Vehicle Safety Ratings

National Traffic Fatality Trends (NHTSA FARS, 2014–2024)

2024 Fatalities (est.)

39,345

down from 40,901 in 2023

2024 Rate (est.)

1.20

per 100M vehicle miles traveled

10-Year Trend

Elevated

peaked at 1.37 in 2021; was 1.08 in 2014

2024 data is an early NHTSA estimate subject to revision. Bars show total fatalities (left axis); line shows rate per 100M VMT (right axis).

Fatalities by Road User Type (NHTSA FARS, 2014–2023)

Passenger Car Occ.

—

of 2023 fatalities

Light Truck Occ.

—

of 2023 fatalities

Motorcyclists

—

of 2023 fatalities

Pedestrians + Cyclists

—

of 2023 fatalities

Occupant Fatality Rate by Vehicle Class (per 100M VMT)

Motorcycle Rate

31.39

per 100M VMT (2023)

~29x the passenger car rate

Rates calculated from NHTSA FARS fatality counts and FHWA VM-1 vehicle miles traveled. Per-model VMT is not publicly available; these rates apply at the broad vehicle-class level only.

Analysis & Insights

What does a decade of federal crash data and IIHS testing actually tell us about which cars keep people alive — and which ones don't? Here are the most important findings from the data above.

Size Kills — But Not the Way You Think

The single strongest predictor of driver death rate in the IIHS data is vehicle size. Minicars dominate the "most dangerous" list: the Mitsubishi Mirage G4 tops the chart at 205 driver deaths per million registered vehicle years — more than 5× the overall average of 38. The Mirage hatchback, Hyundai Accent, and Chevrolet Spark all cluster above 150.

But here's the twist: size also makes vehicles more dangerous to everyone else. The IIHS "danger to others" metric reveals that very large pickups — the Ram 3500, Ford F-350, and Ram 2500 — kill other road users at rates of 120–189 per million registered vehicle years. That's 3.6× the overall average of 53. Your truck might keep you safe, but the physics are zero-sum.

Key takeaway: A Ram 3500 Crew Cab kills other drivers at a rate of 189 per million registered vehicle years — nearly as high as the Mirage G4's rate of killing its own driver. The arms race in vehicle size has real casualties.

The Safest Vehicles Are Boring (And That's Great)

Seven vehicles in the IIHS dataset recorded zero or near-zero driver deaths per million registered vehicle years: the BMW X3, Lexus ES 350, Mercedes E-Class, and Nissan Pathfinder all hit 0. The Audi Q5, Toyota C-HR, and Volvo XC90 were at 2–4. Notice a pattern?

They're mostly midsize SUVs and luxury cars — vehicles with extensive safety tech, solid engineering, and (crucially) drivers who tend to be older, wealthier, and statistically less likely to speed or drive impaired. Separating vehicle engineering from driver behavior is the hardest problem in safety data.

Best bets for safety-conscious buyers: Midsize SUVs with 4WD consistently top both the IIHS and FARS lists. The Subaru Outback (5 deaths/million), Toyota C-HR (2), and Honda Odyssey (6) offer excellent occupant protection without the outsized risk to other road users that very large pickups carry.

IIHS vs. FARS: Where the Data Agrees (and Doesn't)

The IIHS data covers 85 model-year 2020 vehicles with death rates per registered vehicle year. The FARS per-model data covers 337 models over a decade (2014–2023) with estimated rates per 100 million VMT. They measure different things, but the broad strokes align:

Both agree

SUVs are safer

Midsize SUVs consistently show low occupant fatality rates in both datasets. The Ford Escape (FARS rate: 0.95) and Toyota RAV4 appear among the safest in both.

Both agree

Sports cars are deadly

The Ford Mustang has a FARS rate of 6.02 per 100M VMT — the highest among common models. Dodge Challenger and Chevrolet Camaro mirror this in IIHS data.

They diverge

Pickup rates

IIHS shows pickups as moderate for drivers but terrible for others. FARS per-model rates for F-150 (1.04) and Silverado (1.25) look deceptively average — because they're normalized by VMT and pickups rack up a lot of miles.

