As Kevin Clark explains in an excellent Ringer article, the NFL’s analytics revolution is finally kicking into gear. Teams’ analytics departments have swelled and the development of more advanced statistics for player evaluation, implementation of player tracking data, and ever-improving knowledge of salary cap best-practice have followed, bringing a new wealth of information for front offices and coaching staffs to leverage. The tendrils of the new information are far reaching—data can be implemented from on-field schemes to drafting and acquisition decisions, meaning teams who use analytics wisely gain widespread benefits.
The NFL finds itself at a very particular moment in its analytical development. Information is becoming abundant, but only some teams have an understanding of how to put it to use, and ‘smart’ teams find themselves with a tremendous advantage. Not only can they consistently outscheme opponents, but they are also much better positioned to evaluate players and construct their rosters according to salary cap restraints. As the implications of being analytically savvy are so extensive, the question is begged: which teams are smart? And what does that really mean?
Perhaps the foundational discovery of the NFL’s analytics revolution is that passing the football is a lot more effective than running it. The finding is analogous to the discovery that threes are better than twos in the NBA. There is a mathematical proof for why throwing the ball is better which relies on Expected Points Added (EPA), but a basic understanding of the phenomenon comes from the realization that averaging four yards per carry is pretty good, while four yards per pass is not so good.
Understanding passing compared to run efficiency is undeniably powerful. The spread-it-out, pass-crazy trend has been going on for years—the Patriots’, Colts’, and Saints’ offenses of the mid-2000’s certainly fit that bill—but this season the tendency seemed to jump to a new level. Teams like the Rams, Saints, and Chiefs dominated the 2018 regular season with explosive passing attacks. While running back Todd Gurley is a centerpiece in the Rams offense, we watched head coach Sean McVay transform former 1st-overall pick Jared Goff from apparent bust to the leader of a high-octane passing attack featuring a healthy dosage of deep balls to wideouts Brandin Cooks, Cooper Kupp, and Robert Woods. The Saints have been doing the pass-happy thing for a while now under Sean Payton and Drew Brees, but over the past two years they have reached new heights with the addition of the elusive Alvin Kamara who acts as half running back, half receiver. The Saints use Kamara masterfully and create matchup nightmares for opposing defensive coordinators as they send him out of the backfield on a variety of different routes which he consistently burns linebackers on. The most breathtaking example of passing bringing a team towards the promised land is the Kansas City Chiefs. Head coach Andy Reid has long been ahead of the curve when it comes to innovative offensive schemes, but pairing MVP Patrick Mahomes’ cannon-blaster-appendage that he calls an arm with weapons like Tyreek HIll and Travis Kelce produced an incredibly lethal offense in KC.
The culmination of the offensive explosion seemed to come on a Monday night in November as the Rams beat the Chiefs 54-51 in a thrilling shootout which saw a total of ten touchdown passes between Goff and Mahomes. It was a truly amazing game, and commenters moved swiftly to forecast the future of the NFL; as teams tried to emulate the offensive juggernauts in LA and Kansas City, defenses would die and the NFL would slowly but surely begin to look like the Big 12. ‘Smart’ teams would be the ones who figured out how to put a high-octane offense, led by a stellar quarterback, on the field..
The 2019 playoffs painted a contrasting picture to the football of the future that many imagined after the Rams-Chiefs MNF classic. The first round produced a series of ugly games featuring head-scratching play-calling and subsequent twitter-dunking over the heads of several offensive coordinators. The Seahawks’ Wild Card game performance in Dallas was a particularly galling example of a team which has responded to the incorporation of new information with an emphatic “NO!”. Under the guidance of head coach Pete Carroll, the Seahawks have attempted to harken back to the glory days of 2013, when they fed Marshawn Lynch the football and watched him bulldoze unfortunate defensive opponents. Against Dallas, offensive coordinator Brian Schottenheimer, a notorious rock-pounder, dialed up run after run with Chris Carson and Rashaad Penny. The ugly display culminated with Schotty calling for a halfback draw on a crucial 3rd and 10 in Dallas territory.
The great irony of the 2018 Seahawks is that, in Russell Wilson, they have one of the best quarterbacks in the NFL. Pro Football Focus’ (PFF) 2019 Quarterback Annual raves about the Seattle signal-caller: “Wilson’s ability to take care of the ball and make big-time throws down the field were key to him having a top-five season (amongst quarterbacks).” He earned an 87.2 PFF grade this past season, the second highest of his career and seventh best among qualifying QBs. More importantly, Wilson had the fourth-highest grade from a clean pocket and the highest big-time throw percentage in the NFL. PFF has shown these metrics to be the most stable season-to-season and are therefore useful for separating how much of a quarterback’s success, or lack thereof, is a product of luck. The fact that Wilson is so good made the Seahawks’ season and first-round playoff exit extremely frustrating. When set free from Carroll and Schottenheimer’s stubborn play calling in the Wild Card match up with the Cowboys, Wilson provided flashes of what could have been, connecting on multiple deep shots with receiver Tyler Lockett.
