful.

Comparison without forced conclusions

Abduction and experiencer research has a separate destination: The Verg Abduction Matrix (VAM). Future case records can carry VCM, VAM or both, applied to distinct claims; unreviewed assessments remain pending.

A two-axis view could distinguish cases with credible claims and a strong conventional account from cases with credible claims but an inadequate explanation. Cases with weak documentation may remain inconclusive regardless of how attractive either interpretation sounds.

The linked matrix presents the original four approved provisional records and the requested Cobden and McMinnville source reviews. Community scoring, automated scoring and a validated comparison method are not available; future revisions will need explicit source review and accountable judgment.

Human curiosity. AI-assisted inquiry.

The VCM grows out of Micha Verg’s long-standing approach to case evaluation and the possibilities of developing it through human–AI collaboration. AI-assisted organization and comparison can support inquiry; they cannot replace source verification, accountable judgment, or the distinction between an observation and its interpretation.

The objective is a transparent, revisable research framework that helps people ask better questions and discover relationships across the archive. Its methodology will be refined as the project develops.