The NSC Just Published Its Memorial Day Body Count Forecast. It’s 393.

The National Safety Council released its Memorial Day 2026 motor vehicle fatality estimate this week, and the number lands with the blunt precision of an insurance actuary filling in a spreadsheet she has filled in every May for thirty years: 393 dead.^[1] That covers the 3.25-day window from 6 p.m. tonight through 11:59 p.m. Monday. A 90% confidence interval of 330 to 461 means the NSC is telling you, with near-clinical certainty, that somewhere between a packed regional jet and a sold-out Amtrak Acela will not survive a long weekend built around backyard grills and mattress sales.

The estimate is 11% below last year’s 443-death forecast. Celebrate that, if you want. It still means nearly six people will die every hour from Friday evening through Monday night, and the NSC’s own track record shows these numbers are not hypothetical: their 2024 Memorial Day estimate undershot the actual body count by 7.3%.^[1]

What makes this number interesting is not the number itself but the granularity we can apply to it using three decades of prior data. A peer-reviewed analysis of FARS records from 1981 to 2016 calculated that Memorial Day weekend carries 9% higher odds of dying in a crash compared to an equivalent non-holiday period, after controlling for seasonality and day of week.^[2] Averaged over 36 years, the holiday logged 142 deaths per day versus 130 on control days. Saturday accumulates the highest raw count, roughly 40% of the weekend total, which would put Saturday’s allocation at approximately 157 of this year’s forecast. But Monday carries the steepest per-day risk increase at 18% above normal. Researchers attribute Monday’s spike to end-of-weekend fatigue, longer return trips, and the hangover coefficient nobody puts in the regression model but everyone understands.

Alcohol math on Memorial Day is ugly and consistent across every year in the NHTSA archive. Nationally, alcohol-impaired driving accounts for 30% of all traffic fatalities in a given year.^[3] During Memorial Day 2024, that figure jumped to 38%. Apply 38% to the 393 forecast and you get approximately 149 deaths this weekend where the highest blood-alcohol concentration among involved drivers or riders was at or above 0.08 g/dL. For context, our own FARS toxicology dataset (2014–2023) shows a baseline alcohol-positive rate of 15.1% among drivers in fatal crashes.^[4] Memorial Day more than doubles that rate. Memorial Day is not more dangerous because more people are driving; it is more dangerous because a disproportionate fraction of them are drunk.

Then there are the seatbelts, or the absence of them. According to the NSC, 147 lives will be saved this weekend by people who do buckle up, and an additional 96 lives could be saved if every vehicle occupant wore one. That second number represents people who will actually die between now and Monday night, whose deaths would be prevented by a mechanism invented in 1959 and federally mandated since 1968.^[1] Among fatally injured vehicle occupants nationally, 54% were not wearing a seatbelt at the time of death.^[1] The overlap between unbuckled and impaired is not a coincidence. Unbuckled occupants in alcohol-involved fatal crashes outnumber their buckled counterparts by roughly two to one.

A separate BMC analysis surfaced another pattern worth noting as you pack the minivan: passengers face 17% higher odds of dying in a Memorial Day crash compared to control weekends, versus only 5% for drivers.^[2] The researchers speculate this reflects family travel, with more passengers per vehicle making more trips on unfamiliar roads. Rural areas show a 14% increase in holiday fatality odds versus 8% for urban zones, consistent with longer-distance travel to destinations that are, by definition, farther from trauma centers.

None of this is new information in any meaningful sense. NSC has published Memorial Day forecasts since at least the 1970s. The patterns are identical year after year: alcohol involvement spikes, seatbelt non-use concentrates among the dead, Saturday claims the most bodies while Monday claims the highest per-day risk, and rural corridors absorb more than their demographic share. A dead-accurate actuarial model of holiday carnage apparently has no deterrent value whatsoever. In 2024, the actual exceeded the estimate. This year’s 11% decline from 2025 mostly reflects the broader national drop in fatalities rather than any holiday-specific improvement.^[5]

Meanwhile, this weekend also marks the unofficial start of what AAA calls the “100 Deadliest Days” for teen drivers, the stretch from Memorial Day to Labor Day that has averaged 8 teen driver-related deaths per day over the past five years.^[6] AAA’s data shows speeding (28%), drinking (17%), and distraction (9%) as the leading contributing factors. Those percentages, applied to the 393, sketch a weekend where the machines on the road are the least dangerous component of the system.