Similar anecdotes can be told about other first round playoff teams. The Cowboys offensive scheme against Seattle was a continuation of what we saw in the regular season: ride running back Ezekiel Elliott and limit QB Dak Prescott to short throws. Dallas perhaps has a better justification than the Seahawks for using a run-heavy offense, as Dak is no Russell Wilson, but the but the limits of this type of scheme were on full display in their Wild Card matchup. They won the game, but struggled to put away an opponent in the midst of a pitiful offensive performance.
In the AFC, the Ravens, who rode into the playoffs on the back of quarterback Lamar Jackson and a stout defense, saw their surprising success come to an end in a fiery offensive implosion against the Los Angeles Chargers. With Joe Flacco handing over QB1 duties to the rookie Jackson midseason, Baltimore went on a nice run, seemingly thanks to the use of a strategy which hadn’t been employed in the NFL for some time: using the quarterback like a running back. Jackson posted more than 20 carries in multiple weeks and the Ravens saw middling offensive success with the scheme. The Chargers, who had lost to the Lamar-led Ravens in the regular season, came armed with a strong defensive plan for the rematch and completely dismantled the Baltimore offense. The ordeal made the success of the QB run-based scheme in the regular season seem like a fluke which relied on its uniqueness compared to the typical NFL offense. Overall, the Wild Card round reinforced the notion that if you’re still pounding the rock, you’d better get in front of a mirror and take a good hard look at yourself.
While teams who relied heavily on the run in the Wild Card round met their demise, the subsequent playoff rounds were a spectacle that simply didn’t jive with what seemed so true in the regular season: that pass-based offenses would bring victory. Many expected to see offensive firepower and spectacular quarterback performances from the league’s top teams, but we watched the Patriots go on yet another championship spree by running the football down the Chargers’ throats, pitching a first-half shutout against the Chiefs, and keeping the Rams out of the end zone for an entire sixty minutes.
There will be more to come on the Pats in a second, but other strange playoff happenings must first be addressed. For starters, the Rams beat the Cowboys in the divisional round with no small thanks to 100+ yard rushing performances by running backs Todd Gurley and, the more unlikely hero, CJ Anderson. It was a fairly slow game and LA stayed committed to the run throughout. In their divisional matchup, the Saints defeated the defending champion Eagles in a low-scoring game in which New Orleans was bailed out by a pair of Marcus Lattimore interceptions. The NFC conference championship between the Saints and Rams was an odd game as well, which saw both Brees and Goff throw for south of 300 yards. Now, to the Patriots, who won three straight games in the most mysterious of ways. As was touched on earlier, the Pats gave the Chargers a heaping helping of running back Sony Michel from the outset, before shifting to a more pass heavy attack, the combination of which resulted in a complete slaughter of a talented LAC team. Next, in their AFC Championship game against Kansas City , New England held one of the most prolific offenses in NFL history scoreless through the first half. While the Chiefs heated up in the next thirty minutes and Tom Brady two-minute-magic was necessary to ensure victory, the first half shutout remained shocking. In short, after the first round, the playoffs presented some head-scratching developments for proponents of: “passing is good and offense wins.” In the Super Bowl, we got the icing on the cake. The Patriots were crowned after winning a 13-3 grinder against the Rams, the ultimate ironic ending to a season filled with offensive fever.
What happened in the playoffs? Were so many of us wrong about what the future of the NFL will look like? Sure, it was just a couple games, but a closer look reveals that analytical forces reshaping football are much more complex and nuanced than, ‘spreading it out and passing the ball is good.’ A look at season-long pass-run splits immediately reveals that the eyeball test, which led us to believe that smart teams were winning with passing-based offense, was extremely deceiving. The four horsemen of the analytics revolution—the Patriots, Chiefs, Saints, and Rams—ranked 26th, 10th, 28th, and 24th respectively in pass vs. run playcalls. If passing is supposedly so much better than running, why did the league’s most successful teams throw the ball at some of the lowest rates in the league?
