Where the vehicle is kept
Home, overnight parking and other parked locations.
- Theft and vandalism exposure
- Weather and flood exposure
- Parking environment
A postcode contains several different risks
Location combines several mechanisms that need separate testing.
The executive takeaway
Geographic differences may capture several distinct risk mechanisms.
Different exposures need different location evidence
Conceptual framework. Match each mechanism to where the vehicle is exposed.
Where the vehicle travels
The roads, conditions and times of its journeys.
- Road layout and traffic
- Weather along the route
- Time and amount of travel
Weather can affect a parked or moving vehicle. A home address and a collision location describe different exposures.
What geography does the released evidence cover?
- Road outcomes
- County / unitary authority · published
- Weather
- Country · published
- Vehicle crime
- Police force area · published
- Flood hazard
- Lead local flood authority; England; snapshot only · published
Different geographic scales and denominators need to be reconciled before combining them. This visual does not show fitted risk weights or a measured pricing improvement.
Keep in mindCollision location is not the policyholder’s home exposure or an insurer price factor.
What supports this view?
Established: Public road data describes events and traffic where they occur.
Evidence to change the assessment: Reassess after decomposed geographic features improve later, held-out claims.
Department for Transport
RAS0403: collisions and casualties by local authority
- Department for Transport: RAS0403: collisions and casualties by local authority
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
Department for Transport
Local authority motor vehicle mileage
- Department for Transport: Local authority motor vehicle mileage
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
Evidence dates & releaseEvidence published 4 Oct 2026
- Evidence period
- 2016–2025The time measured. Publication and event dates appear with the individual sources.
- Assessment changed
- Change date unavailable in this releaseLast recorded material change to the insurance assessment.
- Explainer changed
- Change date unavailable in this releaseLast recorded change to its evidence or explanation. A routine source check does not reset it.
- Source checks
- 22 Sept 2026 – 26 Sept 202624 of 24 sources have a check date in this release. These dates describe inspection, not new evidence.
Approved evidence published 4 Oct 2026This topic, its downloads and linked Data Lab views use the same selected release.
Inspect the sources and their datesRead the explanation, evidence and limitations
A postcode contains several different risks
Location can conceal road use, environment and other hazards. Separating those mechanisms makes geographic differences easier to understand and test.
Established: Public road data describes events and traffic where they occur.
Developing: Combining carefully aligned evidence can explain more of the geographic pattern.
Still uncertain: Collision location is not the policyholder’s home exposure or an insurer price factor.
What changed in the evidence?
RAS0403: collisions and casualties by local authority
- Department for Transport: RAS0403: collisions and casualties by local authority
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
Local authority motor vehicle mileage
- Department for Transport: Local authority motor vehicle mileage
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
RAS0302: road-type collision and casualty rates
- Department for Transport: RAS0302: road-type collision and casualty rates
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
Insurance implication
Geographic differences may capture several distinct risk mechanisms.
Current response: Test
Watch next: Reassess after decomposed geographic features improve later, held-out claims.
Collision location is not the policyholder’s home exposure or an insurer price factor.
Released sources
- Department for Transport: RAS0403: collisions and casualties by local authority
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Department for Transport: Local authority motor vehicle mileage
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Department for Transport: RAS0302: road-type collision and casualty rates
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Office for National Statistics: Counties and Unitary Authorities December 2025 boundaries
Evidence period: December 2025Checked in this release: 23 Sept 2026
- Met Office: Rainfall — England
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Rainfall — Wales
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Rainfall — Scotland
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Air frost — England
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Air frost — Wales
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Air frost — Scotland
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Wet days ≥1 mm — England
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Wet days ≥1 mm — Wales
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Met Office: Wet days ≥1 mm — Scotland
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
- Home Office: Recorded vehicle offences by police force · Open data file for Recorded vehicle offences by police force
Evidence period: Calendar 2016–2025; source updated 23 July 2026Checked in this release: 23 Sept 2026
- Office for National Statistics: Police Force Areas December 2023 boundaries
Evidence period: December 2023Checked in this release: 23 Sept 2026
- Environment Agency: Properties in river and sea flood hazard
Evidence period: June 2026Checked in this release: 23 Sept 2026
- Environment Agency: Properties in surface-water flood hazard
Evidence period: September 2025Checked in this release: 23 Sept 2026
- Department for Transport: STATS19 annual releases
Evidence period: Latest available releaseChecked in this release: 26 Sept 2026
- Department for Transport: Road safety table index
Evidence period: Latest available releaseChecked in this release: 23 Sept 2026
- Police.uk: Police street-level crime
Evidence period: Monthly releasesChecked in this release: 23 Sept 2026
- Environment Agency: River and sea flood hazard
Evidence period: Current hazard modelChecked in this release: 23 Sept 2026
- Environment Agency: Surface-water flood hazard
Evidence period: Current hazard modelChecked in this release: 23 Sept 2026
- Department for Transport: Resident travel and rural/urban classification
Evidence period: Annual series; 2024 classification breakChecked in this release: 23 Sept 2026
- DVSA: Anonymised MOT data
Evidence period: Annual releasesChecked in this release: 22 Sept 2026
THE ACTUARY LENS · GEOGRAPHIC RISK
Beyond Postcodes
Location is a container of different risk mechanisms—not an explanation by itself. The public road evidence can test geographic differences and persistence; it cannot establish an insurer pricing lift.
