UFO Waves and Flaps: Why Sightings Come in Clusters

The Sky Doesn't Cluster. The Reports Do.

If UFOs were a purely physical phenomenon showing up at random, sighting reports would spread out roughly evenly across time and geography, the way, say, lightning strikes do. They don't. They come in bursts — sudden, sharp, short-lived explosions of reporting activity concentrated in one region over weeks or months, followed by long stretches of comparative quiet. Ufology calls these bursts "waves," or, in the field's older and blunter slang, "flaps." Our own data-mining of the 138,296-report database has surfaced 31 distinct geographic-temporal flap clusters — tight concentrations of dozens of reports crammed into a single month or two, in a single small region — and studying why they cluster turns out to be one of the most genuinely useful exercises in the entire field, regardless of which side of the debate you sit on.

The Mechanics of a Flap

A flap is, at minimum, self-amplifying by construction. One striking, well-publicized sighting draws local media coverage. Coverage sends more people outside looking up, at exactly the moment awareness and expectation are both freshly primed. Some of those people see something genuinely anomalous; many more see an ordinary light, plane, or planet through the lens of a story they just read, and misidentify it with total sincerity. Each additional report feeds the next news cycle, which sends more people outside, which produces more reports — a feedback loop that can run for weeks before it exhausts itself. This dynamic is not in serious dispute among researchers on either side of the ETH debate; where they disagree is whether the loop is amplifying real underlying activity, manufacturing an illusion out of noise, or both at once in different proportions for different waves.

Our own flap-detection data shows this pattern playing out with remarkable geographic consistency: the largest clusters we've identified recur in the same handful of locations across multiple separate years — a corridor of Illinois reports spiking repeatedly between 2000 and 2005, an Arizona cluster reappearing in 1997, 2000, 2004, 2015, and 2017, and steady recurring bursts around Washington state, South Carolina, and the New York City region across nearly two decades. That kind of repeat clustering in the same places, year after year, is itself a data point worth sitting with — it's consistent with a "flap-prone" population effect (some regions are simply more primed to notice and report, thanks to local media habits, terrain, or flight paths) as much as it is with anything about the sky itself.

1954: Proof a Flap Doesn't Need American Media to Happen

Not every major flap fits the media-contagion story as neatly as the American examples below, and the exception is instructive. France's autumn of 1954 produced the densest concentration of close-encounter reports ever recorded — hundreds of landed-craft and occupant sightings in just ten weeks, overwhelmingly from isolated rural witnesses: farmers, gendarmes, and villagers with limited access to the kind of saturation press coverage that drove American waves. Aimé Michel's "orthoteny" theory — that sighting locations aligned along straight geographic lines — was later dissolved by statisticians who showed random points align just as easily, but the wave's underlying geographic spread across a media landscape far thinner than 1940s-70s America remains the strongest single data point for researchers who argue flaps aren't purely a manufactured media artifact. It's also the wave that built Jacques Vallée's entire career and produced the term "orthoteny," a reminder that even a debunked theory can leave permanent, useful vocabulary behind.

1947: The Wave That Invented the Vocabulary

The founding flap of the modern era compressed an entire cultural phenomenon into six weeks. Kenneth Arnold's June 24 sighting near Mount Rainier — nine crescent objects moving, in his own description, "like a saucer skipped across water" — was mangled by wire-service reporting into "flying saucer," and within days the country was seeing them everywhere: researcher Ted Bloecher's later count found roughly 850 press-documented sightings nationwide in that short window, against just 122 the Air Force logged as worth investigating. The wave's second act arrived on July 8, when Roswell Army Air Field announced, then retracted within hours, the recovery of a "flying disc" — an event that would lie dormant for three decades before becoming the most famous UFO case in history. The entire vocabulary of modern ufology — the shape, the term, the press dynamic — was assembled in this single flap.

