DROP · VANDALISMMARCH 2026 BRIEFINGSAN FRANCISCO · 23.5K residents

Nob Hill Crime Rate Trends — San Francisco

Nob Hill is a steep, wealthy district crowned by historic luxury hotels like the Fairmont and the Mark Hopkins, the Gothic Grace Cathedral, and the cable cars that climb its slopes. Long associated with the railroad and silver-mining elite who rebuilt the area after the 1906 earthquake, it remains a residential and hotel district with sweeping bay views.

VANDALISM · 24-MO COUNT03 2026 · 6
0163212-mo avg: 16.6
NOB HILLCITYWIDE TREND (RESCALED)-21% 12MO YOY
-25%MoM
-24%12mo YoY
199last 12mo
6this month
01 · TL;DR

Seven categories moved in Nob Hill this March — four ran below trend on a single-month basis, three registered as sustained multi-month structural shifts. The overall shape is broadly downward across both property crime and violent categories, with no spikes or rare-event signals anywhere in the mix.

Vandalism, motor vehicle theft, and theft from vehicle lead the top three signals, all below trend. The 12-month vandalism total of 199 sits well under its multi-year baseline, and the category's trailing 12-month count has fallen 23.5% against the prior year. Theft from vehicle and motor vehicle theft are moving in the same direction — down 48.7% and 40.9% year-over-year respectively — while robbery's 12-month total of 41 is half the 82 recorded the year before. Other Larceny is the one counter-trend: up 18.0% against the prior 12 months, with 472 incidents against 400 the year before.

4 drops3 sustained shifts
02 · Notable signals

Notable signals 4

DROP · VANDALISMZ = 4.49

Vandalism

The past 12 months saw 199 incidents — about 36% below the 310 average from prior years.

DROP · MOTOR VEHICLE THEFTZ = 3.68

Motor Vehicle Theft

The past 12 months saw 101 incidents — about 51% below the 207 average from prior years.

DROP · THEFT FROM VEHICLEZ = 3.62

Theft from Vehicle

The past 12 months saw 203 incidents — about 67% below the 616 average from prior years.

DROP · BURGLARYZ = 3.08

Burglary

The past 12 months saw 121 incidents — about 53% below the 260 average from prior years.

03 · By category

All categories, last 24 months

Each panel: recent monthly count vs. trailing 12-month context. MoM is the most recent month vs. the one before; 12mo YoY compares the trailing year to the year before that.

Homicidebelow threshold
2024-042026-03
Robbery-50%
2024-042026-03
Aggravated Assault-23%
2024-042026-03
Sexual Assaultbelow threshold
2024-042026-03
Burglary-25%
2024-042026-03
Theft from Vehicle-49%
2024-042026-03
Other Larceny+18%
2024-042026-03
Motor Vehicle Theft-41%
2024-042026-03
Vandalism-24%
2024-042026-03
Arson+120%
2024-042026-03
05 · Forecast

What next month likely looks like

Forecasts trained through March 2026, with a likely range we're 95% confident the actual count will fall inside. Categories with too little recent volume — or violent categories at the neighborhood level — show no forecast and are surfaced through signals above instead. See the methodology page for the gating rules.

Aggravated Assault

NO FORECAST

Too low-volume per neighborhood for a reliable point forecast — see the rare-event and streak-break signals above instead.

Arson

NO FORECAST

Below the volume threshold for a reliable forecast — too few incidents in recent months to project from.

Burglary

APRIL 2026
Most likely 14 next month — likely between 2 and 27.
+41% vs 12-month average (≈10.1)

Homicide

NO FORECAST

Too low-volume per neighborhood for a reliable point forecast — see the rare-event and streak-break signals above instead.

Motor Vehicle Theft

APRIL 2026
Most likely 11 next month — likely between 0 and 22.
+27% vs 12-month average (≈8.4)

Other Larceny

APRIL 2026
Most likely 29 next month — likely between 9 and 49.
26% vs 12-month average (≈39.3)

Robbery

NO FORECAST

Too low-volume per neighborhood for a reliable point forecast — see the rare-event and streak-break signals above instead.

Sexual Assault

NO FORECAST

Too low-volume per neighborhood for a reliable point forecast — see the rare-event and streak-break signals above instead.

Theft from Vehicle

APRIL 2026
Most likely 10 next month — likely between 0 and 31.
41% vs 12-month average (≈16.9)

Vandalism

APRIL 2026
Most likely 21 next month — likely between 10 and 33.
+27% vs 12-month average (≈16.6)
06 · Context & comps

How Nob Hill compares

Peer neighborhoods picked by closest 12-month vandalism volume — a pragmatic v1 of peer matching. Demographic / housing-stock peer matching isn't built yet (we deliberately don't ingest income or race data alongside crime). Volume similarity has the right intuition: “neighborhoods experiencing comparable vandalism levels.”

07 · Patterns

Recurring local terms (last 12 months)

Top terms in incident descriptions for Nob Hill, excluding generic crime taxonomy. Useful as texture — what kinds of specifics show up here that don't show up elsewhere.

warrantshopliftinglockedpossessionlostfraudulentsuspiciousforciblenarcoticsparaphernaliaphoneorderfoundbuildingoccurrenceaggravatedrecoveredunlawfuldeathforcecardcausefalsemoneyinvestigation
When does it happen?

Hour-of-day, day-of-week, and seasonality

Distribution of bucketed incidents in this neighborhood across the full analysis window. Useful for routine context — shopping-strip thefts vs. late-night assaults read very differently when you can see when each typically happens.

HOUR OF DAY · ALL CATEGORIES
06081,21612am6am12pm6pm11pm

Hour 0 is mildly inflated by reports without a known time defaulting to midnight — see methodology.

DAY OF WEEK · ALL CATEGORIES
01,4012,802MonTueWedThuFriSatSun
MONTH OF YEAR · ALL CATEGORIES
08221,643JanFebMarAprMayJunJulAugSepOctNovDec
08 · Methodology

How we built this page

Data → Anomalies → Forecast → Page

Incident data is pulled from SFPD's open dataset, mapped to 10 NIBRS-aligned categories, and aggregated to neighborhood × category × month. Anomalies are surfaced using strict thresholds (~p < 0.01). Forecasts are Prophet with low-count gating; violent categories at the neighborhood level skip the forecast and show rare-event / streak signals instead.

Spike rule: 12-mo total > baseline mean + 2.5σ AND ≥ 20 incidents AND 6-mo confirms. Drop rule: 12-mo total < baseline mean − 2.5σ AND baseline mean ≥ 20. Rare event: any incident in the last 90 days, no prior comparable in ≥ 5 years.