What 10,285 Case Files Look Like in Aggregate

This piece is not history. It is a description of what we found by machine-reading the National Archives catalog metadata for this collection, and it comes with its own caveats attached.

We parsed the title of every file unit in the case-file series. Those titles follow a convention — place, then date, sometimes an incident number — which makes them machine-readable in bulk: "Hamburg, New York, June 1947". From 10,743 archival descriptions we resolved 10,276 sightings, and recovered a location for 96.3% of them.

We restrict that count to records that actually sit in a case-file series. A trailing date alone is not enough: "Ford Press Releases, January - April 1966" parses perfectly and is not a sighting. Nine such records were excluded once we required membership of a case-file series as well as a parseable title.

The timeline, and a correction worth making

The distribution is not smooth, but it does not have the shape most accounts of this subject imply. Three findings, none of which match the popular summary:

The heaviest single month is November 1957, not August 1952. November 1957 accounts for 364 case files — almost three times August 1952's 128, and the largest concentration anywhere in twenty-two years. It is not close.

1957 is the heaviest year: 970 files. The famous summer of 1952 places sixth, with 590. The top five years are 1957, 1966 (866), 1965 (726), and 1956 and 1967 tied at 614.

The 1960s produced almost as much paper as the 1950s — 4,890 files against 4,907. The idea that this is a 1950s phenomenon that faded is not supported by the filing record; volume held up until the project closed.

We are describing the files, not the sky. A concentration like November 1957 could reflect a genuine wave of sightings, a burst of press coverage that prompted people to report, a change in how the office logged reports, or some combination. The metadata cannot distinguish those, and we are not going to pretend otherwise. What it does establish is where the paper is — and the paper is not where the familiar story says it is.

The map, and the problem with it

Ranked by state, the top of the table looks like this:

California, New York, Texas and Florida are unremarkable — they are populous states with a lot of sky and a lot of aircraft. Ohio is the anomaly. It leads the entire country, and by city the single most-represented place in the archive is Dayton, Ohio.

Here is the caution, and it matters more than the ranking: Project Blue Book was headquartered at Wright-Patterson Air Force Base, which is in Dayton. A filing system located in Dayton, staffed in Dayton, generating correspondence in Dayton, will accumulate Dayton in its metadata for reasons that have nothing to do with what was in the sky over Ohio.

We have not established that this explains the Ohio figure, and we are not asserting it. But any reader who takes the state ranking as a map of sightings rather than a map of files is likely to be misled, and the correlation between the top city and the project's own address is too tidy to ignore.

What the metadata will not tell you

Three limits worth stating:

• Roughly 4% of case files have no recoverable location. Their titles read

[illegible] because the typewritten originals had already degraded when they were transcribed. That is a property of 1950s paper, not of the sightings.

• About 480 sightings happened over open ocean and are recorded only as

latitude and longitude, in formats like 3030N 15835E. They are real records with no state and no city, and they vanish from any analysis that groups by state.

• A file is not an event. Some incidents generated several files; some files

cover several reports. Counting files is the only thing the metadata permits, and it is not quite the same as counting sightings.

The honest summary

The aggregate view is good for two things: seeing when the government's attention spiked, and finding your way to individual documents. It is bad for inferring where unusual things happen. The strongest signal in the geography is plausibly the location of a filing cabinet in Ohio.