INSURANCE IMPLICATIONS
The five conclusions
- Direction
- Geographic differences may capture several distinct risk mechanisms.
- Magnitude
- Public road-outcome differences are not insurer pricing relativities.
- Timing
- Annual evidence supports a held-out geographic study.
- Evidence
- Authority-level collisions and traffic; not residential policy exposure.
- Response
- Test before changing portfolio assumptions.
What would change this view? Reassess after decomposed geographic features improve later, held-out claims.
Different road outcomes
The approved 2016–2025 authority-level evidence aligns police-reported injury collisions with motor vehicle miles. Compare frequency, serious injury and traffic separately in the Lab.
What drives the difference?
A location factor can mix exposure, road environment, behaviour, severity and correlated hazards. A more useful pricing question tests those components individually and asks whether differences persist in later years.
No claims lift claimed
These public measures are not postcode-level insurer claims or resident mileage. A pricing benefit would need held-out claims, credible exposure and governance on an insurer’s own portfolio.
Inspect the evidence
Use the Lab for the interactive maps, annual series, source records and study protocol. Filters from an older map link are carried into the Lab.
Open Postcode Risk Lab →To explore how road outcomes change over time, continue to Road Risk →
UPDATES
Evidence and assessment changes
Evidence dates
Observation periods and publication dates appear with the supporting sources in Data & sources. A source check records an inspection; it does not make the evidence newer.
Assessment changes
The assessment changes when the evidence justifies a different insurance conclusion. Its change date is recorded separately from source checks and the date this release was published.
Data & sources
Trace the assessment to its publications, datasets and observation periods.
Evidence dates & releaseEvidence published 4 Oct 2026
- Evidence period
- 2016–2025The time measured. Publication and event dates appear with the individual sources.
- Assessment changed
- Change date unavailable in this releaseLast recorded material change to the insurance assessment.
- Explainer changed
- Change date unavailable in this releaseLast recorded change to its evidence or explanation. A routine source check does not reset it.
- Source checks
- 22 Sept 2026 – 26 Sept 202624 of 24 sources have a check date in this release. These dates describe inspection, not new evidence.
Approved evidence published 4 Oct 2026This topic, its downloads and linked Data Lab views use the same selected release.
Inspect the sources and their datesData and downloads
Evidence limit: Collision location is not the policyholder’s home exposure or an insurer price factor.
Sources behind this view (24)
Dates belong to the selected release. A source check is an inspection, not new evidence. External source links open in a new tab.
Jump to a source
Source 1 · Department for Transport
RAS0403: collisions and casualties by local authority
Qualification: All roads; latest geography in the workbook. Severity-adjusted FSC and KSI. Reporting warnings for Avon and Somerset and Staffordshire in 2022–2023 are excluded from persistence tests.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Scope and supported claims for RAS0403: collisions and casualties by local authority
Used for: Department for Transport
Source 2 · Department for Transport
Local authority motor vehicle mileage
Qualification: Motor vehicle miles; pedal cycles excluded. Exact authority code and year join. Independent rate reconciliation against RAS0403; mismatches are suppressed.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Scope and supported claims for Local authority motor vehicle mileage
Used for: Department for Transport
Source 3 · Department for Transport
RAS0302: road-type collision and casualty rates
Qualification: National road-type rates include pedal-cycle miles. Kept separate from local-authority motor-mile rates.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Scope and supported claims for RAS0302: road-type collision and casualty rates
Used for: Department for Transport
Source 4 · Office for National Statistics
Counties and Unitary Authorities December 2025 boundaries
Qualification: ONS BGC source generalized to approximately 500 m for display. Boundaries are not address-level hazard determinations.
