How To Improve The Offense in 2021 and Beyond

As Pelissero noted, Mike Zimmer was clear at the end of the season about what he wanted from the offense: More of the same.

“I love the scheme that we’re running offensively, I love the wide zone offense, I love the play-action passes,” Zimmer said during his end-of-year press conference. “All those things. A coach told me one time that your offense should be what your quarterback is best at. And that’s what I feel Kirk is best at. Those kind of things are what makes him really good. So to me, that is really important.”

Most of Cousins’s career has been spent playing in some form of the Kubiak offense. In Washington he played for Mike and Kyle Shanahan and then Sean McVay was his OC in 2016.

Over the last two years with Kevin Stefanski and Gary Kubiak, Cousins’s numbers have reached new heights with the Vikings’ QB averaging 8.2 yards per pass attempt and registering a 106.1 quarterback rating. He also graded by PFF as the fifth and 10th best QB in the NFL in those two seasons.

Both the run game and play-action passes have been extremely effective in the Kubiak system. The Vikings were fourth in yards per carry last season and Cousins had the fourth best QB rating when using play-action, only behind Aaron Rodgers, DeShaun Watson and Tom Brady (per PFF).

The biggest question facing Klint Kubiak is not whether the system can work to create a good environment for Cousins, it’s whether he can take it to the next level.

Even with improvements on defense, it’s going to be challenging for the Vikings to get back to their top-five defensive rankings of the past, meaning the difference between being a mid-pack NFC team and a potential division winner is going to be whether they can be more productive than 11th in points and 18th in percentage of drives in which they produce points.

Will Klint Kubiak have the freedom under Zimmer to throw more often in certain situations? Will he be able to modernize the offense more along the lines of motions and quick game? Will he find more ways to get Dalvin Cook involved in the passing game? And will the front office give him everything he needs a la more weapons and improved interior offensive line?

If Klint Kubiak runs everything the same exact way, the Vikings would still project to have a good offense. The 33-year-old is taking over a situation with two top-notch receivers, one of the NFL’s best running backs and an up-and-coming tight end. But with the top four teams in points reaching championship weekend, the Vikings need to set their sights higher than just having a good offense.

There is the question about whether the Vikings should have gone another direction. From a schematic standpoint, the answer is probably no. That would have required young talents like Justin Jefferson and Irv Smith Jr. to learn a new offense and Cousins to play in his third system in four years in Minnesota. In a year in which the Vikings want to get back to the top of the NFC North, overhauling the offense would have been risky. Plus it’s unclear whether this offseason will be “normal” or not.

Systematically, the Shanahan/Kubiak style is thriving all over the NFL. The Packers led the league in scoring last season with a Kubiak disciple at the helm. Tennessee had a top-five offense running a very similar philosophy, Kevin Stefanski won Coach of the Year and the 49ers were in the Super Bowl in 2019 with Jimmy Garoppolo operation all sorts of play-action passes.

All of these offenses, however, have taken what Mike Shanahan and Gary Kubiak created and built upon it. Can Klint do the same?

Ultimately, the freedom in which Klint has to make improvements and push the offense to do new things will be determined by Zimmer. In past years, the Vikings’ head coach has appeared to dictate the offensive philosophy. To his credit, he has also made tweaks and changes after playoff-less seasons that resulted in turnarounds the following year. Will the offense stay status quo with a status quo hire? We won’t know until we see it in action.

Minnesota Vikings quarterback Kirk Cousins capped off his 10-game resurgence with a walk-off NFC offensive player of the week honor after throwing for 405 yards and three touchdowns in a win over Detroit in the season finale.

It's the second time in five weeks Cousins received the award, a fitting footnote to a season rife with early turbulence only to finish on a positive note.

Cousins went from 10 interceptions in his first six games to capping off his sixth season as a starting QB with a career-best 35 touchdowns and his second-highest passer rating (105). Despite some struggles, Cousins still finished as a top-10 quarterback (ninth, according to PFF, down from fifth in 2019). Six of the eight QBs ahead of him led their teams to the playoffs.

