STATS19 DATA LAB · GREAT BRITAIN

STATS19 Data Lab

Police-reported injury collisions in Great Britain · geographic history 2016–2025 · record analysis 2021–2025

Coverage research

Do fuel differences remain after allowing for miles?

Fuel-specific rates depend on compatible mileage; stock is not exposure.

Charts displayed; exact values remain available with each chart.

VEHICLE MILEAGE · PHASE 4

Do fuel differences remain after allowing for miles?

Start with the supported car-and-taxi rate. Inspect fuel counts and the MOT mileage sample before interpreting a fuel comparison.

Selected comparison: 2025 · All supported MOT fuels

Change comparison 2 applied filters2025 · All supported MOT fuels

YearMOT profile fuel

2025 · derived descriptive rate0.0483Car-and-taxi involvements per 100,000 car-and-taxi miles

Official aggregate car-and-taxi mileage262.41bn126,826 involved cars and taxis · all fuels and ages

Fuel-specific per-mile comparisonWithheldPassenger-car mapping and calibrated group mileage are missing

Derived descriptive rate

Numerator / population: Each involved car or taxi counted once per collision · All cars and taxis; all fuels and ages
Denominator / unit: Car-and-taxi vehicle miles · Involvements per 100,000 matching vehicle miles
Scope: Great Britain · 2025
Exposure basis: Official aggregate mileage
Coverage: 126,826 involvements and 262,411,300,000 miles. Fuel-specific mileage remains unavailable.

Measurement details · outcome and adjustment

Outcome: Car-and-taxi involvements

Adjustment: Descriptive; no age or fuel adjustment

Interpretation limit: Involvement does not imply fault. Licensed-car stock and MOT sample distance are separate indicators, not fuel-specific motor-mile denominators.

Supported broader comparison

Car-and-taxi involvements per 100,000 matching miles

All cars and taxis · all fuels and ages

Select a point, or use the Data tab for exact values. Missing observations remain gaps.

Exact values · also in Data (5 rows)

Car-and-taxi involvements per 100,000 matching miles
Year	All cars and taxis · all fuels and ages	All cars and taxis · all fuels and ages · conditional 95% interval
2021	0.0583	0.0579–0.0587
2022	0.0554	0.055–0.0558
2023	0.053	0.0526–0.0534
2024	0.0502	0.0499–0.0506
2025	0.0483	0.048–0.0487

95% collision-cluster approximation where available, conditional on the official mileage estimate. DfT mileage uncertainty is not included. 2021 travel was affected by the pandemic. Vehicle involvement does not establish fault.

Exact values · also in Data (5 rows)

Great Britain · all cars and taxis; numerator and exposure use the same year and vehicle class.
Year	Involvements	Matching miles	Per 100,000 miles	Conditional 95% interval
2021	129,067	221,359,300,000	0.0583	0.0579–0.0587
2022	135,235	243,994,000,000	0.0554	0.055–0.0558
2023	133,237	251,299,700,000	0.053	0.0526–0.0534
2024	128,678	256,129,700,000	0.0502	0.0499–0.0506
2025	126,826	262,411,300,000	0.0483	0.048–0.0487

Evidence details · Fuel support, MOT mileage and sensitivity

Fuel counts, stock and mileage are different measures

Fuel-specific rates per mile and age-adjusted comparisons remain unavailable. The stock indicator below uses licensed cars and does not account for how far they travel.

Exact values · also in Data (6 rows)

2025 · STATS19 cars and taxis; GB licensed-car stock. Unknown and other fuels remain in the counts.
Fuel	Involvements	Mean quarter-end licensed cars	Per 10,000 licensed cars	Matched miles	Per 100,000 miles	Age-adjusted rate
Diesel	34,194	9,789,092	34.93	Unavailable	Withheld	Withheld
Battery electric	5,062	1,542,617	32.81	Unavailable	Withheld	Withheld
Hybrid electric (combined)	14,635	Unavailable	Unavailable	Unavailable	Withheld	Withheld
Other propulsion	83	Unavailable	Unavailable	Unavailable	Withheld	Withheld
Petrol	59,337	18,541,662	32	Unavailable	Withheld	Withheld
Unknown propulsion	13,515	Unavailable	Unavailable	Unavailable	Withheld	Withheld

Stock is an approximate annual denominator, not exact active vehicle-years. Missing hybrid stock mapping is retained. Neither the fuel counts nor the stock indicator isolates a causal fuel effect.