They diverge

Older sedans

FARS 10-year data captures older fleets. The Chevy Impala (rate: 5.0) and Ford Taurus (2.74) score poorly, but both are discontinued — their FARS rates reflect aging vehicles, not current safety standards.

The COVID Effect: A Permanent Shift?

The national FARS data shows a sharp inflection point in 2020. Despite 13% fewer miles driven (2.83 trillion VMT, down from 3.26T in 2019), fatalities rose 7% to 39,007. The fatality rate per 100M VMT jumped from 1.11 to 1.34 — a 21% spike in a single year.

2021 was even worse: 42,939 deaths at a 1.37 rate, the highest since 2005. What happened? Emptier roads enabled faster driving. Seatbelt use dropped. Impaired driving increased. The behavioral shift proved stickier than the virus: even as VMT recovered to 3.25 trillion by 2023, the fatality rate (1.26) remains 14% above pre-pandemic levels.

The quiet crisis: From 2019 to 2023, annual traffic deaths increased by 4,546 — equivalent to roughly 12 additional deaths every day. The 2024 estimate of 39,345 suggests improvement, but we're still well above the trajectory we were on before the pandemic.

The Pedestrian and Cyclist Emergency

Buried in the "by road user type" chart is the most alarming trend in U.S. traffic safety. Pedestrian fatalities rose 49% from 4,910 in 2014 to 7,318 in 2023. Cyclist deaths rose 49% too — from 723 to 1,075. Meanwhile, passenger car occupant deaths actually fell 20% over the same period (12,543 → 10,096).

Cars are getting safer for the people inside them. They're getting more dangerous for everyone else. The shift from sedans to SUVs and trucks — with their higher front-end profiles and greater mass — is a major factor. A pedestrian struck by an SUV is 2–3× more likely to die than one struck by a sedan at the same speed, according to IIHS research.

What This Means If You're Buying a Car

If occupant safety is your priority, the data is clear:

Choose a midsize or larger vehicle. The physics of crash protection favor mass and crumple zone length. Small cars carry a real and measurable penalty.
4WD/AWD correlates with lower death rates — but this is partly a confounding variable. AWD vehicles tend to be newer, more expensive, and driven by lower-risk demographics.
Avoid vehicles with "performance" reputations. The Dodge Challenger, Charger HEMI, and Chevy Camaro have high death rates that likely reflect driver behavior as much as vehicle design.
Consider your impact on others. If you don't need a full-size truck, choosing a midsize SUV instead means dramatically lower risk to pedestrians, cyclists, and other drivers.

Methodology note: This analysis uses the data as presented in the dashboards above. IIHS rates are based on registered vehicle years (MY 2020, 2018–2021 fatality data). FARS per-model rates use estimated VMT with assumptions about fleet size and annual mileage that introduce uncertainty — treat them as directional, not precise. See the Methodology section below for full details on limitations.

Methodology & Sources

What does this data show?

Driver death rate = number of driver deaths per million registered vehicle years. This is the gold-standard metric used by safety researchers because it normalizes for how many of each vehicle are on the road — unlike raw fatality counts, which would simply reflect sales volume.

What is a "registered vehicle year"?

One vehicle registered for one year = 1 registered vehicle year. If 500,000 Toyota Camrys were registered across the study period, that's 500,000 vehicle years. This normalization means a bestselling car isn't penalized just because millions exist.

Why not normalize by miles driven?

The ideal metric would be deaths per mile driven (VMT — vehicle miles traveled), which would control for the fact that some vehicles are driven more than others. However, per-model VMT data does not exist publicly. The FHWA publishes aggregate VMT by broad vehicle class (cars vs. trucks), but not by specific make/model. IIHS "deaths per million registered vehicle years" is the best available normalized metric at the individual model level. See the NHTSA class-level fatality rates above for VMT-normalized rates at the vehicle-class level.

"Danger to Others" metric

The other-driver death rate counts deaths of drivers in other vehicles struck by the subject vehicle, per million registered vehicle years of the subject vehicle. This reveals how dangerous a vehicle is to people outside it. Very large pickups (Ram 3500: 189) cause dramatically more deaths to other drivers than small cars (Buick Encore: 6).

