How Does Kirk Cousins Affect the Vikings' Super Bowl Odds?

The Vikings, having demonstrated that they might have one of the best — and certainly one of the most unique — defenses in NFL history, have confronted the reality of their offensive situation.

Since 2000, 379 teams have posted winning records through 9 games. The 2026 Minnesota Vikings rank 379th in offensive success rate. As any national or local commentator will tell you, the situation was “untenable.”¹

Trading for Kirk Cousins — a roster move seemingly made at the last minute after seeing how poorly Carson Wentz performed on Monday Night Football against a weak Bills defense — is quite clearly their attempt to salvage this season.

Will it work? Let’s dive into our available evidence and find out.

So, let’s go over it one more time. How bad were the Vikings at quarterback, and how good was Cousins with the Raiders?

Instant upgrade, right?

Possibly. But bringing in a new quarterback midseason rarely pays off dividends.

How quarterbacks perform when thrust into a new environment midseason is well studied. Carson Palmer forced a trade from the Cincinnati Bengals to the Oakland Raiders, where he put up some abysmal numbers upon arrival before correcting course. Joshua Dobbs did the opposite after being forced into action following his trade to the Minnesota Vikings.

And Vikings fans well remember “the Josh Freeman game” on Monday Night against the New York Giants.

The sample isn’t large, but it is helpful. Joe Flacco has done it twice, and Josh Johnson has done it even more times.

Some quarterbacks (Dobbs, Baker Mayfield, Kyle Orton and Joe Flacco (2025)) immediately put up their best numbers with their new team before fading. Others (Palmer and Jimmy Garoppolo) improved over time.

For the most part, as you’d expect, quarterbacks moved into this new role perform worse in their initial outings than they did over their three prior seasons. As a result, we generally see below-average play.

But we know this isn’t a typical midseason quarterback trade. Many quarterbacks moved partway through the season are moved largely because the acquiring team is desperate and the team moving on doesn’t see much more value in them.

Dobbs wasn’t making Pro Bowls. Flacco, at 40, had left his best days behind him. This isn’t really the case with Cousins, who was putting up top-ten EPA per play numbers, ranking near C.J. Stroud and Patrick Mahomes.

Even Garoppolo, whose two starts with the Patriots gave reason for optimism before his trade to the 49ers, wasn’t expected to play at the level Cousins is playing at now.

So, can we assume Cousins picks up where he left off? He’s familiar with the offense, and the Vikings supporting cast is better than Vegas’ by most estimations. But the offense has systemic issues too; not every problem is built around who is taking snaps at quarterback.

We can play with a couple of assumptions: the first is that Cousins plays at the level he played at in Las Vegas. The second is that he plays exactly like Kyler Murray did. A third assumption midpoints between the two: a rough approximation that acknowledges that there may be some chemistry issues to begin the process and that the Vikings’ systemic issues continue.

The Vikings move from a projected 10-6-1 record and a 70 percent playoff rate to winning between 11 and 13 games with odds ranging between 76 and 97 percent. For the first overall seed, the ranges run between nearly one in two odds, with an ideal scenario of one in twelve when staying the course.

But it’s not that simple. There needs to be another discussion: age. Kirk Cousins is no longer the 30-year-old that signed with the Vikings in 2018. At 38, there’s plenty of reason to worry about his decline, something commentators echoed in the moments after the trade.

Concerningly, there’s some evidence that suggests this age problem doesn’t appear until partway through the season — about the moment Cousins takes his first dropbacks with Minnesota.

Cousins, at 315 attempts this year, is frighteningly close to this cutoff. If it’s meaningful.

This was worth testing independently. The relationship is there, but it's much softer than the strong relationships implied throughout this discussion. 

One problem with this analysis is that it didn’t account for other factors. Not only do all quarterbacks decline as the season goes on, but very few quarterbacks over 35 take snaps for teams, so there’s a sample size issue. On top of that, the older cohort of quarterbacks is better — they have further to fall, and even after the decline, they’re (in aggregate) better than the under-35 group.

Why do quarterbacks generally decline late in seasons? There are probably a dozen reasons. In addition to wear building up over the course of the year, weather turns harsher later on (this analysis, at least for the moment, does not split outdoor and indoor games), rookies hit a “rookie wall,” and defenses might be coming into their own.

Injuries to supporting casts could impact offenses more than they do defenses, too, with more expected rotation and less drop-off for players at cornerback and defensive line than for those at receiver and offensive line.

Below, we can see both cohorts decline, with the younger group of quarterbacks declining gradually from +0.100 EPA per play to about +0.075.

An initial model on this question suggested a drop-off, but there’s an important caveat. Those shaded bands on either side of the lines represent uncertainty (at a 90 percent confidence level). The true value is somewhere in that band, not necessarily on the line.

The data tells us that dropoffs have happened, and it also suggests that reading into those dropoffs might be trivial (and that dropoffs occur around dropback 400, not 350). I also struggle to explain this without feeling like there also needs to be an explanation for why there’s a spike up to +0.170 at dropback 325 or so. Do quarterbacks feel invigorated after 300 passes but get tired after 400?

Even so, a player who was playing at +0.150 EPA per dropback (Cousins was operating at +0.164 after adjusting for opponent) would drop down to +0.100 EPA per dropback — a huge upgrade over what the Vikings have put at quarterback this year and a significant upgrade over the typical quarterback, who drops to +0.075.

Kyler Murray’s opponent-adjusted EPA per dropback was a pitiful -0.069.

One common practice in these conversations was to exclude Tom Brady because he was an outlier. Indeed, once one removes Brady from the sample, the drop-off looks much sharper and starts sooner. But there is just as large an outlier pulling down the average: Brett Favre.

Eliminating him from the sample makes the entire drop-off disappear.

So, let’s eliminate both Brady and Favre.

It looks very dramatic, but the takeaway here is the uncertainty bands, not the median line. The overall performance of plus-35 quarterbacks is lower — instead of skating along +0.150 and peaking at +0.170, it rides along the +0.135 line and peaks at +0.150.

Because the lines intersect, it looks like a steep drop. But in reality, it’s a drop of about 0.040 EPA.

Still, it’s perhaps worth accounting for. On top of that, giving Cousins full credit for his Raiders-level performance without accounting for his other performances in recent seasons might not be good practice. For quarterbacks who switch teams and have a strong start to their season, the evidence suggests that their three-year average prior to the season gives us more information than their performance thus far in the same season.

The true value of their rest-of-season performance happens to combine those levels, and a rough even split between performance thus far and prior-season performance gets us pretty close.

If we incorporate all of that and push in some dropoff later in the season (driven almost entirely by lower depth of target, as completion rate, sack rate, interception rate, and touchdown rate stay even), we can get our best guess at a model.