Car Safety Ratings Explained: NHTSA vs IIHS

A car safety rating is a standardized score that shows how well a vehicle protects its occupants in a crash — and, increasingly, how well it avoids crashes in the first place. In the U.S., two organizations publish them: NHTSA rates crashworthiness from 1 to 5 stars, and the IIHS rates it Good, Acceptable, Marginal, or Poor.

The terms get used loosely — "crash rating," "collision rating," and "safety rating" all refer to the same idea: results from controlled crash tests, scored on a published scale so you can compare one car against another. They are not opinions or reviews; they come from physically crashing vehicles in a lab and measuring the outcome. The rest of this guide breaks down who runs each test, what the scores mean, and how to read the two scales together.

Two independent bodies: NHTSA, a government agency that scores crash protection on a 1–5 star scale, and the IIHS, an insurer-funded nonprofit that rates vehicles Good, Acceptable, Marginal, or Poor on tougher tests.

Because they test different scenarios on different scales, their results don't always line up — and neither one alone tells you the whole story. The sections below break down what each measures, followed by how to read them together.

NHTSA is the government's 5-star baseline; IIHS runs tougher, more varied tests on a Good–Poor scale. The table below shows exactly where they differ.

The short version: a 5-star NHTSA rating tells you a car is fundamentally sound, while IIHS results — especially the small overlap front test and the Top Safety Pick awards — tell you how it stacks up against the best on the market.

NHTSA crash-tests frontal impact, side impact, and rollover resistance, then combines them into a 1–5 star overall rating. Five stars is the top score.

The frontal test uses a full-width barrier at 35 mph, side impact uses a moving barrier and a pole, and rollover is a tipping-stability calculation. The overall rating is a weighted average. Because nearly every new vehicle now scores four or five stars, the star rating tells you a car clears a high bar — but it doesn't do much to separate the very best from the merely good.

One limit to keep in mind: the full-width frontal test is good at catching problems when the entire front of the car hits something, but it does not reveal how the structure holds up in a partial-overlap crash — where only one corner strikes a tree or pole, which is common in the real world. Stars also measure crash protection only, not a car's ability to avoid the crash in the first place.

The IIHS rates vehicles Good, Acceptable, Marginal, or Poor across a tougher and broader set of tests, so its results separate strong designs more clearly than star ratings do.

Its crashworthiness tests are more demanding than NHTSA's, which is why a 5-star NHTSA vehicle can come back Marginal at IIHS. Here is what each IIHS test actually simulates:

On top of crashworthiness, IIHS separately rates headlights and crash-avoidance systems, including pedestrian-detecting automatic emergency braking — areas NHTSA's star rating doesn't cover at all. MotiveGrid scores that crash-avoidance equipment in a separate driver assistance pillar.

Whatever the scale, every crash test comes down to three measurements: how much the structure deforms, what the crash-test dummies record, and how well the restraints keep occupants in place.

1. Structure and intrusion. After the impact, engineers measure how far the cabin collapsed at points like the footwell, brake pedal, A-pillar, steering column, and door sill. Less intrusion means more survival space. This is the crux of the small overlap front test: because only 25% of the front strikes the barrier, the crash bypasses the main crumple structures — the frame rails designed to absorb energy — and drives force straight into the wheel, suspension, and firewall, which can be shoved back into the footwell and crush the driver's legs. A vehicle earns a Good rating only if its safety cage stays largely intact and the wheel doesn't intrude.

2. Dummy injury measures. Instrumented dummies record the forces on the head, neck, chest, pelvis, and legs. Those readings map to real injury risk — head acceleration to concussion and brain-injury likelihood, chest compression to rib and organ damage, femur load to leg fractures. NHTSA converts these injury probabilities into a relative-risk figure that drives the star count; IIHS folds them into each test's Good–Poor rating.

3. Restraint kinematics — does the dummy stay put? High-speed cameras track whether the head stays on the airbag, whether the seatbelt holds the torso, and whether the dummy "submarines" — slides forward under the lap belt. Poor restraint behavior is exactly why IIHS added a rear-seat dummy to the moderate overlap test: back-seat passengers, often children, were being protected less well than those up front, and the updated test now penalizes that directly.

Two more ratings round out the picture. Roof strength is measured by pressing a plate onto the roof until it crushes — a vehicle must hold at least four times its own weight to earn Good, the margin that matters in a rollover. Headlights are scored on a dark track by how far usable light actually reaches on straight and curved roads, with penalties for glare into oncoming traffic — a real-world factor most buyers never think to check. Every one of these results appears, per agency, in the safety section of each vehicle page.

Both awards now require good ratings in the toughest crash and crash-avoidance tests; the difference between them comes down to how well the vehicle protects back-seat passengers.

Front crash prevention and headlights used to be where models fell short — now they are essentially table stakes. What separates the two awards is the updated moderate overlap front test, which IIHS revised to add a rear-seat dummy: an acceptable rating earns the base Top Safety Pick, and a good rating earns the higher Top Safety Pick+. Because the test now rewards rear-passenger protection, the 2025 criteria sharply cut the field — from 71 qualifying models to 48 at the same point a year earlier.

IIHS tightens these requirements almost every year, and headlight and crash-prevention results can vary by trim and option package — so always check the award for the specific model year and trim you are considering.

The safest pick does well on both scales. Treat a strong NHTSA star rating as the baseline and IIHS Good ratings — ideally a current Top Safety Pick+ — as the tiebreaker between good cars.

MotiveGrid's safety score combines crash protection (NHTSA and IIHS structural ratings) with crash avoidance (the standard active-safety equipment a vehicle carries), into one number per trim.

Vehicles that haven't yet been crash-tested by either agency receive a "Provisional" badge and are capped at 80 out of 100, so a newly launched model can't post a full score before independent testing confirms how its structure performs. You can see how every model stacks up on the safest vehicles ranking, or read the full safety methodology for how the score is built and which factors carry the most weight. You can also compare any two vehicles' safety side by side, or browse the safest cars by class.