Photo: Matthew Huang/Icon Sportswire
If you ask around at Vegas sportsbooks – or any other legally sanctioned sportsbook, for that matter – they’ll tell you that the Seattle Seahawks have about a 70% chance of winning the Super Bowl this Sunday. I might have a few more coins clinking around in my pocket if I were up to the task of figuring out exactly how they got that number, but I’d imagine the possibility of a Sam Darnold implosion factors in somewhere.
The wisdom of the herd comes with a big grain of salt, but the consensus win condition for the Patriots goes something like this: Drake Maye plays well against a Seahawks defense that has a bit more firepower than him and his comrades, and Darnold turns into a pumpkin. If one team’s quarterback plays well while the other team’s doesn’t, the better QB wins – cutting-edge analysis, to be sure. But, what if we’re underrating the ability of Seattle to withstand Darnold laying an egg?
If you look at games in which one quarterback, but not the other, has a net negative EPA, the results are pretty dismal. When quarterbacks go into the red, their teams have historically won just 8% of the time. That obviously gets worse the further negative you go, and no team in our database (2015-present) has ever won if their quarterback netted -15 EPA or worse.
Darnold is actually pretty remarkable in this regard. Of the 26 quarterbacks who have played 40+ games since 2022, he is one of just three quarterbacks to have a winning record in games in which he had a net negative passing EPA, and his win rate is easily the best at 69% (the average is 27%). It would be absurd to suggest that he has some special quality that allows his team to win in spite of him, but it is nevertheless interesting.
More interesting is that, despite his reputation for turning into a pumpkin, it doesn’t happen very often relative to the league-average rate. He does, however, sport the third-lowest EPA per dropback in bad games (-0.27) among the aforementioned group. In Darnold’s case, it’s ‘quality’ – if you can even call it that – over quantity that feeds his reputation.
This year, the Seahawks have gone 4-2 despite such performances by Darnold, including wins over the Texans in Week 7 and the Rams in Week 16 in which Darnold registered -14 EPA both times. Regression to the mean seems more likely than not, but it’s still worth looking at.
In the wins against Houston, Minnesota (Week 13), and Carolina (Week 17), Seattle’s defense had three of the fifteen best single-game defensive performances by EPA in the NFL this year. That alone seems enough to carry them, but let’s keep in mind that the Vikings and the Panthers were two of the worst passing offenses in the league, and while the Texans were more respectable, all three turned in some of the worst single-game passing performances in those contests. Seattle also held Minnesota and Carolina to 0 and 10 points respectively despite each starting one of their drives in Seattle territory.
The Week 16 win over the Rams is the outlier here. It was just one of just two games this year in which a team that had a 99+% win probability at the start of the 4th quarter ended up losing. This is also the only game in which it might be said that the run game carried the Seahawks (+6 EPA), as it was a dismal defensive performance from them.
The losses are perhaps more instructive. They lost to the Rams in Week 11 by just two points in spite of the facts that Sam Darnold threw four interceptions that game and that Los Angeles began four drives in plus territory. Not coincidentally, that was easily the Rams’ worst passing performance by net EPA this season. We’re pretty far removed from Week 1, but the defense also fared relatively poorly in that game against the 49ers as well. Both these games were decided by a single score.
The Texans win and the overtime win over the Rams were also decided by a single score, and win-loss record in one-score games infamously tends toward 50%, at least year-over-year. But these are still one-score games that were close in spite of poor play from Darnold, which is the point.
All the discussion about Darnold coming undone in the final hour as a key to a Pats victory has ignored the fact that a defense that ranks 1st in EPA/play allowed against the run and 5th against the pass, has carried them when Darnold does wet the bed. They’ve kept it close, even in the losses, and it’s been harder to break them than it has been to break Darnold.
Perhaps it’s better said that a Patriots win might require a bad showing from Darnold and a lackluster effort from the defense, and that’s much, much more daunting if you’re a New England fan.
