In the aftermath of any close and controversial Vikings loss, there is a lot of ALL-CAPS typing and second-guessing. Even players are left to lament the razor-thin differences between wins and losses.
“One more play, one more yard, one more stop, things like that,” wide receiver Adam Thielen said after the Vikings’ crushing 27-26, last-minute loss at Seattle on Sunday night. “It’s just we’re so close, and that’s probably why it’s so disappointing.”
In a game with so many twists, turns and moments, it can be hard to separate decisions from outcomes. Indeed, this is one of the most challenging thing about Monday morning quarterbacking, so to speak: Are you second-guessing the thought that went into a play … or the execution of the play … or are you just mad that it didn’t work in one particular instance?
In times like this, it can be helpful to be armed with data. Numbers can at least help clarify what went right or wrong because — unlike a fan base with 60 years of baggage and perhaps a few adult beverages in them late on a Sunday — they don’t have feelings.
And hey, maybe you’ll feel like yelling at me instead when I say this: I think the data suggests that Mike Zimmer and his staff made two correct decisions at two critical moments Sunday.
1) Going for a two-point conversion after cutting the deficit to 21-19 with 3:39 left in the third quarter and 2) the bigger (but related) decision to go for a first down on 4th-and-1 from Seattle’s 6 when they were up 26-21 with 2 minutes left.
We discussed both things on our postgame video, and both are great fodder for debate. In the calmer light of day, though, here’s what math and win probability data tells us:
*The conventional wisdom I have stuck in my head is that a team shouldn’t “chase the points” — i.e. go for 2 — after cutting a deficit to two points until the fourth quarter. So I initially thought it was too early for the Vikings to try it even though it was late in the third quarter.
As it turns out: Per FiveThirtyEight, data suggests, “When a team is down 2, it should go for 2 — pretty much any time, but especially in the third quarter and beyond.”
In general when a team is down 2 points, going for 1 and making it only increases win probability by 1.8% while going for 2 and making it increases win probability by 8.4%. But what happens if they miss it?
Pro Football Reference lets us get even more specific with the situation. Plugging in all the scenarios to the win probability calculator — including that Seattle was a 7-point favorite — shows the Vikings had a 26.2% chance to win after missing the conversion. They would have had a 30.7% chance to win had they successfully kicked to cut it to 21-20. And a 35.6% chance to win if the two-point try was successful.
It’s a thin margin, but the upside of going for 2 and making it increased their win probability by 4.9% over the expected safe outcome of kicking, and missing it only decreased it by 4.5%. And that’s assuming they would have made the kick — likely, though not a given on a rainy night in Seattle. It was a smart decision based on data.
*That said, the play the Vikings tried — culminating in Kirk Cousins trying to run the ball in out of a shotgun snap and an empty backfield — was not ideal. Cousins has some mobility, but he’s not Russell Wilson. Perhaps something with motion and a rollout, which play to Cousins’ strengths, would have yielded a better result.
And it should be noted that the decision to go for 2 directly impacted the second decision we need to talk about. Had the Vikings gone for 1 and made it, then the rest of the game played out the same, they would have been ahead 27-21 while facing that fourth down at the 2 minute warning.
A field goal at that point would have put the Vikings up by 9 — putting their win probability in the high 90s and forcing the Seahawks to score twice, plus convert an on-side kick in between, to pull the game out.
Instead, of course, Zimmer and the Vikings faced a real dilemma: Kick a field goal and force Seattle to go (in all likelihood) 75 yards for a touchdown and two-point conversion just to tie … or try to make half a yard to essentially seal the game.
Here again we visit the Pro Football Reference win probability calculator and find:
The Vikings’ win probability as they faced 4th-and-1 from Seattle’s 6 with 2 minutes left and a five point lead was 90%.
A field goal would have decreased that slightly to 89.9%, assuming Seattle got the ball at its own 25 down 8 with just under 2 minutes left.
Going for it and making it would have increased the probability to 99%, while getting stopped (the outcome that happened) decreased that to 77.9%.
Given that NFL teams convert roughly 2 of every 3 tries on fourth-and-1 into first downs, the math again was on the Vikings’ side. There was a good chance they were going to essentially seal the game by making a yard, especially since Seattle had just one timeout left.
And that was the only scenario by which they weren’t going to give the ball back to the comeback king and possible future Hall of Famer Wilson, at least with any meaningful chance to win or tie. Maybe if they’re facing more of a Ryan Tannehill type of QB, taking the points would have been a more arguable position.
But regardless of opponent, going for it was the data-driver correct move.
And the yard was there to get. If Alexander Mattison cuts right instead of running into a plugged middle, Zimmer is being lauded today for his guts instead of being lambasted.
Ah, but there’s that emotion again. We were trying to shoo that away to recognize this: Lamenting outcomes is natural. Questioning execution is fair. But second-guessing the decisions?
In two key instances Sunday, Zimmer was right even if both times everything went wrong.