Vehicle Fatality Trends

Which vehicles are involved in Victoria's road deaths — and are they killing their occupants, or everyone else? Every figure below is computed from the Victorian Government's open road crash dataset.

Victoria · 2012–2025 Source: DTP Victoria Road Crash Data (CC-BY-4.0) “Involved in”, never “caused” — see method
Lives lost
Fatal crashes
2012 – Oct 2025
Casualty crashes analysed
all severities
Vehicles involved
362.7K
vehicle records linked

The brand race — fatal-crash involvements since 2012

Cumulative fatal crashes in Victoria involving each brand's light vehicles (cars, wagons, utes, vans). Raw counts reflect how many of each brand are on the road as much as how they perform — the sections below adjust for that.

2012

Ford is highlighted: it ranks third on raw involvement, but tops the severity table below. 2025 covers January–October only.

When it crashes, how often does someone die?

Share of each brand's crash involvements that were fatal — a severity rate that removes the “there are simply more of them on the road” effect. Victorian light vehicles, brands with ≥2,000 involvements.

The grey hairline is the all-brand average. A higher bar means that when this brand's vehicles crash, the crash is more likely to kill someone — a blend of vehicle age, crashworthiness, where and how they're driven.

Killers vs killed — who dies when each model crashes?

For each high-involvement model: deaths of its own occupants (horizontal) vs deaths of other parties — pedestrians, riders, occupants of other vehicles (vertical). Above the diagonal, a model's fatal crashes kill others more than its own. This is the “aggressivity vs crashworthiness” lens used by European (ETSC/VIAS) and MUARC research.

Data table

The heavy-vehicle severity ladder

Fatality rate per crash involvement for every model with ≥1,500 involvements. Heavy 4WDs and utes dominate the top — Victorian data reproducing the European mass/size findings.

Old vehicles are deadlier

Fatality rate by vehicle age at the time of the crash (light vehicles). A 20+ year-old vehicle's crash is more than twice as likely to be fatal as a near-new one's.

Consistent with European (ETSC) findings on fleet age — and Australia's fleet is old. Age = crash year − year of manufacture.

The SUV & ute shift

Vehicle involvements in fatal crashes by class and year. As Australia's fleet swapped sedans for SUVs and dual-cab utes, fatal-crash involvement followed.

SUV/4WD identified by model (the crash dataset has no SUV category — see method). 2025 is a partial year.

Who is dying

Victorian road deaths per year by road user.

Method & honesty

Source. Victoria Road Crash Data (Department of Transport and Planning, CC-BY-4.0), ACCIDENT / VEHICLE / PERSON tables, crashes from 1 January 2012 to 31 October 2025 (portal snapshot of 8 June 2026). The dataset records police-reported casualty crashes; property-damage-only crashes are absent.

Involvement, not fault

Every brand and model figure counts vehicles involved in fatal crashes. Crash data does not attribute legal fault, and neither do we. In multi-vehicle crashes the same death is counted against each involved vehicle, so columns are not summable across brands.

Denominators

Raw counts are dominated by fleet share — Toyota tops them because Victoria drives more Toyotas. The severity rate used here, fatal involvements ÷ all casualty-crash involvements, is computed entirely within the dataset (single source), so no external registration series is mixed in. It answers: given this vehicle crashed, how likely was the crash fatal? That blends the vehicle's crashworthiness and aggressivity with who drives it, where, and at what speeds — it is not a pure vehicle-safety score. For engineering-grade per-model ratings see MUARC's Used Car Safety Ratings, which corroborate the pattern shown here.

Attribution of deaths

“Own occupants” are fatalities recorded in the vehicle (linked person records); “other parties” are all other fatalities in crashes that vehicle was involved in, including pedestrians (who are in no vehicle). Deaths of motorcyclists struck by a car count against the car as other-party deaths.

Data quality

Make and model are free-text fields truncated to six characters; we harmonise the common values (e.g. MITSUB → Mitsubishi, COMMOD → Commodore). Records with unknown make (~9% of vehicles) are excluded from brand rankings and disclosed here. The dataset has no SUV body category — SUVs are tagged via a model list (LandCruiser, Prado, RAV4, CX-5, Territory, Forester …), a documented heuristic. Model-level views suppress cells with fewer than 5 fatal involvements; severity rates are only shown at ≥1,500 involvements (brands) or ≥1,500 (models). 2025 is a partial year (to 31 October).

Context

The “killers vs killed” framing follows current European research: the VIAS Institute (Belgium) found a +300 kg mass difference raises the other party's fatal-injury risk by ~80%, and pickups triple pedestrian/cyclist fatal risk; the European Transport Safety Council is campaigning on large-vehicle risk in urban areas. This dashboard is, to our knowledge, the first public interactive view of that dichotomy in Victorian data.