Getting Hit by a Car Is Twice as Deadly as a Decade Ago. Deaths Fell Anyway.

I ran the numbers, then I ran them again. They didn't get better, but they did get weirder. The Governors Highway Safety Association published its full-year 2025 pedestrian report in July, and the headline is good news: drivers killed a projected 6,732 people on foot last year, down 7 percent, the third straight annual decline.^[1] That is 505 fewer deaths. Champagne, right? Hold the cork, because buried in Part 2 of the same report is a number that should ruin your afternoon: of all the times a car hit a pedestrian in 2024, 10.1 percent were fatal, against 4.0 percent in 2015. The crash got more than twice as deadly while the death toll fell.^[1]

Deaths equal crashes times lethality. That is the whole equation, and it is the only honest way to read this report, so plug in the numbers: in 2015, 5,494 pedestrians died and 4.0 percent of pedestrian crashes were fatal, which implies roughly 137,000 pedestrian crashes. In 2024, 7,080 died at a 10.1 percent fatality rate, implying roughly 70,000 crashes. Read that again: the number of times cars hit people roughly halved in a decade. Each hit got two and a half times more likely to kill. That death toll is the product of two trends sprinting in opposite directions, and the falling one is winning lately. Barely.

As for what made the crash deadlier, the report points at the obvious suspect: the vehicles. Light trucks, meaning SUVs, pickups, and vans, were the striking vehicle in 50.4 percent of 2024 pedestrian deaths versus 40.6 percent for passenger cars.^[1] The gap between those two categories opened in 2019 and widened every year through 2023. Then, for the first time, it narrowed slightly in 2024. One year does not make a trend, but it is the first crack in a six-year wall. The mechanism has a number attached: IIHS research found tall, blunt front ends are associated with a 43.6 percent increase in pedestrian fatality risk compared with low, sloped ones.^[3] When the thing hitting you is shaped like a brick wall and weighs 5,500 pounds, physics does the rest. Pedestrians now account for nearly one in five of all traffic deaths, 18 percent in 2024, a share that kept climbing even as the raw count fell.^[1]

On why there are fewer crashes at all, here the report gets honest in a way I respect: it mostly shrugs. GHSA credits enforcement, technology, education, and infrastructure, the standard four horsemen of the grant application. Some of it checks out, starting with the share of deaths where no sidewalk existed, which fell from 68.3 percent in 2021 to 62.2 percent in 2024; concrete is being poured somewhere.^[1] Hit-and-run deaths dropped 4 percent to 1,669. But the technology story has a hole in it you could drive a Suburban through. Pedestrian automatic emergency braking cuts pedestrian crashes by roughly a quarter in daylight, and it is spreading through the fleet ahead of NHTSA's 2029 mandate.^[5] Problem: 76.5 percent of pedestrian deaths happen after dark, where the systems are far less effective.^[1] The tech cannot explain most of the improvement. Something else is moving the crash count, and nobody has published what.

Time for the paragraph where I argue with myself, because the lethality number deserves an interrogation. The 4.0-to-10.1 percent series comes from CRSS, which estimates total crashes from sampled police reports.^[1] Deaths are counted well; fender-benders are not. If police respond to fewer minor crashes than they did in 2015, and there is good reason to think they do, the denominator shrinks while the numerator holds, and the fatal share rises on paper without a single car getting deadlier. GHSA itself calls its vehicle-size explanation a "possible reason," not a finding. The honest read: some of the doubling is real, anchored by the IIHS front-end data and the light-truck share, and some of it is almost certainly a reporting artifact, though the split between those two is unknown. Anyone who tells you otherwise is selling something.

What is not ambiguous is the geography: California cut pedestrian deaths from 1,067 to 828, a 239-life reduction that accounts for nearly half the entire national decline.^[1] North Carolina went the other way: 289 to 383, up 94, the largest increase in the country. Twenty-eight states improved and seventeen got worse, while the national rate fell below 2.0 per 100,000 for the first time since before the pandemic, landing at 1.97.^[1] And the economic toll of the 2025 deaths is nearly $94 billion, a number so large it stops meaning anything, which is itself a kind of meaning.

Here is what you do with this: if you walk, the night owns the risk: more than three quarters of deaths happen after dark, so reflective gear and lit routes are the highest-leverage personal intervention in the entire report. If you are buying a car, pedestrian AEB is worth having and effectively standard on new vehicles, but know its limits after sunset. If you are shaped like a policymaker, the question is California's: one state delivered half the country's progress while North Carolina added 94 bodies. Figure out what Sacramento did and copy it before your state becomes the next Raleigh. And if anyone tells you the pedestrian crisis is over because deaths fell three years running, show them the lethality curve. The crisis did not end; it consolidated into fewer crashes, deadlier ones, and a decade-long trend that says the car hitting you matters more every year.

Sources & References

1. Governors Highway Safety Association, Pedestrian Traffic Fatalities by State: 2025 Preliminary Data, July 2026. ghsa.org (PDF)
2. GHSA press release, “U.S. Pedestrian Deaths Fall for Third Straight Year,” July 14, 2026. ghsa.org
3. Hu, W., Monfort, S. S., & Cicchino, J. B., “The association between passenger-vehicle front-end profiles and pedestrian injury severity in motor vehicle crashes,” IIHS, September 2024. iihs.org
4. NHTSA, Fatality Analysis Reporting System (FARS) and Crash Report Sampling System (CRSS). nhtsa.gov
5. IIHS, pedestrian crash avoidance research: systems with pedestrian detection reduce pedestrian crash rates by roughly 27%. iihs.org