For a decade the NFL’s go-to answer to CTE research was a methods complaint: of course the donor rate looks catastrophic, only families who already suspect something is wrong sign the papers to hand over a brain. It was a real criticism. Researchers at Boston University built a study specifically to kill it, and the number that came back was still catastrophic. The NFL’s response to the CTE study never changes, whether the methodology behind it was thin or, like this one, built to withstand the league’s own objection.
Start with the raw number. Boston University reported that researchers examined 387 donated brains of former players who died between 2008 and 2023, and found CTE in 362 of them, nearly 94 percent. ESPN put the pool of donated brains at 396 and landed on the same nearly-94-percent rate. That’s the number the league has waved off for a decade, because a brain donor isn’t a random sample of dead football players. Families who watched a father or husband lose himself to confusion and rage are the ones who sign the forms. The rest of the dead just get buried, uncounted.
So the researchers went and found the hole themselves. For the other 1,966 players in that same 2,362-player window who never donated anything, they pulled National Death Index records and found that players with documented cognitive symptoms were roughly five times more likely to have had a family member donate their brain than players without symptoms. That’s selection bias, quantified, which is the exact homework assignment the league insisted nobody had bothered to do.
Correct for that bias across the full population and the number drops to 74.5 percent. Call it three in four deceased players had CTE, because the math landed there whether you round up or not. That’s a real twenty-point drop from the raw donor rate, and it dropped exactly because the correction worked the way it was designed to. Nobody buried a flaw here. The NFL’s own argument got run through the data it demanded, and the data still said catastrophic.
None of that means three in four guys suiting up this Sunday are walking around with an undiagnosed brain disease. This is a study of men who already died, a population weighted toward families who suspected something before anyone offered them a consent form. It doesn’t predict who’s playing now or who develops CTE later. That’s in the study and in every outlet’s coverage, and it’s the entire reason the league built its defense on self-selection in the first place.
The lead author, Julia Gallini, is a BU biostatistics PhD candidate working with Mass General Brigham and the Concussion and CTE Foundation on NIH money, which is a lot of rigor to throw at a problem the league keeps describing as a sampling issue. Matt Hasselbeck, the three-time Pro Bowl quarterback who serves as an ambassador for BU’s related Diagnose CTE study, called the findings concerning: “we have to learn from this data to make the game safer.” That’s about as measured as a response gets. He’s not calling for anyone’s job. He’s asking the league to read its own study.
https://x.com/AdamSchefter/status/2115745361973420399
The league’s actual on-record answer was that it “continuously strives to make the game of football safer, including by implementing strategies to reduce concussions and head impacts.” It cited Guardian Caps. It cited rule changes. It pointed former players toward what it called “a robust, and expanding, set of resources” for health concerns, as if several hundred rotted brains were a benefits enrollment problem instead of exactly the pattern the league spent a decade insisting was sampling bullshit. This is the same paragraph the NFL has been sending reporters since the first wave of donor studies, word for word, regardless of whether the study behind it was sloppy or, like this one, engineered specifically to survive the league’s own stated objection.
That’s the tell the raw percentage can’t give you on its own. A number can be wrong, overstated, badly sampled, and the NFL knows exactly how to say so in a press release. A number can also be bulletproof, built to answer the league’s own complaint on its own terms, and the league reaches for the identical safety paragraph anyway. The response isn’t calibrated to the evidence. It’s calibrated to the calendar, because the games are on Sunday either way.
Philadelphia just watched a smaller version of the same reflex up close, when the Eagles knew about Jalen Hurts’ blood clot and played him anyway, and Sirianni’s answer never changes either. Different building, same drawer the statement comes out of.
Gallini and her colleagues did the NFL’s homework for it, closed the exact loophole the league spent ten years hiding behind, and handed back a number that should have ended the argument on the spot. The league filed it next to the last one. Same paragraph, same drawer, same Sunday.