What mileage does the MOT sample observe?

2025 · All four supported MOT fuels · Class 4 normal-test sample, ages 3–30. This profile selection changes only the MOT sample, not the all-car road rate or the fuel table.

27,503,477 / 29,603,251 selected vehicle-years have an observed eligible prior mileage interval (92.9%). Missing intervals are not zero mileage.

Exact values · also in Data (4 rows)

2025 · All four supported MOT fuels · Class 4 normal-test sample, ages 3–30
Vehicle age	Selected vehicle-years	Valid intervals	Missing prior interval	Coverage	Median annual-use bounds	50,000+ tail
3–5	5,099,940	3,489,892	1,610,048	68.4%	5,000–<10,000	4,846
6–9	9,504,146	9,364,004	140,142	98.5%	5,000–<10,000	9,732
10–14	8,978,442	8,819,253	159,189	98.2%	5,000–<10,000	7,135
15–30	6,020,723	5,830,328	190,395	96.8%	0–<5,000	6,807

These are descriptive annualised-use bands, not calendar-year passenger-car miles. Median bounds reflect grouped readings, not confidence intervals. Class 4 includes other vehicles. The open tail means the mean, total mileage and outlier contribution cannot be reconstructed from these bands. Exposure imputation share: unavailable.

Read the MOT cohort, cleaning and selection methodology

Exposure and calibration checks

Exact values · also in Data (9 rows)

Checks determine whether a mileage comparison can be published.
Check	Status	Evidence
Broad car-and-taxi exposure	Supported	Same-year GB official car-and-taxi miles; all fuel and age groups combined.
MOT observed sample	Available for description	Reviewed Class 4 normal-test cohort, ages 3–30; existing quality exclusions remain.
Passenger-car population	Blocked	Class 4 is not car-only. No verified body-type mapping aligns MOT testers with licensed cars and STATS19 cars/taxis.
Exact interval mileage	Blocked	Grouped annualised use with an open 50,000+ tail does not identify means, calendar mileage, outlier sums or vehicle-cluster uncertainty.
Quarterly fleet expansion	Partial	GB quarterly petrol/diesel/BEV licensed-car stock is available. The archived stock file has no age field; UK age tables cannot be substituted or multiplied by GB fuel shares.
Vehicle-age numerator	Blocked	0 of 183,948 vehicle involvements have known age (all vehicle types). Fuel-by-age rates require matching joint exposure.
Young vehicles and attrition	Blocked	No approved imputation for pre-MOT mileage, non-return, entry/exit, SORN or selection into observed tests.
DfT calibration	Unavailable	Raw expanded mileage, residual and calibration factor remain unavailable until population and interval checks pass. Overseas/foreign-vehicle coverage also needs assessment.
Uncertainty and headline stability	Withheld	Fuel headlines require collision-cluster and exposure uncertainty plus calendar, outlier, young-vehicle, attrition and mapping sensitivities. No exposure-imputation proportion can yet be estimated.

Exact values · also in Data (1 rows)

A missing expansion is never treated as zero or filled by forcing a calibration factor.
Raw expanded MOT miles	DfT benchmark miles	Pre-calibration residual	Calibration factor	Imputed exposure share
Unavailable	262,411,300,000	Unavailable	Unavailable	Unavailable

Sensitivity and uncertainty requirements

Exact values · also in Data (5 rows)