Data details

Source: Insurance Institute for Highway Safety (IIHS), July 2023
Model year: 2020 (and equivalent models up to 3 years earlier if not substantially redesigned)
Fatality period: Calendar years 2018–2021
Adjustment: Rates are adjusted for driver age and gender
Minimum threshold: 100,000+ registered vehicle years OR 20+ deaths in the study period
Overall average (driver): 38 deaths/million — up from 36 (2017 models) and 28 (2011 models)
Overall average (other-driver): 53 deaths/million
Coverage: This page shows the highest and lowest rated vehicles from the IIHS publication. The full IIHS dataset includes additional vehicles with mid-range rates.

NHTSA FARS national data

The Fatality Analysis Reporting System (FARS) is a census of all fatal motor vehicle crashes in the United States, maintained by NHTSA. Unlike the IIHS per-model data above, FARS covers all crashes nationally and can be normalized by vehicle miles traveled (VMT) — but only at the broad vehicle-class level (passenger cars, light trucks, motorcycles), not per make/model.

VMT data comes from the FHWA Highway Statistics Table VM-1, which estimates total miles driven annually by vehicle type. Dividing FARS fatalities by VMT yields the "fatality rate per 100 million VMT" — the standard metric used in NHTSA Traffic Safety Facts publications.

2024 data is an early estimate based on NHTSA's preliminary projections and is subject to revision.
Fatalities by road user type (2014–2023) are final FARS counts.
Per-class VMT rates use FHWA VM-1 data matched to FARS occupant fatality counts for the corresponding vehicle type.

FARS per-model estimated rates

The FARS per-model section aggregates all occupant fatalities across 2014–2023 from NHTSA FARS bulk CSV downloads, grouped by make/model. Unlike the IIHS data (which covers a single model-year cohort of driver deaths), FARS per-model data includes:

All occupant fatalities (drivers + passengers), not just driver deaths
All model years on the road, not a single MY cohort
All vehicles involved in fatal crashes, regardless of registration volume

Since per-model VMT data does not exist publicly, estimated fatality rates use a proxy method:

Fleet estimate: publicly reported average annual US sales × fleet multiplier (12.5 yr average vehicle age × 0.70 survival discount ≈ 8.75 effective fleet years)
Annual VMT estimate: estimated fleet × NHTS class-average annual miles (sedans: 11,500 mi; SUVs: 12,500 mi; pickups: 13,500 mi; vans: 11,800 mi; sports cars: 8,000 mi)
Rate: 10-year total deaths ÷ (estimated annual VMT × 10 years ÷ 100,000,000)

Key caveats:

Sales figures ≠ registrations — fleet size estimates are approximate
All vehicles within a class are assumed to drive the same annual miles
Does not account for driver demographics, geographic variation, or vehicle age distribution
Includes models with 50+ deaths or significant annual sales (>1k) for rate comparison
IIHS uses proprietary S&P Global Mobility registration counts — our estimates are approximations, not directly comparable

Limitations

IIHS data: Does not control for miles driven per vehicle (VMT). Vehicles driven more miles/year will accumulate more exposure.
Does not separate highway vs. urban driving patterns, which differ by vehicle type.
Specific drivetrain configurations (2WD vs 4WD) are listed where IIHS reported them separately — rates can differ significantly.
Small sample sizes for rare vehicles can produce volatile rates (a single death can swing the rate dramatically).
NHTSA/FARS data: VMT normalization is only available at the vehicle-class level, not per make/model. Class-level rates mask variation within each category.
FHWA VMT estimates are modeled from traffic counts and may not perfectly reflect actual travel.
FARS per-model: Estimated rates depend on sales-as-fleet-proxy assumption. Vehicles with much higher or lower than average usage will have distorted rates.

Sources

IIHS driver death rates → | IIHS press release →
NHTSA FARS database → | NHTSA Traffic Safety Facts →
FHWA Table VM-1 →
FARS bulk CSV downloads → | NHTS (National Household Travel Survey) →