Do sighting reports track Starlink launches?

Pre-registered 2026-09-13 · readout no earlier than Mon Mar 15 2027 00:00:00 GMT+0000 (Coordinated Universal Time) · status: collecting

The question

Satellite trains are a known and frequently confirmed cause of misidentified sightings. How much of the reporting volume do they account for?

Hypothesis, stated in advance

Daily counts of newly opened sighting clusters rise in the 72 hours following a Starlink deployment, relative to matched control windows.

What would count as support

A positive and statistically significant association at the 72-hour lag after multiple-comparison correction across the tested lags.

Method

Daily counts of clusters opening in the UAP and Sky & Earth beats, regressed on a Starlink launch indicator at lags 0 to 5 days, using a negative binomial model with day-of-week and a smooth time trend as covariates. Lags are declared in advance; Benjamini-Hochberg correction applied across the six.

Data

thestaticwire.clusters
thespacedevs.launchlibrary2

Known confounds

Written down before the result, so they cannot be discovered conveniently late.

Launches cluster in time and correlate with other space-news coverage, which independently raises reporting volume.
Visibility depends on orbital geometry, local sunset time and weather. The naive indicator ignores all three, so a null result here is weak evidence of absence.
Reporting lag between a sighting and its publication is unmeasured and varies by outlet.

Result

Not yet. Data is still accumulating and the readout date has not been reached. Whatever it says when it arrives — including nothing at all — will be published here.

Data access

/api/v1/analyses returns this register as JSON, including every entry that found nothing.