The answer is context. While passing has been shown to be more efficient than the run on average, so much of the outcome of any given snap is determined by what the opposing team thinks is going to happen. Put yourself in the mind of an NFL coach and it is quickly apparent that each play call is a massive, game-theory web. An offensive coordinator must think: what does the opposing defensive coordinator tend to call in this situation? What seems to be the defensive scheme this game? What plays have been called recently? What play does he think I’m going to call right now? The reason the Pats, Chiefs, Saints, and Rams had such success this season is because their coaches are masters of this mind game within the game. The evidence lies in efficiency metrics. Despite New England, New Orleans, and Los Angeles throwing at some of the lowest rates in the NFL, they ranked as follows in passing offensive defense-adjusted value over average (P-ODVOA), a measure of offensive efficiency controlled for pass plays: 5th, 4th, and 2nd. In other words, when these teams decided to pass, they did so extremely effectively. What is even more interesting are the juggernaut’s ODVOA numbers when controlled for run plays. They ranked 9th, 8th, and 1st and the Chiefs, who passed at a high rate compared to the other three (and ranked first in P-ODVOA), ranked 4th in R-ODVOA. These stats show that the best teams in the NFL are simply operating at a different level when it comes to play calling. Their high efficiency numbers in both the pass and run reveal that the sequences in which they call plays and formations lead to an unpredictable code which is difficult for opposing defenses to crack.
While the math is true and passing on the whole is more efficient than running, unpredictability is really what football analytics is all about. Smart teams use information to make the game theoretical play-calling decisions other teams face as difficult as possible. The Patriots and their 2019 Super Bowl run perfectly embody this. Especially in the playoffs, they were an amorphous blob of a football team, completely lacking a single identity. Game to game, and in some cases, quarter to quarter, New England looked like an unrecognizable team. A perfect example of this was the divisional round game against the Chargers. LA played nearly their entire Wild Card game against the Ravens with seven defensive backs in the game and stubbornly held to this strategy against New England. The Pats’ coaching staff had clearly done their homework and ran the ball early and often, resulting in a complete smashing of the small-statured Baltimore defense. The ordeal was a beautiful lesson in the importance of being flexible in scheme design. On one side, Belichick and the Patriots perfectly adapted to the opponent at hand and on the other, head coach Anthony Lynn and the Chargers’ were rigid. Clearly the inflexible team paid a heavy price.
Defense was a huge part of the Patriots’ playoff success as well. A unit which had looked fairly pedestrian for much of the season suddenly was wreaking havoc on the most explosive offenses in the NFL. The mastermind of all of this was, of course, Bill Belichick. No one has made better use of available information for longer than New England’s overlord, and his seventh Super Bowl run was his most beautiful masterpiece yet. His willingness to completely change what his team looked like week to week allowed Belichick to both mold the Pats into a kind of Boggart, which took the form of the opponent’s greatest fear, and make prediction nearly impossible for opposing coaches. At the other end of the spectrum are teams like the Seahawks. While it probably isn’t great that the aforementioned Brian Schottenheimer is such an outspoken proponent of the run game, what’s worse is the all-so-predictable run on first down, run on second down, throw on third down sequence that he so often employs. Such a pattern is easily identified by an opposing defensive coordinator and neutralized.
Another dimension of unpredictability is roster construction and the notion of “positionless football.” We’ve seen this same phenomenon in basketball, as traditional positions have gradually been smushed together and players of certain body types are being asked to do things previously thought physically impossible. The strategy puts stress on opposing defenses as each player on the floor can do a variety of tasks. The trend has had analogous consequences in the NFL. It is embodied by running backs like Alvin Kamara, Todd Gurley and seemingly any Patriots back from the past six years. These guys still run the ball often, but are also catching balls at a high rate. Also, tight ends like Travis Kelce and Rob Gronkowski have been plugged in all over formations and excel as both receivers and blockers. The list of players who are beginning to blur positional lines goes on and on, but what they all have in common is the ability to do multiple things well on the football field, and usually have at least one skill which is not traditionally found at their position. These players are perfect for a coach like Bill Belichick. They are what allows him to treat his team like a lump of clay and make his next move so unpredictable. Gronkowski is just as likely to mash an inside linebacker on a power block as he is to run up the seam and catch a deep ball from Tom Brady: a perfect recipe for defensive confusion.
The NFL’s future is unpredictability. Smart coaches who leverage information to gear schemes to opponents’ weaknesses will succeed. Smart front offices who identify and sign players that can play a variety of roles on the field will aid their coaching staffs in the quest for this unpredictability. The trouble is that there are only so many really smart people out there. Even if these principles are widely accepted, only the special few—disciples of Belichick, Reid, Payton and McVay—will be able to effectively implement them. As more teams figure out how to keep opponents guessing, those who can’t will be left in the dust. In the coming years, look for the unpredictable teams to turn the gap between themselves and predictable teams into an abyss.