Strangeness Index & IFO Filter (2025-01-01 to 2026-03-10)

The analysis is based on a global dataset of 3,680 UAP sightings, sourced primarily (99.8%) from the National UFO Reporting Center (NUFORC). The date range is January 1, 2025, to March 10, 2026. Each sighting was assigned a Strangeness Index score (0-100), a systematic metric evaluating the anomaly of the report after accounting for prosaic explanations. Sightings scoring 60 or higher were classified as 'High Strangeness,' forming the core subject of this filter analysis. The analytical method involved comparative statistics between the full dataset (N=3680) and the high-strangeness subset (N=259). Key dimensions analyzed included: shape distribution, temporal patterns (hour-of-day), geographic distribution (top locations and coordinate clusters), and witness metrics (corroboration rate, average witnesses). Conventional object indicators (IFO signatures) such as near-airport linear flights, short-duration single-witness events, and Starlink formation patterns were also assessed to establish a baseline of identifiable phenomena.

The Strangeness Index score distribution for the 3,680 sightings is heavily skewed toward lower scores. The majority, 92.2% (N=3394), fall in the 'Low' category (20-39). Only 7.0% (N=259) achieve a 'High' score (60-79), with zero in the 'Very High' (80-100) category. The overall average score is 37. The high-strangeness subset comprises 259 sightings. Within this subset, the shape distribution is: Orb (N=138, 53.3%), Other (N=63, 24.3%), Disc (N=21, 8.1%), Cigar (N=16, 6.2%), Triangle (N=14, 5.4%), Fireball (N=4, 1.5%), and Chevron (N=3, 1.2%). Temporally, high-strangeness sightings peak during dusk hours (5pm-10pm), accounting for 70.7% (N=183) of cases. This contrasts with the full dataset, where dusk sightings represent 47.3% (N=1740). Night (10pm-5am) accounts for 18.1% (N=47) of high-strangeness cases, day (9am-5pm) for 6.9% (N=18), and dawn (5am-9am) for 4.2% (N=11). Geographically, the top locations for high-strangeness reports are Florida (N=31), Arizona (N=29), California (N=24), Colorado (N=20), and Texas (N=15). Witness data shows a low corroboration rate across the entire dataset (6.7%, N=248 events with multiple witnesses), with an average of 1 witness per report. For the high-strangeness subset, specific corroboration data is not provided, but the average witness metric is consistent with the overall dataset. IFO indicator analysis found minimal signatures of conventional objects: one short-duration single-witness event, and zero near-airport linear flights or Starlink pattern matches.

The application of the Strangeness Index filter isolates a phenomenologically distinct subset of reports. The high-strangeness residual is not a proportional microcosm of the full dataset but exhibits specific skews. Most notably, the 'Orb' shape, while dominant in the full dataset (49.3%), becomes even more predominant in the high-strangeness subset (53.3%). Conversely, 'Triangle' sightings, which constitute 8.5% of all reports, drop to 5.4% of high-strangeness cases, suggesting many triangular reports may be attributable to conventional aircraft formations or other IFOs. The extreme temporal concentration at dusk (70.7%) is statistically significant and requires explanation. This could indicate an observational bias (increased sky-watching during evening hours) or a potential characteristic of the phenomena themselves, such as luminosity being more visible in specific lighting conditions. The geographic clustering in Florida, Arizona, and Colorado is notable, as these states rise in relative rank within the high-strangeness subset compared to the full dataset (where California, Florida, Texas are top). This may suggest regional factors—whether environmental, demographic, or related to reporting culture—that correlate with higher-strangeness reports. The near-absence of IFO signatures (e.g., Starlink, airport patterns) in the high-strangeness subset indicates the filter is effectively screening out these known conventional explanations. However, the overwhelming prevalence of 'Unknown' for weather (100% of high-strangeness cases) and duration (98.9% of all cases) represents a major data quality limitation, preventing analysis of correlations with atmospheric conditions or event length.

This analysis demonstrates that a small fraction (7%) of UAP reports from a 14-month global dataset resist conventional explanation after applying a systematic Strangeness Index filter. The resulting high-strangeness profile is characterized by simple, luminous 'Orb' shapes observed primarily in the early evening hours, with notable geographic clusters in the southwestern and southeastern United States. Confidence in these specific morphological and temporal patterns is rated as Medium, based on the moderate sample size (N=259) of the filtered subset and the clear statistical deviations from the full dataset. Primary limitations include severe gaps in foundational data fields (weather, duration, altitude, movement patterns), which constrain deeper environmental and kinematic analysis. The low corroboration rate (6.7%) also limits the ability to verify events independently. It is recommended that future data collection efforts mandate basic environmental and kinematic fields to improve analytical rigor. Further research should focus on the specific dusk-time observational cluster and the geographic hotspots identified, employing targeted sensor deployment or analysis of existing atmospheric data in those regions to search for correlated physical signatures.