It took the first half of the season for the Vikings to figure out how to best use Cousins in this scheme. It yielded the best version of the quarterback since he arrived as a free agent in 2018. The biggest takeaway for Vikings' brass is figuring out how to make this version of Cousins appear again in 2021.

Cousins is not standing in the way of the Vikings reaching their playoff goals. His success, however, hinges on an improved offensive line and the resurgence of a rebuilding defense.

Coach Mike Zimmer believes this was the most explosive offense he's had in his Minnesota tenure. The Vikings' 430 points scored are the third-most in franchise history but ranked 11th league-wide in a season in which scoring was up across the board. Minnesota was 10th in expected points added per play and fourth in yards.

The Vikings are just the ninth team in NFL history to feature a 1,500-yard rusher (Dalvin Cook) and 30-plus touchdown passer, according to ESPN Stats & Information. Each of the other eight teams won at least 10 games, including the 2020 Tennessee Titans. Minnesota finished the season with a 7-9 record and missed the playoffs.

While the Vikings' offense flourished at times -- Justin Jefferson had one of the best rookie seasons ever by a wide receiver -- the defense stumbled in a season that will be remembered for turnover, injuries and a steep learning curve for young players.

Cousins and the Vikings showed they could hang around with some of the league's best offenses like Tennessee and Seattle, even New Orleans. But Minnesota's defense, which ranked 29th, couldn't hold up its end. The 475 points allowed are also the third-most in team history. For the first time ever, the Vikings were outscored in a season where they scored at least 400 points.

"Let's be honest, you judge wins and losses in this league," Vikings offensive coordinator Gary Kubiak said. "Everybody's trying to be a playoff team and then be the best team for one month. I know that's what this league is all about. We've done some good things on offense, we moved the ball, some guys had some good numbers. That's great, but that's all about winning."

The offensive continuity Zimmer so desperately wanted to carry over from 2019 to 2020 seemed to pay off, even if they couldn't always carry the weight of the entire team. The Vikings had 10 games decided by one possession or less this season.

With Cousins returning to an offense in 2021 with Jefferson, Cook, receiver Adam Thielen and tight end Irv Smith Jr., the issue won't be the weapons around him. The offensive line is a different story. Minnesota finished below the middle of the pack in pass protection (18th in pass-block win rate) with 20 of Cousins' 39 sacks attributed to the interior of the offensive line. Some of those sacks might be on Cousins for holding on to the ball too long, but it's evident the Vikings need to upgrade certain spots on the line.

"I like that group, but we'll have to see and if we can improve, we'll improve," Zimmer said.

The Vikings could benefit from tweaking their offensive philosophy, like being more aggressive passing early in games like they are late (Cousins had a 114.1 second-half passer rating and threw 25 second half/OT touchdowns compared to 10 in the first half). But don't expect their overall method to stray far from a run-first scheme.

The success of this Vikings' offense is reliant on the play of its defense. Zimmer's philosophy of having a strong defense combined with an offense that controls time of possession and limits turnovers won't work without an improved pass rush. The Vikings ranked 30th in rushing the passer. And they must get better at stopping the run (they ranked 32nd in run-block win rate). Addressing those needs in free agency and the draft is as important as getting the players who missed time this season back in 2021.

"A priority for me is that we continue to get more pass-rushers," Zimmer said. "So we need people that can rush the quarterback, whether it's a linebacker, a cornerback or safety you've got to have guys that can cover in today's NFL. So that makes a big difference."

Cousins showed his ability to bounce back from a disastrous start. Against Detroit in Week 17, Cousins became the fifth QB with 250 passing yards and three touchdowns in a half this season, joining Tom Brady, Baker Mayfield, Patrick Mahomes and Josh Allen.

It's play like that, along with his improved scrambling ability and his accuracy (Cousins finished with the eighth-lowest percentage of uncatchable passes thrown, according to PFF), that leaves the Vikings confident QB play isn't their issue. They don't have to try and orchestrate a blockbuster trade to move away from Cousins and his massive contract this offseason, but they do need to address the areas of need around him this offseason.

