Entering the 2018 NFL draft, Josh Allen was seen as half man, half meme—a caricature of the big-armed quarterback who couldn’t hit water if he fell out of a boat. This was largely due to his collegiate stats: Allen completed barely half of his passes (56 percent) in three years at the University of Wyoming, and he ranked last in every statistical category among draft-eligible quarterbacks in 2018, according to an analysis conducted by NumberFire. But it wasn’t just the numbers that turned Allen into a punch line—it was the way he was missing.
All this evidence caused many to write the QB off early in his career. Before Allen had even taken a professional snap, then-Jaguars All-Pro cornerback Jalen Ramsey called him “trash.” Football Outsiders wrote that Allen was “a parody of an NFL prospect.” We here at The Ringer referred to him as “a football fantasy,” and I, in my tribal exuberance over a Jewish quarterback, put Josh Rosen over Allen in a quarterback prospect ranking.
Fast-forward three years, and things have changed a bit. Allen and the Bills nearly reached the Super Bowl last season, and Allen finished second in NFL MVP voting behind Aaron Rodgers. The Buffalo QB also became the first player in NFL history to have 4,500 passing yards, 35 passing touchdowns, and five rushing touchdowns in a single season. This new level of play was a direct result of Allen’s improvements in decision-making and accuracy. In 2018 and 2019, Allen ranked dead last in completion percentage; in 2020, he rose all the way to fourth. His jump from completing 53 percent of his passes in 2018 to 69 percent was the largest two-year increase in NFL history.
Few people thought that kind of leap was even possible, let alone that Allen would be the one to make it. Conventional wisdom said that a quarterback’s accuracy couldn’t be significantly improved once they reached the NFL. Sure, throwing motions could be tinkered with, footwork could be solidified, and timing could become more precise—but those tweaks were supposed to round out a quarterback’s edges, not alter the core of how they play.
Then last year happened. Allen’s season upended what we thought we knew about quarterback development in the NFL. Now, as we head toward a draft that might have more passers taken in the top 10 than ever, has Josh Allen’s rise changed how teams look at quarterbacks?
Sage Rosenfels believes quarterbacks can become significantly more accurate in the NFL. After all, he managed to do so as a pro.
When Rosenfels was at Iowa State, he never had a completion percentage better than 54 percent. But Rosenfels spent nearly a decade in the NFL as a backup quarterback, and in 2007, when he saw his most single-season action (five starts for the Texans), he completed 64 percent of his passes. While completion percentage is not a perfect gauge for a quarterback’s accuracy, the two are related. And Rosenfels says his improved accuracy came from a combination of physical and mental changes that altered both his outlook at the line of scrimmage and his abilities as a passer.
“Not only did I get used to throwing these passes over and over and over, and hitting that same seam route over and over and over, but more importantly I understood the defense so much better,” Rosenfels says. “Also in the NFL, things are very precise as far as the timing and the depth and the distance for the wide receivers. So you do that enough, and you do just get better at it.”
Footwork is a window into a quarterback’s mind, and much of what is needed to become a more accurate passer is mental. Every step a quarterback takes has to be in sync with the moves his receivers are making. It’s a dance. And Rosenfels—who is now a private quarterbacks coach for the QB Collective—says there’s a reason coaches want passers to run an offense in rhythm.
“Footwork creates timing,” Rosenfels says. “You can’t see a guy open and throw it to him. Almost every time, you have to throw it before he is open. So you are about to throw it, and he is about to start looking.”
From a quarterbacking perspective, accuracy means putting the ball in a catchable location for the wide receiver. But the receiver must also be in the correct spot.
“I tell my quarterbacks all the time, ‘Accurate quarterbacks throw to accurate receivers,’” Rosenfels says. “You need to accurately run the route the way it’s supposed to run. When [the quarterback’s] back foot hits, you should be coming out of your break.”
This was part of the reason Allen struggled in his first two seasons, according to Bills general manager Brandon Beane. “Some of the issues that we had in Year 1 that Josh would never say a word about, we had receivers that didn’t always run the right routes and didn’t maybe run them to the right depth,” Beane told Ringer staff writer Nora Princiotti in October. “We’re not going to publicly criticize those guys, but there were plenty of times when people were getting onto Josh when it wasn’t Josh’s fault.”
To change this dynamic, the Bills went out and added as much receiving talent as they could to give Allen a better support system. The team signed slot receiver Cole Beasley and speedster John Brown in 2019. And ahead of the 2020 season, Buffalo traded for Stefon Diggs, the best deep-ball tracker in the NFL. With that extra help, Allen went from the second-worst deep-ball passer in the league in 2019, according to Pro Football Focus, to the sixth best in 2020, one spot behind Tom Brady. And Diggs led the NFL in receiving yards and receptions last season as a result.
