IIHS Gave Out 63 Awards. The Vehicles Killing You Didn't Qualify.

IIHS published its toughest award criteria in history this year: mandatory standard AEB, good rear-seat moderate overlap for both TSP and TSP+, and a brand-new vehicle-to-vehicle crash prevention gauntlet at 31, 37, and 43 mph.^[1] Sixty-three vehicles passed. We cross-referenced every winner against a decade of NHTSA FARS fatality data, and the result was striking: the lab finally agrees with the road.

Forty-seven of those 63 winners are SUVs or crossovers.^[1] In the FARS database covering 2014 through 2023, SUVs as a class carry a fatality rate of 0.63 deaths per 100 million vehicle miles traveled, the lowest of any major vehicle category, and an occupant lethality ratio of 0.524.^[2] Sedans, at 1.02 per 100M VMT and 0.645 lethality, collected 14 awards. Pickups, sitting at 1.09 per 100M VMT despite their mass advantage, managed exactly two: a Toyota Tundra crew cab and a Tesla Cybertruck, whose FARS track record is too short to measure.

Minicars, minivans, and small pickups each earned exactly zero awards across both TSP and TSP+ tiers.

IIHS President David Harkey called the minivan results "disappointing," noting persistent rear-seat protection failures.^[1] Disappointing undersells it considerably; the van class carries a 0.63 fatality rate per 100M VMT in FARS aggregate, but that number is pulled down by commercial vans that rarely carry children in the second row, and the specific structural weakness IIHS flagged is moderate overlap protection in exactly the seats where families put their kids. So a vehicle class marketed as the safest family hauler on the road cannot pass a test designed to protect families.

We computed the convergence directly: of the vehicle classes that carry 75.7% of all FARS deaths from 2014 to 2023 (sedans, pickups, minivans, minicars), only 25.4% of IIHS 2026 awards went their direction.^[2] Meanwhile, SUVs accounted for 24.3% of total FARS deaths and took home nearly three-quarters of the awards. For the first time, the IIHS criteria function almost like a FARS death-rate prediction model, sorting vehicles into the same hierarchy the highway has been building with actual corpses for ten years running.

Pickup buyers should pay close attention to what the two-out-of-sixty-three number actually means: the F-150, Silverado, Sierra, and Ram, which together account for 26,529 FARS deaths over the study period, all failed to qualify.^[2] Ford's Ranger, with a rate of 2.91 per 100M VMT and 3,089 deaths, is in a class that received zero awards at any level. IIHS has essentially declared that the four bestselling vehicles in America do not meet its 2026 safety floor.

What to do with this: If you are shopping for a vehicle right now, start with the 2026 TSP+ list and work backward. A Kia K4 at $22,290 passed the same battery of tests that eliminated every full-size American pickup.^[1] If you drive a minivan and have kids in the back seat, check whether your model's rear-seat moderate overlap rating is "Good" at iihs.org/ratings. If it is not, understand that IIHS has now made that deficiency a disqualifying condition at both award levels, because the crash data justified it.

Strongest counterargument: This convergence might just be a proxy for vehicle age. Modern SUVs dominate IIHS awards because they are new, heavy, and packed with standard safety tech. They show low FARS rates for the same reasons, plus their limited exposure time in the 2014–2023 dataset means they have not accumulated enough miles to build proportional death counts. A 2003 Corolla does not fail IIHS 2026 criteria because Corollas are inherently dangerous; it fails because it was designed 23 years ago. But this counterargument does not explain the minivan gap: the 2025 Honda Odyssey and 2025 Chrysler Pacifica are current-model-year vehicles that still cannot pass the rear-seat test, which means age alone is not the whole story.

Limitations: FARS data covers 2014–2023 and aggregates all model years within a nameplate, so fleet age composition distorts rate comparisons, while IIHS tests current model-year vehicles only and cannot be mapped one-to-one against FARS denominators. We cannot directly compare model-for-model because FARS does not isolate individual model years with VMT denominators. Van class FARS rates blend commercial vans with consumer minivans, likely depressing the combined figure below where minivans sit alone. Estimated rates use VMT approximations, not odometer readings, introducing ±15% uncertainty for low-volume models.