Around the same time that Vikings interviewed Tolbert, Mike Zimmer — a defensive head coach who is very hands-off with both the offense and special teams — stated that he wanted to add to the offense.

This might come as a surprise. Last season the Vikings’ offense finished 4th in yards per game with 393.3 last year, and the defense was a liability. But on some level, it makes sense.

Despite being efficient after some early season hiccups, the offense sometimes felt stale and predictable. This season, we saw even more creativity from the NFL’s younger offensive minds while it felt like the Vikings stuck to a safer formula that took no risks. I don’t recall a single trick play, and the only new wrinkle in the offense was when they lined up Chad Beebe as a shotgun running back on third down.

I’m not saying that the offense was primarily to blame for Minnesota’s 7-9 record — Justin Jefferson, Adam Thielen, Kirk Cousins, and Dalvin Cook all statistically had great seasons — but the Vikings could use some innovation to enhance a formula that has proven to be effective.

But who will be the mastermind before the new approach?

Situationally, the Vikings have shown a pair of odd tendencies that are neither efficient nor progressive in their thinking. They run it constantly on second and short. And they run it constantly on second and long.

These two trends may be the best way to quantify why the Vikings seem to be perpetually toeing the line between a good offense and a great offense, one that frequently scores in the mid-to-upper 20s but only scored 30 points in a game three times this year (and lost two of them). When the Vikings were 1-5 at the bye week, Kirk Cousins said the offense was hitting triples and home runs but needed to focus on hitting more singles. Minnesota actually delivered better results following this remark with a 5-1 stretch, even though it would make any modern baseball manager faint. But as the Vikings struggled to beat the Carolina Panthers and Jacksonville Jaguars, then lost to the Tampa Bay Buccaneers and Chicago Bears, it felt like Gary Kubiak and Mike Zimmer were calling for bunts when trailing by four in the eighth inning. The singles approach may have a place, but there’s reason to question its long-term sustainability.

SECOND-AND-SHORT

On Sunday, the Vikings faced eight situations where they had a second down and short (1-3 yards), considered to be an opportunity where offenses can take deeper shots without fear of setting up a third and long. Teams in 2020 run it, on average, 66% of the time in these situations, or two out of three. The Vikings chose to run it with Cook seven out of eight, or 88%. They only opted to throw in the final minute when they needed to drive the field in less than a minute with no timeouts. It seems like an unbalanced tendency, but that’s how the Vikings have handled it throughout the year.

Minnesota runs on second-and-short 82% of the time, the third-most in football. If this were a foolproof method that led to the most first downs in the league — like Tom Brady and his famous QB sneak — it might be an understandable tendency. But the Vikings are only average at converting: 65.2%, 16th in the NFL. You can’t really use the small sample size argument here either, since the Vikings have had more second-and-short opportunities than any team in the NFL (66). That’s another way of complimenting them for their work on first down, where they are also No. 1, averaging a robust 6.8 yards per play. So many second-and-shorts… so little productivity.




The Vikings’ approach prioritizes moving the chains but limits their explosive potential with two of the league’s top receivers. Again, if this were a team with a best-in-the-league bottom line on offense, it may not be an issue. But Minnesota is merely 20th in scoring percentage (drives ending in points) and were 13th in DVOA heading into Week 15. They are in line with the league average at moving the chains on second-and-short, but their yardage on those plays is minimal: 2.9 yards per carry, or 27th in football. When the Vikings choose to pass, Cousins does well. He is 9-for-11 in those situations with nine first downs, 7.8 yards per attempt, two touchdowns, and a 138.8 passer rating. Those pass attempts, however, tend to only come when the Vikings are trailing late.

