911Sentinel

Research · 2 min read

Crime Data Is a Process That Runs Late

Recent occurrence-date totals are structurally incomplete, offense-specific, and lack any published revision history.

Crime data looks like a census. It is a process, and it runs late. In three years of Seattle property records the median offense took 32.7 hours to surface, motor-vehicle theft appeared in under a day while other larceny took almost three, and the source keeps no usable revision history — so we built our own.

Our analysis of 116,754 valid intervals, reports published july 1, 2023–june 30, 2026.

How quickly do Seattle property-offense records actually appear?

Among 116,754 valid Seattle property-offense intervals from 2023 through 2026, the median offense-to-report lag was 32.7 hours, only 44.2% appeared within a day, and 16% took more than a week. Motor-vehicle theft surfaced in 16.7 hours; all other larceny took 65.8.

At a glance

The numbers behind the answer

Selected measures only. Denominators and interpretation stay attached so the headline cannot stand alone.

32.7 hours

Median offense-to-report lag

Across 116,754 valid nonnegative intervals in 2023–2026.

44.2%

Published within 24 hours

Slightly fewer than half of records appeared within a day.

16.7h vs 65.8h

Fastest vs slowest offense

Motor-vehicle theft versus all other larceny median lag.

01The distribution

A median of a day and a half, with a long tail

Half of property-offense records surface within 32.7 hours of the recorded start; 44.2% within a day. But 16% take more than a week, the 90th percentile reaches 286 hours, and the 95th reaches 590. A weekly dashboard that treats recent days as final is reading an unfinished page.

The gaps are not uniform. Motor-vehicle theft has a median lag of 16.7 hours, while all-other-larceny runs to 65.8 hours — different events get discovered and reported on very different clocks.

Evidence visual

Median offense-to-report lag by offense type

Hours from recorded offense start to publication, 2023–2026.

Motor vehicle theft16.7 h
Burglary / breaking & entering32.4 h
Theft from motor vehicle40.3 h
All other larceny65.8 h
02Survival view

The clock is offense-specific

Treating the interval as a survival process, offense type predicts how quickly a record appears: relative to the baseline category, motor-vehicle theft shows a hazard ratio near 1.9 and shoplifting near 1.6 — they surface sooner. The concordance of the model is modest (0.57), which is itself the point: lag is noisy, but offense type clearly matters.

Use it as

A freshness policy: label recent periods provisional and never compare them with settled history.

03Provenance

The source keeps no revision history, so we keep one

SPD's crime dataset exposes no row-level revision history — the revisions endpoint returns 404. That means the only way to know how a published number changed is to snapshot the dataset yourself and diff it. We began that log in this program; it is the kind of record an open dataset does not publish about itself.

One caution throughout: this interval is not emergency-call, dispatch, arrival, or police response time. It is the gap between when an offense was recorded as starting and when its report was published.

  • Show a data-as-of label on every map.
  • Mark recent periods provisional.
  • Never compare immature current periods to settled history.
  • Keep offense-specific expectations, not one buffer.
  • Snapshot the source if you need revision history.
04Useful answers

Questions property teams ask

Does a 32.7-hour median mean police took that long to respond?

No. It measures how long after the recorded offense start the report was published, not any response time.

Is a seven-day buffer enough?

It depends on the offense and the decision. 16% of records exceed seven days, and some exceed three weeks.

Why are some offense times exactly midnight?

They likely encode missing precision rather than a real midnight event; they are counted and handled as a sensitivity.

Why build a revision log?

The public dataset publishes no row-level revision history, so change over time is only visible if you snapshot it yourself.

05Inspect the work

Our methods, limits, and sources

How we calculated this

This is original 911 Sentinel research — we gathered the records, ran every calculation below, and published the aggregate dataset.

We computed the interval between SPD offense_date and report_date_time for accepted property-offense records and modeled it with Kaplan–Meier and Cox proportional hazards.

  1. We pulled 116,846 property-offense records (2023–2026) and deduplicated by offense id.
  2. We flagged sentinel dates, negative intervals, and exact-midnight timestamps.
  3. We computed the lag distribution and by-offense medians.
  4. We fit Kaplan–Meier and a Cox model with offense type, and recorded a dataset revision-log baseline.
What this analysis cannot establish
  • offense_date is an offense start and can be a broad or estimated window.
  • Exact-midnight timestamps may be imprecise.
  • Only finalized reports appear; recent weeks undercount.
  • This is not emergency, dispatch, arrival, or police response time.
  • Socrata exposes no row revision history, so our log is a snapshot diff, not an official record.
Sources

The raw records come from the sources below; the study design, analysis, charts, and conclusions are our own.

New coverage — intake open

Treat freshness as part of the decision

A property review should use data with a visible as-of date and a provisional window for recent periods, not last week treated as final.

911 SENTINEL — PROTECTIVE OPERATIONS STANDARDOPERATING RECORD AVAILABLE ON REQUEST SEATTLE, WA