POSTCODE RISK LAB

Do geographic differences persist?

Explore road outcomes across authorities and test whether differences recur in later years. These public proxies do not measure postcode-level insurance claims.

Explore the geographic evidence

2016–2025 · Counties and unitary authorities, December 2025

THE ACTUARY LENSGEOGRAPHY · EVIDENCE · PERSISTENCE

Beyond Postcodes — Geo Risk

Unpack geography. Follow the risk. Test whether the differences endure.

Read the white paper
Official sources · 2016–2025207 areas · 2,070 area-yearsPublic evidence, not a validated insurance tariff
Tracker publication · 25/09/2026
Evidence JSONFull CSV
Overview · County Durham · Injury collision frequency · Map year 2025 · History 2016–2025 · Assessment 2022–2025
Release and terminology

lens-2026-09-25-a1fc94c9307e

KSI = killed or seriously injured. FSC = fatal or serious collisions. GB = Great Britain. UK = United Kingdom. DfT = Department for Transport. ONS = Office for National Statistics. PFA = police force area. NRD = National Receptor Dataset. These public outcomes are not insurance claim costs.

Do geographic differences endure?

The question

Do recorded road outcomes differ between areas, and do those differences repeat in later years?

The evidence

Official area-level observations, fixed discovery groups and a separate later-year assessment. Explore exposure, frequency and injury severity separately.

The limit

Authority-level patterns support investigation. They cannot prove within-postcode differences or insurance pricing lift; that requires geocoded insurer exposure and held-out claims.

What the released evidence says

Injury collision frequency · formal assessment 2022, 2023, 2024, 2025 · all 207 authorities.

  • 71 — consistently above
  • 111 — consistently below
  • 9 — mixed across years
  • 16 — incomplete evidence

These counts describe direction against the fixed-panel benchmark. Mixed and incomplete evidence remain visible. They are not probabilities or a significance test.

Geography can describe several different mechanisms. The evidence should keep them separate.

Beyond Postcodes · Johan Strydom

Through The Actuary Lens

Authority-level patterns support investigation, but cannot establish within-postcode differences or pricing lift without geocoded insurer exposure and held-out claims.

Published measures & limits

Injury collision frequency

Police-reported injury collisions on roads in this area, divided by motor vehicle miles travelled in the same area. Not all insurance claims.

Unit: collisions / 100 million motor vehicle miles

Serious injury exposure

Killed or seriously injured casualties, adjusted by DfT for reporting changes. Not claim cost or the probability of a claim.

Unit: killed or seriously injured casualties / 100 million motor vehicle miles

Fatality exposure

People killed, divided by motor vehicle miles. Road casualty outcomes do not establish monetary claim severity.

Unit: fatalities / 100 million motor vehicle miles

Severity conditional on collision

DfT-adjusted fatal or serious collisions divided by all reported injury collisions. The denominator is collisions, not miles.

Unit: % injury collisions fatal or serious

Traffic through the area

Traffic occurring on roads in the area. This is not the annual mileage of residents or vehicles kept here.

Unit: billion motor vehicle miles

About the five Postcode Risk Indices

These five public-data measures are different from Johan’s proprietary Postcode Risk Indices. The indices await validated definitions and inputs; no index values are published here.

Sources & release

Analysis period: 2016–2025 · Approved release: 25/09/2026 · Validation: 25/09/2026

The analysis above is from the approved release. Later checks cannot change published findings until reviewed and approved.

23 sources need attention. The source table is open below; each row explains whether a check failed or new evidence awaits review.

Inspect 24 relevant sources
SourceRecorded source periodChecked in releaseValidated in releaseCurrent source check
RAS0403: collisions and casualties by local authority2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Local authority motor vehicle mileage2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
RAS0302: road-type collision and casualty rates2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Counties and Unitary Authorities December 2025 boundariesDecember 2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Rainfall — England2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Rainfall — Wales2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Rainfall — Scotland2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Air frost — England2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Air frost — Wales2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Air frost — Scotland2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Wet days ≥1 mm — England2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Wet days ≥1 mm — Wales2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Wet days ≥1 mm — Scotland2016–2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
Recorded vehicle offences by police forceCalendar 2016–2025; source updated 23 July 2026Not recordedNot recordedNew evidence awaiting review — not yet in findings
Police Force Areas December 2023 boundariesDecember 2023Not recordedNot recordedNew evidence awaiting review — not yet in findings
Properties in river and sea flood hazardJune 2026Not recordedNot recordedNew evidence awaiting review — not yet in findings
Properties in surface-water flood hazardSeptember 2025Not recordedNot recordedNew evidence awaiting review — not yet in findings
STATS19 annual releasesLatest available release21/09/2026Not recordedNew evidence awaiting review — not yet in findings
Road safety table indexLatest available releaseNot recordedNot recordedNew evidence awaiting review — not yet in findings
Police street-level crimeMonthly releasesNot recordedNot recordedNew evidence awaiting review — not yet in findings
River and sea flood hazardCurrent hazard modelNot recordedNot recordedNew evidence awaiting review — not yet in findings
Surface-water flood hazardCurrent hazard modelNot recordedNot recordedNew evidence awaiting review — not yet in findings
Resident travel and rural/urban classificationAnnual series; 2024 classification breakNot recordedNot recordedNew evidence awaiting review — not yet in findings
Anonymised MOT dataAnnual releases22/09/202625/09/2026No source issue recorded
Technical release IDlens-2026-09-25-a1fc94c9307e