Consider this: If you isolate second-and-short situations to the first three quarters, the Vikings become the most extreme team in the NFL. Their rush percentage shoots up to 91.5%, but their first-down percentage regresses to 18th in the league. This isn’t just about running vs. passing. This is about predictability. There are few, if any, play-calls that should be made over 90% of the time in a certain situation. If the data showed Minnesota was passing it 90% of the time in those situations, that would also be grounds for criticism.

Let’s look at it another way. How do teams with the most second-and-short success call plays? In terms of yards per play, seven of the 10 most efficient teams pass the ball more than the league average of 34%, led by Tampa Bay at 7.4 yards per play and a 36% pass percentage. The Houston Texans (second-best YPP) and Pittsburgh Steelers (fourth-best) both pass it over 50% of the time, and they both sit in the top 10 in first-down percentage. Hmm.

What about recent Vikings teams under different play-callers? Kevin Stefanski was a bit better calibrated in 2019 with a 72% run percentage on second-and-short, though the total normalizes to 68% when removing the fourth quarter, where the Vikings were sitting on a lot of leads. More interesting is John DeFilippo’s balance in 2018, which leaned heavier on the pass and eventually got him fired. DeFilippo was 50/50 however you splice it, first three quarters, whole game — perfectly split. If the Vikings had done better in those situations, maybe DeFilippo would have survived the season. Minnesota was only 25th in first-down percentage, dealing a costly-blow to the passing-over-rushing advocates out there.

SECOND-AND-LONG

The Vikings are hyper-conservative on second-and-short, and to compound matters, they don’t make up for it with a lot of hyper-aggression elsewhere. Take second-and-long as an example (10 yards or more), where the Vikings distance themselves even further from their peers in run-pass ratio.

Ironically, the Vikings are closer to achieving balance here than they are on second-and-short, but balance is not a good thing in this case. Facing 2nd and 10 or longer, Minnesota has run it 43 out of 93 times, or 46%. Only the Denver Broncos have a higher percentage (47%), and the Broncos have a much worse passing game. Isolated to the first three quarters, the Vikings are No. 1 in the NFL at a 45% run rate.

It’s worth asking again: Have the Vikings found a market inefficiency here? Is running the ball actually producing good results, and that’s why they continue doing it? The answer is no. The Vikings convert first downs 11.6% running the ball in second-and-long, 17th in the NFL. They’re averaging 4.7 yards per carry, which is actually worse than their overall team average, so basically, they are aspiring for a 3rd and 5. They haven’t been great passing out of these situations either, but still more effective at 6.0 yards per attempt. Overall, run and pass, the Vikings are converting first downs 14% of the time on second-and-long when the league’s best team, the New England Patriots, is more than double that.

What’s jarring is the discrepancy between the Vikings’ run-pass ratio and the rest of the league. Sure, there are outliers like Denver and former Vikings play-caller Pat Shurmur that do things similarly, but the league average is 31% for rushing on second-and-long. Five out of the top seven most efficient teams in those situations (by yards per play) pass it more than the league average (Dallas, Arizona, Miami, San Francisco and Tampa Bay). The Vikings, on the other hand, are 15% more conservative than average in those situations yet aren’t enjoying success.

As we did above, let’s compare the data this year to what we’ve seen with past Vikings’ play-callers. Believe it or not, they were actually more extreme in 2019. They were the only team remotely close to a 50/50 run-pass split on second-and-long with 48% running plays, but again, that team had more reason to run because it held more leads. DeFilippo, predictably, was closer to league average, calling runs just 33% of the time.

The current Vikings offensive brain trust prides itself on an atypical scheme that uses the fewest three-receiver-sets in football in the name of keeping tight ends on the field as run blockers. This worked well for the 2019 Vikings, who had a better defense that kept games lower scoring. It hasn’t worked as well for the 2020 team with its young defense, and the play-calling hasn’t done much to adjust. They are predictable on second-and-short when they should be unpredictable. They are unpredictable on second-and-long when the predictable thing would actually be more effective.

The Vikings are getting their money’s worth out of Cook, but they aren’t exactly swinging for the fences.

