Lossline observed fire benchmark

The change behind the count.

A week-over-week view built only from completed reporting periods. See which incident categories and observed states actually moved instead of reading a fresh total without context.

Latest completed weekSeptember 21, 2026UTC reporting period
Qualifying alerts107-81.2% week over week
Observed areas42active source areas

Completed weeks

One comparable timeline.

The current partial week is excluded so an unfinished period is never presented as a completed decline.

18
214
254
315
512
568
107

Category movement

Which signals changed most.

Comparing the weeks of September 14, 2026 and September 21, 2026.

Commercial Fire-82.8%122 → 21 alerts
Structure Fire-81.4%188 → 35 alerts
Residential Fire-80.5%257 → 50 alerts
Working Fireflat1 → 1 alerts

Observed state movement

Where the volume landed.

01Florida49 alerts · -83.9%
02New York15 alerts · -63.4%
03California10 alerts · -80.8%
04Delaware10 alerts · -75.0%
05Maryland7 alerts · -73.1%
06Arizona6 alerts · -78.6%
07Pennsylvania6 alerts · -87.5%
08New Jersey2 alerts · -80.0%
09Minnesota1 alerts · -80.0%

Benchmark limits

Direction, not a national fire rate.

The benchmark compares qualifying alerts observed in Lossline-monitored areas during complete UTC weeks. Changing monitored coverage can affect totals as the network expands.

It is not a complete U.S. fire census and does not measure damage, claim severity, owner interest, or closed restoration work.

Inspect every source report · Download the public dataset