But receiving talent was just one part of the equation. The Bills also prioritized keeping their offensive line intact to give Allen time and protection in the pocket. And offensive coordinator Brian Daboll ran a system that passed more on first down than any other team in the league (62 percent). That gave Allen more opportunities to throw against favorable defensive looks, since defenses have to pay more attention to the run game on first down. All of that together put Allen in a position to succeed. But the QB still had plenty of work to put in on the physical side.
As Allen’s private quarterbacks coach, Jordan Palmer, told Princiotti, Allen bounced on his toes when he threw in college, which affected his accuracy. When a QB bounces, Palmer says, “you’ve added movement that you don’t need. And every time you stride you’re going to overstride, which means your foot is going to land a little differently and every ball is going to come out a little differently.”
Basically, if a quarterback is late in rhythm, they’ll often step into the throw more than usual to put more oomph on the ball. That causes their front leg to land farther out, which tilts the back (throwing) shoulder down and changes the trajectory of the throw. So the act of trying to throw the ball harder not only alters the amount of force on the ball, but also its angle—which can lead a quarterback to sail a ball over a receiver’s head.
Under Palmer’s tutelage, Allen learned to keep his feet closer together, and made a number of other physical changes. Allen practiced on the beach to make his upper-body mechanics more consistent. He spoke with Tony Romo about tucking his left arm in and keeping his head in the same place on every throw. And he worked with Buffalo’s quarterbacks coach, Ken Dorsey, to bend his knees more, like a boxer.
But it’s unclear how much those mechanical changes stick once a quarterback is on the field. Footwork can certainly be improved. But Nate Tice, a former NFL staffer and quarterback at the University of Wisconsin, believes that when a quarterback gets into trouble, they revert back to the throwing motion they’ve done their whole life.
“You throw how you throw,” Tice says.
That seems to fly in the face of the adage that practice makes perfect. After all, if Knicks forward Julius Randle can go from a 29 percent 3-point shooter to a 40 percent 3-point shooter in his seventh NBA season, why can’t NFL quarterbacks become more accurate too?
“It’s not a basketball shot where you can improve the mechanics and just rep it, rep it, rep it. In a basketball shot, you’re sitting there, your feet are set, then you’re shooting,” Tice says. “In football, you’re all over the place. You have to throw off-platform, and make a play throwing sidearm, and over the top, and the next one you’re throwing off your left foot. And I think that’s why it’s so much harder to get more accurate, is because there’s so many more mechanical things happening when you’re throwing a football.”
Allen’s metamorphosis likely came from a mix of better footwork, a dash of mechanical fixes (that may or may not have stuck), and a lot of help from a front office dedicated to giving him the right supporting cast of coaches and teammates. “Calm feet and calm mind have led to [Allen’s] improvement,” Tice says.
The question is whether Allen’s improvement is a one-off, or whether it can be replicated.
They say one great comedian inspires 10,000 bad ones. Could Allen inspire teams to take risks on inaccurate quarterbacks in the hopes they can strike gold? Should teams rethink their approaches entirely and treat accuracy as something that can be learned?
Both of those things are a lot harder than they sound. Tice believes Allen came into the NFL with a lot of room to grow, but that his circumstances are not replicable.
“Usually a guy drafted in the first round like this, their feet are going to be pretty damn tight,” Tice says. “Trevor Lawrence has perfect footwork. Justin Fields has pretty good footwork. These guys have good footwork already. So once they get to the NFL, there’s going to be some more consistency, a little bit of improvement, but you aren’t going to see much accuracy growth from footwork improvement.”
Mike Renner, the lead draft analyst at Pro Football Focus, says Allen’s development changed the way he judges quarterback prospects—to a degree. He’s now willing to give greater looks to flawed players who have higher-end upside. But Renner also points to Christian Hackenberg and Drew Lock as cautionary tales of teams chasing quarterbacks with big arms and big frames despite accuracy issues. Both players were second-round picks. Hackenberg flamed out of the league without ever playing a down, while Lock finished dead last in completion percentage last season. (Sound familiar?) Looking at Allen’s rise as proof that a player like Lock—or a future prospect with accuracy issues—can succeed may be missing the forest for the trees.
We have a tendency as football observers to assume that whoever a player becomes in the NFL is who they were always going to be. The reality is every prospect has a range of outcomes, and both landing spot and coaching—along with individual growth—help determine how good or bad a player becomes. Allen seems to have landed in the perfect situation to hit the high end of his potential. But perhaps if he’d gone third overall to the Jets in 2018 and been coached by Adam Gase, you would not be reading this story.
“A lot of guys come in with his profile,” Renner says. “Not a lot of guys have turned into Josh Allen.”