In the mid-19th century, the USS Central America would regularly sail between New York to Panama. This was the middle of the California Gold Rush, so the boat was often full of prospectors and their hauls. In 1857, the USS Central America sank somewhere off Charleston’s coast, dragging down over 500 passengers and over $100 million in gold.

Fast forward 130 years, when a treasure hunter named Tommy Thompson decided he wanted to find that loot. Finding a single shipwreck in the ocean is impossibly hard. By just guessing and checking, it would be about as likely as stumbling upon a single grain of sand in a four-bedroom floor plan. So Thompson needed to deploy some advanced math.

Bayes’ Theorem is a mathematical concept that helps determine how probabilities change when new information is introduced. When the NFL expanded the playoffs to 14 teams, for example, the probability that the No. 1 seed would win the conference increased. Bayes’ Theorem is the tool used to calculate the difference.

Thompson used Bayesian math to help streamline his search. Each patch of ocean that didn’t turn up bountiful treasure was also a piece of information. Given that the treasure wasn’t here, it changes the probability that it’s over there. Using Bayes’ Theorem, he could sort more quickly through the vast open sea and hone in on the USS Central America’s remains.

After just three summers of searching, despite bad weather, faulty equipment, and clashing egos, Tommy Thompson found the haul he was looking for.

Bayesian probability is a potent tool that can be applied to football as well. We can get a better understanding of target shares, quarterback decisions, and general play-calling. And Kirk Cousins is perfectly suited for such a study.

This won’t be an evaluation of Cousins. He is just a good example we can use to learn more about NFL progressions. Because Cousins is extremely consistent in how he reads plays, it’s easy to learn the basics through watching him play. Where other quarterbacks may leave the progression in certain scenarios, Cousins applies a very constant approach. That can be good or bad depending on the situation, but when trying to learn more about NFL play design, it makes him a helpfully controlled variable.

Often, Cousins will begin a play with an immediate read, called an “alert.” For a common example, if the defense blitzes, Cousins will pull out of the progression and throw a pass that punishes that blitz.



After that point, the rest of the progression is informed by the fact that there is no blitz. If there were, we wouldn’t have gotten to that point. So the rest of the play can be comfortably designed for a non-blitz environment. This is why “getting pressure with four” is so oft-repeated by TV analysts. At this point in the play, it must be taken as a given that at least seven defenders will be in coverage. Knowing that helps to design the next portion of the read.



If the defensive coverage addresses the short middle with no blitz, it’s likely a 2-hi coverage. That means Cover 2 or Cover 4. Against those, a “Hawk” or “Ohio” route concept is likely to be successful. It’s not the best against man coverage or Cover 3, but against those coverages, we wouldn’t have gotten this far into the read. Here, the Seattle Seahawks show only one high safety but play a modified version of Cover 4 with man-match concepts. It’s a defense that beats many things, but once we’ve sorted out the blitz and several coverages, our odds get a lot better.





Simply running “Hawk” against a random defense isn’t comprehensively successful. But running “Hawk” against a defense that we know isn’t a blitz or Cover 3 is a much better bet. By designing and reading plays with this Bayesian idea in mind, most coaches and quarterbacks can remove the element of surprise. With each layer, the odds increase that they’ll have a counter to the defense’s play call worked in.

Perhaps the Minnesota Vikings could use some more complex ideas, with all the continuity on their roster from 2020 to 2021. A new offensive coordinator is unlikely to install a brand new scheme or core terms, so they could afford to up the complexity a bit. The 2020 Vikings used a lot of “mirrored” concepts, pictured below. These are simple to run but also simple to defend. An example:



In 2021, the Vikings could lean on their quarterback’s in-structure prowess by simply expanding the structure. By adding more layers that narrow down the possible counters, they could be more prepared for varied defenses. They could use the right side and left side of the field to layer in additional contingencies. That wouldn’t require moving on from their heavy outside zone, play-action approach. It would juice up their straight drop-back passing game and increase their chances when behind the 8-ball. Crucially, the players are set up to handle that.