1952: The Flap That Reshaped Government Policy

Project Blue Book's own numbers make 1952 the biggest officially logged flap in American history: 1,501 reports for the year, 717 of them crammed into June through August alone. Its centerpiece, the Washington, D.C. radar-visual events of two consecutive July weekends, briefly out-headlined the Korean War and triggered the largest Pentagon press conference since World War II. The flap's true legacy, though, is bureaucratic rather than evidential: it directly produced the CIA-convened Robertson Panel of January 1953, whose recommendation to actively debunk UFO reports as a security nuisance shaped official U.S. posture toward the subject for the next 65 years — a single flap's political fallout outliving the flap itself by nearly seven decades.

1973: The Flap With No Official Home

By 1973, Blue Book had been dead for four years, and the biggest post-war American flap had nowhere official to land — reports poured instead into police switchboards, local press, and the civilian networks that filled the vacuum, chiefly MUFON, founded in 1969 and coming of age in this exact moment. October 1973 alone delivered Pascagoula's robotic-entity abduction and the Coyne helicopter near-collision within a single week, against a backdrop of Watergate and the oil crisis — arguably the high-water mark of American institutional distrust in the twentieth century, a psychological climate several researchers consider inseparable from the flap's intensity and its unusually high rate of occupant reports.

1989-90: When an Air Force Admitted It Was Stumped

The Belgian triangle wave stands apart from every other flap on this list for one reason: institutional cooperation instead of institutional silence. Across roughly 2,000 collected witness reports, 143 of them on the very first night, a national air force ultimately scrambled jets, briefly locked radar onto accelerating targets, and then held a public press conference admitting it could not explain what happened — a level of candor no government has matched before or since. The wave arrived in the same extraordinary autumn as the Voronezh, USSR giant-humanoid sighting, amid the fall of the Berlin Wall and a sudden burst of glasnost-era Soviet media openness — a reminder that flaps often crest during periods when the broader information environment is already unusually unsettled.

2024: The Same Mechanics, Filmed on Every Phone

More than three decades later, the New Jersey drone flap proved the underlying mechanics haven't changed at all — only the recording equipment has. Beginning in November 2024, more than 5,000 reports of mysterious "drones" flooded the FBI in six weeks, dragging Congress into demanding answers and pushing nightly news coverage that only accelerated the reporting rate further. The eventual federal attribution — authorized commercial drones, hobbyist aircraft, conventional planes, and ordinary stars, misread under a genuinely primed and anxious public — is close to a controlled experiment in flap dynamics, run in real time with smartphone documentation of nearly every stage: FAA-trackable aircraft became, within days of media saturation, "car-sized drones" on national television. It is the same 1947 loop, updated for an era when everyone carries a camera and every report reaches a national audience within minutes rather than a next-day newspaper cycle.

How We Actually Find a Flap in the Data

It's worth explaining, briefly, how the 31 flaps in our own database get identified in the first place, because the method matters for how much weight to put on the result. Rather than relying on anyone's subjective sense that "a lot happened that month," our clustering approach groups reports by geographic proximity (using geohash grid cells) and time window, then scores each cluster by a combination of raw case count and how sharply it spikes above that region's normal background reporting rate — a "peak score" that separates a genuine short-term surge from a location that's simply a consistently busy reporting area year-round. The largest flap our system has isolated, a 2004 cluster centered on Illinois, contains 77 individual reports compressed into a single peak month, dominated by Light and Triangle shapes — a scale of concentration that, whatever its ultimate cause, is statistically real and not an artifact of anyone's storytelling after the fact.

Do Flaps Prove Anything, Either Way?

The self-amplifying mechanism behind flaps is real and well-documented, and it genuinely does account for a large share of any given wave's report volume — that much is not seriously contested by researchers on either side. But treating "flaps are partly self-amplifying" as equivalent to "flaps are entirely manufactured by media contagion" skips over a harder question the skeptical case has never fully closed: why do the flaps that receive the deepest institutional scrutiny — Belgium's radar-confirmed triangles, the Nimitz encounter that opened the modern UAP era — tend to leave behind a stubborn, unexplained residue even after every mundane candidate in the reporting surge has been stripped away? Flaps are, at minimum, real sociological events, reliably triggered, reliably self-amplifying, and reliably clustered in specific places and eras that reward close study on their own terms. Whether they are also evidence of something physical clustering in the sky above those same places, at those same moments, remains the one question thirty-one flaps and eight decades of data still haven't settled.