Bills 24, Chargers 16: The Field Called the Winner Right, Pete Prisco Nearly Called the Number
Buffalo beat Los Angeles 24-16 in Week 3, and the pre-game record on this one wasn't close. We tracked 19 graded calls on this game — 15 straight winner picks and four exact-score predictions — and the field went 16-for-19. Every pundit who called the score got the winner right too, and the only misses were the three who bet on the Chargers.
That's not a coin-flip result to argue over. Buffalo was the favorite in every version of this pick, and 16 of 19 people simply took the favorite. What's worth reading closely is the four who put a number on it, and the three who didn't just take the chalk.
The score calls: close, but not quite close enough
Four pundits went on record with an exact score, and every one of them had Buffalo winning by more than the eight points it actually was.
- Pete Prisco had it "Bills 32, Chargers 23" — a nine-point margin, one off the actual eight. That's the tightest of the four, and it graded out at 97.62 accuracy and a skill score of 29.39, the best mark anyone earned on this game.
- John Breech and the SportsLine model, relayed by Matt Severance, both had Buffalo by 10 — Breech's "Bills 34-24 over Chargers" and Severance's "The SportsLine Projection Model has Buffalo winning 31-21" — each worth 95.24 accuracy and a 24.89 skill score.
- Tyler Sullivan was furthest off among the four, projecting "Bills 31, Chargers 20" — an 11-point spread — for 92.86 accuracy and a 20.39 skill score.
Straight winner picks on Buffalo earned nothing extra here — the site's baseline already assumes the home favorite, so calling BUF to win outright is a shrug, not a skill. The score predictions are where the real signal is, since they're graded against how close the margin actually was, and Prisco's was the closest anyone came.
The three who took Los Angeles
Three pundits bet against the field and lost, all with the same call: the Chargers upset the Bills.
Eric Mangini put it plainly in a taped upset pick: "I'm taking the chargers over the Buffalo Bills... So Chargers over the Bills." It graded at 0 accuracy and a skill score of -189. That's not a fluke for Mangini this season — his overall record sits at 27.25 accuracy across 18 graded calls, with his game-winner picks specifically running 34 accuracy over 10 graded. This miss extended a pattern rather than breaking one.
Eric Williams went further, framing it as a trend reversal: "get ready for the Chargers to do Chargers things on the road against the best team in the NFL in the Buffalo Bills, right? Nope, when everyone expects them to fall on their face, these are the moments the Chargers rise up." That one also graded 0 accuracy, -189 skill. For Williams, though, this looks more like the exception — his winner-pick record otherwise runs 2-for-3 (66.67 accuracy) on a small sample.
David Droegemeier made the call earliest, on September 8, nearly three weeks out: "I feel like the chargers can beat the bills, and I feel like they will. I think that's gonna be a a big shot game." It's his only graded game-winner pick on the season, and it came back at 0 accuracy and -200 skill — the maximum penalty the site hands out, reflecting both the wrong pick and the extra lead time he had to change his mind.
What it touches beyond this game
The Chargers' rough week fits a rough season: game-winner picks about Los Angeles are running just 24.2 accuracy across 62 graded calls this year, easily the shakiest number for either club. Buffalo's are at 71.5 accuracy across 117 — a big gap, and one that makes every remaining Chargers upset pick a harder sell and every Bills favorite pick an easier one.
For the full ledger on either team, see the Buffalo Bills and Los Angeles Chargers pages, and check where this week's callers land on the leaderboard.