Giants vs. Vikings: What the analytics tell us about the Giants’ victory

The New York Giants went on the road to play the Minnesota Vikings in the Wild Card Round of the 2023 NFC playoffs and came home Sunday with a 31-24 win.

This was the first time all season the Vikings lost at home, as well as the first time they weren’t able to complete a fourth-quarter comeback and win a one-score game.

So how did the Giants pull it off? The surface-level analysis is that the Giants outscored the Vikings in a shoot-out, capitalizing on their poor defense and mistakes over the course of the game.

However, the advanced stats suggest quite a bit more context than that — and that the Giants might be in greater danger of losing a valuable coach than we think.

Plays of the game

The Vikings started out with the advantage this game, coming out as (not quite) home-field favorites and sprinting down the field for a touchdown. However, the Giants’ 14-point first quarter and 20-play, 10-minute drive to dominate the second quarter pretty much put paid to that. But despite the Giants’ offensive success, the biggest play of the game belonged to the Vikings.

At the very end of the third quarter we see the Vikings (briefly) regain the advantage in win probability. That dip into purple represents what was the single biggest play of the game in terms of both expected points and win probability added.

That was Kirk Cousins’ pass to TE T.J. Hockenson on fourth-and-2 with 32 seconds remaining in the quarter. It was an incredibly gutsy call by the Vikings, but it worked out to be worth +3.8 EPA and +16 points of win probability (43 percent to 59 percent).

The Giants’ biggest play in terms of EPA was Daniel Jones’ 47-yard pass to Darius Slayton on first-and-10 with 3:06 to go in the first quarter. That pass was worth +3.0 EPA and got the Giants to the Vikings’ 38-yard line and set up their second touchdown of the quarter. That play was also the third-biggest play of the game in terms of win probability added.

The Giants’ biggest play in terms of win probability was Jones’ QB sneak on fourth-and-1 at the Vikings’ 5-yard line with 8:38 to go in the fourth quarter. That added 11 points to the Giants’ win probability, taking them up to a 70 percent chance to win the game.

Mike Kafka’s beautiful game plan

It’s no secret that the Minnesota Vikings have an atrocious defense. After doing a pair of deep dives into their defense this year, I can only think that it is some kind of miracle that they managed to win as many games as they did.

Let’s put it this way: The Vikings defense gave up an average of 5.9 yards per play this season. For context, the Philadelphia Eagles and Detroit Lions (two different flavors of Top 5 offense) averaged 5.9 yards per play this season.

Granted, the Vikings did play the Eagles and Lions (twice) this year, but their defensive play made their opponents, on average, look like elite offenses.

As I’ve noted in both of my defensive previews, Vikings’ Defensive Coordinator Ed Donatell was a long-time assistant to Vic Fangio and has based much of his philosophy on Fangio’s. Underpinning Fangio’s philosophy is the use of Cover 4 shells, which are effective in preventing explosive plays down the field. Much of the NFL has switched to Cover 4 defenses this year, which has both contributed to the general decline of offense league-wide (this has been the worst year for offensive football since at least 2010) and the resurgence of the running game across the NFL. This philosophy can be very effective, as evidenced by the elite defenses that Fangio has produced for multiple franchises.

But it can also be very vulnerable if teams don’t have the personnel to execute it and are going against an offensive mind who can exploit its weaknesses.

The Vikings, clearly don’t have the players on the back end to make the system work the way it should. Patrick Peterson and Harrison Smith have been excellent players in the NFL, but Father Time has caught up with them.

Meanwhile, Mike Kafka is quickly proving to be a master of identifying and exploiting his opponents’ tendencies.

Before we go on, this is what a base Cover 4 defense looks like:

One of the vulnerabilities of the Cover 4 is that it can lead to “death by a thousand cuts” scenarios. Because it sees the corners drop into deep coverage zones and forces linebackers (or safeties in sub-packages) to defend the shallow area of the field, those areas are vulnerable to quick passes.

That was exactly where the Giants attacked. Only 6 of Daniel Jones’ 35 pass attempts traveled further than 10 yards downfield, and he averaged just 4.9 air yards on completions. Part of that is just the nature of the Giants’ offense — they don’t look to throw the ball downfield and no team (not even the Ghost Of Matt Ryan) has averaged fewer than the Giants’ 6.3 intended air yards.

In this case, those quick passes put the Giants’ receivers right in the gooey center of the Vikings’ coverage zones.

Just to illustrate how Kafka dissected the Vikings’ defense, I superimposed Isaiah Hodgins’ routes (courtesy of NextGenStats’ player tracking data) over that diagram of a Cover 4 defense.

Despite my poor artistic ability, you can still see how Hodgins was matched up against safeties and linebackers, or run into the voids between coverage zones.

Of course, that was only one receiver and those routes weren’t run in a vacuum, and not all of them were against Cover 4.

They were a part of route concepts designed to manipulate the Vikings’ defenders and create separation for the receivers. Only six of Jones’ 35 passes were thrown into coverage, and the Giants’ receivers had plenty of room to work.

How effective was the game plan? While the Giants only picked up 171 air yards on the game, they had 130 yards after the catch.

So far, Kafka has drawn interest from the Carolina Panthers, Houston Texans and now the Indianapolis Colts — and the Giants should probably start thinking about who will call their offense next year. Just in case.

Wink Martindale’s gamble

On the flip side of the equation was the Giants’ defense.

Giants fans used to seeing a hyper-aggressive mentality from Wink Martindale likely had heartburn by halftime as the Vikings were able to move the ball pretty much at will.

Remember how the Vikings allowed their opponents to pick up 5.9 yards per play this year? That’s how many yards per play the Vikings picked up against the Giants’ defense in this game.

A big reason for that was Martindale taking a page out of the Vikings’ playbook and playing a much less aggressive and more zone-based defense than we are used to seeing from him. Any time you go against the foundational philosophy of your scheme it’s a big risk. It might catch your opponent off-guard, but it could also backfire spectacularly. In this case, the Giants were more concerned with Justin Jefferson than their defense failing completely.

The Giants clearly came into the game with the intention of not allowing WR Justin Jefferson to take over the game and beat them. Not only was Adoree’ Jackson tasked with shadowing Jefferson, he was bracketed on almost every play and only got 47 yards on the game.

Of course, paying so much attention to Jefferson created opportunities for other pass catchers. Adam Thielan had several big catches on the game, and T.J. Hockenson continues to be a problem for the Giants. In fact, we can see

The similarities between the Giants and Vikings’ defense produced surprisingly similar separation numbers.

The Giants generally only sent four rushers this game (after blitzing on roughly 40 percent of snaps in the regular season), but those rushers were able to pressure Kirk Cousins.

Credit to Cousins for standing in the pocket until the last instant and consistently finding a receiver downfield. He completed 81.6 percent of his passes (not counting throwaways), and was 13.7 percent above his expected completion percentage (93rd percentile since 2010).

Fortunately for the Giants, the Vikings’ defense was just as porous, if not more-so, as New Yorks. Had the Vikings had a even a decent defense that stopped the Giants’ offense a couple times, Martindale’s gamble might have turned out very differently. But often the difference between “brilliance” and “foolhardiness” is “success” and “failure”. This is one of those cases where the gamble paid off, and was brilliant because of it.