When you look up a car's safety rating, you'll almost certainly encounter NHTSA stars. They are the most visible government safety data for new vehicles in the United States. But "5 stars" is frequently misread as meaning "the safest possible" — when it actually means "performed well in this specific test configuration." This guide explains what NHTSA ratings measure, how they are structured, where they fall short, and how to use them alongside other data.
NHTSA — the National Highway Traffic Safety Administration — is a U.S. federal agency within the Department of Transportation. Its New Car Assessment Program (NCAP), established in 1978, tests new passenger vehicles to give consumers comparable safety information across models. Unlike IIHS, which is a private insurance-industry organization, NHTSA is a government agency. Its crash test data is public domain — freely available at api.nhtsa.gov and nhtsa.gov/ratings.
NHTSA produces star ratings across three separate crash configurations. Each is its own test; the overall star rating reflects the worst-performing individual area:
A 5-star overall rating means the vehicle performed well across all three test categories. For frontal and side tests, it means the dummy measurements stayed below federal injury thresholds with enough margin to qualify for the top tier. This is a real and meaningful achievement — a vehicle with fewer stars performed measurably worse.
What it doesn't mean: identical protection to every other 5-star vehicle. Within the 5-star frontal tier, one vehicle may have a head injury reading of 310 (well below the 700 threshold) while another reads 645 (still within 5-star range, but much closer to the limit). The star looks identical; the measurement does not. This is the gap SafeScore was built to make visible.
NHTSA and IIHS both measure crash performance, but they are not the same evaluation. IIHS runs different test configurations — including the small-overlap frontal test that NHTSA doesn't use, a more demanding side barrier test introduced in 2023, and separate headlight and vehicle-to-vehicle compatibility assessments. IIHS uses Good/Acceptable/Marginal/Poor ratings rather than stars. A vehicle that earns top marks from both organizations has been validated across multiple methodologies with different structural demands. A vehicle that earns 5 NHTSA stars but a Marginal IIHS small-overlap rating has passed one test and not another — both pieces of information matter.
SafeScore is not a replacement for NHTSA star ratings — it is a resolution layer within the frontal crash results. NHTSA frontal stars tell you whether a vehicle passed the threshold; SafeScore tells you by how much. A vehicle with a 5-star frontal rating and a SafeScore of 78 left more injury margin than a vehicle with the same star rating and a SafeScore of 57. Use stars as your entry filter, SafeScore to differentiate within the top tier.
Independence: SafeCarCompare is independent and is not affiliated with or endorsed by NHTSA. SafeScore is calculated from NHTSA public-domain data and is not issued by NHTSA. All underlying measurements are verifiable at api.nhtsa.gov.
SafeCarCompare shows injury margins from NHTSA crash-test data — beyond star ratings. Enter any two vehicles to see head, chest, and neck injury margins side by side.