Witness Narrative Mining in United States (2015-01-01 to 2026-03-10) — Triangle

This qualitative research employs narrative analysis on a dataset of 3,466 triangulated UAP sightings within the contiguous United States, spanning from January 1, 2015, to March 10, 2026. The dataset was filtered to include only reports where the primary shape descriptor was 'Triangle.' A subset of 500 witness narratives, comprising the actual textual accounts, served as the primary corpus for linguistic and thematic mining. The average narrative length was 13 words. Analytical methods involved term-frequency analysis to identify the most common descriptors (e.g., 'lights,' 'craft,' 'triangular') and qualitative coding of narrative excerpts to extract recurring behavioral motifs (e.g., 'silent,' 'hovering'). Patterns were cross-referenced with quantitative metadata, including temporal distribution (hourOfDayBreakdown), duration (durationBreakdown), geographic clustering (locationPoints, topLocations), and corroboration metrics (corroborationRate, averageWitnesses). The study's scope is explicitly limited to witness-reported characteristics and does not attempt prosaic identification.

Quantitative analysis reveals distinct patterns. Temporally, sightings are heavily skewed toward low-light conditions: 1,524 (44.0%) occurred at dusk (5 PM–10 PM) and 1,389 (40.1%) at night (10 PM–5 AM), collectively accounting for 84.1% of all reports. The duration was most commonly brief, with 1,869 sightings (53.9%) lasting 1–5 minutes. Geographically, reports are distributed nationwide, with the highest frequency in California (n=338), Florida (n=238), and Texas (n=156). Spatial analysis of location points shows clustering in metropolitan areas, including Phoenix, AZ (lat 33.5, lng -112; n=46) and Los Angeles, CA (lat 34, lng -118; n=37). Corroboration data indicates a low rate of independent witness confirmation. The average number of witnesses per event is 1.8, and only 327 of the 3,466 events (9.4%) are listed as corroborated. This corroboration rate is below the global dataset average of 11.37%. The source is overwhelmingly the National UFO Reporting Center (NUFORC), accounting for 3,465 of 3,466 reports. Term-frequency analysis of narratives shows the most common descriptors are 'triangle' (n=217), 'lights' (n=203), 'craft' (n=139), 'triangular' (n=76), and 'black' (n=65). Key behavioral terms include 'moving' (n=59), 'hovering' (n=36), and 'silent' (n=28).

Narrative mining identifies a highly consistent and specific phenomenological description. Witnesses independently and repeatedly report a configuration of three luminous points arranged in a triangular formation. A notable shared detail is the mention of an additional, often central, light. For example, one narrative states: 'Black triangle with four white pale backlit lights. There was an extra light between two of the corner lights.' This suggests a non-random, structured arrangement of light sources that is consistently perceived across disparate observations. The language patterns strongly associate the triangle shape with descriptors of anomalous performance. Narratives frequently couple the shape with phrases denoting extreme kinematics ('traveling inhumanly fast,' 'instantaneous acceleration') and the absence of sound ('silent'). The phrase 'triangle formation' appears explicitly, indicating witnesses often perceive discrete light sources moving in a coordinated, rigid geometry rather than a single solid object, though some narratives describe a 'giant dark object' or 'craft.' The low average witness count (1.8) and corroboration rate (9.4%) are statistically significant. They indicate that most triangular UAP reports are single-witness events, which presents a challenge for verification and increases the potential influence of misperception. However, the high degree of narrative consistency across thousands of such single-witness reports from different geographic regions and years suggests a common stimulus, whether prosaic or anomalous. The concentration of reports in dusk and night hours is consistent with the increased visibility of luminous phenomena against a dark sky but may also influence perception of shape and motion.

This analysis finds a high degree of consistency in the witness-reported characteristics of triangular UAPs across the United States over an 11-year period. The specific, recurring description of a triangular light formation—often with a central light—coupled with reports of silent operation and anomalous kinematics, constitutes a stable phenomenological signature. Confidence in the descriptive patterns is High, given the large sample size (N=3466) and the independent replication of specific details across the narrative corpus. Primary limitations include the dataset's reliance on self-reported, unverified civilian accounts, the lack of sensor data, and the low corroboration rate. The over-representation of NUFORC as a source may introduce reporting biases. The narrative analysis is also constrained by the brevity of the average report (13 words). Recommendations for further research are: 1) Cross-reference these witness-derived behavioral signatures with radar or satellite data from corresponding times and locations where available. 2) Conduct a comparative narrative analysis with known prosaic explanations for triangular light formations (e.g., formation flying, B-2 Spirit sightings) to identify discriminant linguistic features. 3) Initiate a structured witness interview protocol to obtain more detailed technical descriptions from recent high-quality reports to supplement the brief initial narratives analyzed here.