No numerical sensitivity is claimed when its required exposure inputs are absent.
Assumption	Alternative	Status	Missing evidence
Vehicles before first MOT	Compare externally supported young-vehicle use with older observed vehicles; vary the imputed young-vehicle miles.	Not estimable	No compatible young-vehicle mileage or joint GB fuel/age fleet expansion is present.
Attrition and SORN	Include entry/exit and non-return scenarios; use licensed vehicle-years rather than surviving testers.	Not estimable	Non-return does not distinguish scrappage, SORN, export, missed tests or gaps.
Calendar allocation	Allocate each interval by covered days; compare endpoint-year allocation and 270–450-day test gaps.	Requires raw rebuild	Published use bands lack exact interval endpoints and mileage sums.
Fuel mapping	Retain unknowns and changing-code exclusions; compare strict EL/PE/DI groups with combined HY only.	Mapping limitation	HY does not identify plug-in hybrids. Car body mapping is not validated.
Odometer and outliers	Normalise declared units, retain chronology across missing readings; compare 50,000 and 100,000 annual-mile cutoffs.	Requires raw rebuild	The open 50,000+ band cannot quantify outlier mileage or its effect on expansion.

Headline fuel comparisons stay withheld until plausible assumptions are tested. Exposure uncertainty must be combined with collision-cluster uncertainty; driver age or road use can be adjusted only with mileage in the same strata.

Definitions, sources and reproducibility

The supported rate counts distinct collision/vehicle pairs for cars and taxis against official GB car-and-taxi miles. Involvement does not establish fault, claims or a causal fuel effect.
Fuel rows retain the original STATS19 categories and unknowns. Licensed stock is a separate per-10,000-vehicle indicator; it is never substituted for miles. Quarterly stock is the existing GB mean of four quarter-end observations, not exact licensed vehicle-years.
MOT profiles describe the existing reviewed normal-test Class 4 cohort, ages 3–30. They include vehicles other than passenger cars and condition on a recorded eligible test. There is no vehicle-level link to STATS19.
Recent-use bands annualise an observed odometer change over 180–730 days. They are not calendar-year exposure. A 50,000+ open tail and grouped counts do not identify the sample mean or total mileage. No midpoint-based expansion is performed.
MOT profile missingness is not imputed exposure. New vehicles, SORN/active status, attrition, overseas mileage and foreign vehicles require explicit treatment before expanding and calibrating to GB road traffic.
A total-mileage calibration cannot repair a missing fuel/age allocation or a population mismatch. Fuel-by-age, age-adjusted fuel rates, annual-mileage-band collision rates and causal comparisons remain unavailable in this version.
The vehicle view and its exports read STATS19 and MOT from one immutable release. Previous roadworthiness approval is not approval of a new mileage exposure model. Failed checks retain the broader official-mileage result.

All-vehicle STATS19 age completeness in 2025: 0 / 183,948. This is a coverage check across all vehicle classes, not a fuel-by-age estimate.

MOT source period: 2022-01-01 to 2025-12-31. Downloads retain the MOT build identity and the STATS19 source hashes.

Download full vehicle mileage audit All-year rates and fuel counts CSV Selected MOT age profile CSV Sensitivity register CSV

Full data & reproducibility

Geographic inputs & release manifest Detailed STATS19 data & manifest Read the actuarial interpretation →

Sources & release

Loading source monitoring. Selected results are already available; live checks do not change them.

Source	Evidence period	Source published	Checked in release	Latest check	Monitoring state
RAS0403: collisions and casualties by local authority	2016–2025	Date unavailable	23 Sept 2026	21 Oct 2026	No source issue recorded
Local authority motor vehicle mileage	2016–2025	Date unavailable	23 Sept 2026	21 Oct 2026	No source issue recorded
Rainfall — England	2016–2025	Date unavailable	23 Sept 2026	23 Oct 2026	Source content changed; review pending
Rainfall — Wales	2016–2025	Date unavailable	23 Sept 2026	23 Oct 2026	Source content changed; review pending
Rainfall — Scotland	2016–2025	Date unavailable	23 Sept 2026	23 Oct 2026	Source content changed; review pending
Air frost — England

Loading approved evidence

STATS19 Data Lab

Methods and limitations

All investigations

Do fuel differences remain after allowing for miles?

Do fuel differences remain after allowing for miles?

Supported broader comparison

Fuel counts, stock and mileage are different measures

What mileage does the MOT sample observe?

Sources & release

Sources & release