How long does confirmation take when it comes at all?

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

For the stories that do get corroborated, what does the delay distribution look like?

Hypothesis, stated in advance

The distribution is heavily right-skewed: most corroboration arrives within 48 hours or never, with a thin tail of stories confirmed weeks later.

What would count as support

Median time to corroboration under 72 hours; 90th percentile beyond two weeks.

Method

Kaplan-Meier survival estimate over clusters opening at tier 3 or 4, with the event being first corroboration and censoring at the readout date. Reported separately by opening tier and by beat.

Data

thestaticwire.clusters
thestaticwire.cluster_events

Known confounds

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

Our polling interval is ten minutes, so all timings are quantised to that and biased slightly late.
Publisher-stated publish times are unreliable and frequently backdated, which is why the clock runs on our own fetch time instead.

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.

Amendments

Changes made after registration. Appended, never edited over.

2026-09-13 — Inherits the recognised-publisher requirement from the corroboration-rate amendment of 2026-09-13; the event being modelled is now first corroboration by a recognised tier 1-2 publisher.

Data access

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