- Evidence period
- December 2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 5 · Met Office
Rainfall — England
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 6 · Met Office
Rainfall — Wales
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 7 · Met Office
Rainfall — Scotland
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 8 · Met Office
Air frost — England
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 9 · Met Office
Air frost — Wales
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 10 · Met Office
Air frost — Scotland
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 11 · Met Office
Wet days ≥1 mm — England
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 12 · Met Office
Wet days ≥1 mm — Wales
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 13 · Met Office
Wet days ≥1 mm — Scotland
Qualification: Country-level area average from HadUK-Grid. Descriptive weather exposure, not a local postcode estimate or a fitted weather effect on collisions.
- Evidence period
- 2016–2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 14 · Home Office
Recorded vehicle offences by police force
Open publication: Recorded vehicle offences by police force (opens in a new tab)Open data file: Recorded vehicle offences by police force (opens in a new tab)
Qualification: Sum of offence codes 37.2, 126, 45 and 48. Financial Q4 assigned to the following calendar year. Counts, not theft rates; no parked-vehicle denominator. Force recording practices vary; missing counts suppressed. British Transport Police excluded; Devon and Cornwall explicitly mapped to ONS Devon & Cornwall.
- Evidence period
- Calendar 2016–2025; source updated 23 July 2026
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 15 · Office for National Statistics
Police Force Areas December 2023 boundaries
Qualification: Exact official force name match with explicit Devon and Cornwall alias; no areal allocation of British Transport Police.
- Evidence period
- December 2023
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 16 · Environment Agency
Properties in river and sea flood hazard
Qualification: Current model snapshot using NRD 2023 properties. High ≥3.3% annual flood chance; medium 1–3.3%. Percentage of ALL properties in the authority. Exact LLFA name match to DfT/ONS authority register. Not motor-loss experience and not historical yearly evidence.
- Evidence period
- June 2026
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 17 · Environment Agency
Properties in surface-water flood hazard
Qualification: Current model snapshot using NRD 2023 properties. High ≥3.3% annual flood chance; medium 1–3.3%. Percentage of ALL properties in the authority. Exact LLFA name match to DfT/ONS authority register. Not motor-loss experience and not historical yearly evidence.
- Evidence period
- September 2025
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 18 · Department for Transport
STATS19 annual releases
Qualification: Shared STATS19 monitor; official tables above supply this initial release.
- Evidence period
- Latest available release
- Source published
- Date unavailable
- Source checked
- 26 Sept 2026
Source 19 · Department for Transport
Road safety table index
Qualification: Detect new table URLs and revisions; changed evidence needs review.
- Evidence period
- Latest available release
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 20 · Police.uk
Police street-level crime
Qualification: Approximate incident locations. Recorded vehicle crime is not insured theft frequency.
- Evidence period
- Monthly releases
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 21 · Environment Agency
River and sea flood hazard
Qualification: Modelled hazard, not observed motor loss. England only.
- Evidence period
- Current hazard model
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 22 · Environment Agency
Surface-water flood hazard
Qualification: Modelled hazard; do not replay one model snapshot as historic annual evidence.
- Evidence period
- Current hazard model
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 23 · Department for Transport
Resident travel and rural/urban classification
Qualification: Resident travel is distinct from traffic through an area; old/new classifications are not directly comparable.
- Evidence period
- Annual series; 2024 classification break
- Source published
- Date unavailable
- Source checked
- 23 Sept 2026
Source 24 · DVSA
Anonymised MOT data
Qualification: Use shared MOT Evidence Lab. Test-station postcode area is not keeper home postcode.
- Evidence period
- Annual releases
- Source published
- Date unavailable
- Source checked
- 22 Sept 2026