clock menu more-arrow no yes

Filed under:

Josh Allen Keeps Dunking on Me—and Defying Recent Lessons About QB Scouting

From the time he was a prospect, the Bills QB has been a polarizing figure in football circles. In 2020, he’s emerged as a consensus early-season MVP candidate. What changed? And what can that tell us?

Getty Images/Ringer illustration

It’s unfortunate that as sportswriters we aren’t allowed to change our opinions about athletes. If we were, I would say that Josh Allen has become a shockingly effective quarterback and one of the most fun players in the NFL. Sadly, I am locked in. Before Allen was drafted in 2018, I wrote that “he does not seem especially good at playing football,” and now my name is Rodger “He Does Not Seem Especially Good at Playing Football” Sherman. If Allen leads the Bills to the Super Bowl, I will be obligated to say he still sucks. I will be buried with this take.

Allen’s ability to play quarterback has been a point of contention in football circles for years. Heading into that draft, analysts were divided into two camps. The first was obsessed with his tools and salivated over the prospect with the prototype build and the cannon arm. The second was horrified by his statistics and couldn’t look past Allen’s underwhelming college numbers. The tools camp saw an athletic marvel unlike any other; the stats camp saw a QB who barely managed competency against subpar Mountain West competition. Then Allen reached the pros, and the divide somehow got starker. The Bills went 10-6 and made the playoffs in Allen’s second season thanks largely to one of the league’s best defenses, but statistically Allen appeared to be one of the NFL’s worst quarterbacks. The pro-Allen camp credited Allen for the Bills’ successes; the anti-Allen camp argued he was holding the team back, and that Buffalo would be a Super Bowl contender if it merely had an average quarterback.

But in Year 3 of Allen’s career, there can no longer be debate. Allen is incredible. Even the stats don’t depict Allen as disappointing anymore; through four games, he has not only proved to be significantly better than he was in his first two seasons—he’s significantly better than most other quarterbacks in the NFL. Allen never threw for 300 yards in a game in 2018 or 2019; he threw for 300 yards in each of his first three games in 2020. In Week 3, he threw for 311 yards with four touchdowns while running for a fifth score in a 35-32 win against the Rams. The week before, he threw for 415 yards with four touchdowns in a 31-28 win over the Dolphins. This wasn’t just Allen’s first 400-yard game in the pros—he never surpassed that mark in his three seasons at Wyoming. The Bills are 4-0 largely because of Allen, not in spite of him.

Allen had a negative DYAR (defense-adjusted yards above replacement) in each of his first two seasons in the league, implying that he was actually worse than a replacement-level quarterback. This year, he’s tops in the league in that metric. As a rookie, Allen threw 10 touchdowns and 12 interceptions; this season, he has 12 touchdowns and just one pick. Allen ranks third in adjusted net yards per attempt (9.3), second in passer rating (122.7), and second in QBR (89.8) after failing to rank higher than 23rd in those categories in either of his first two seasons. Pro Football Focus—an outlet that’s been so famously anti-Allen that Bills fans made anti-PFF T-shirts—lists Allen with its fifth-highest passing grade among all quarterbacks, after ranking him 40th in each of his first two seasons. Remember, there are only 32 NFL teams.

PFF highlighted two areas in which Allen struggled horribly during his first two seasons. Despite his powerful arm, he was woefully bad at completing deep passes, going 36-of-131 on targets deeper than 20 yards downfield, with 11 touchdowns and 10 interceptions. This year, he is 11-of-16 on deep throws. Allen also didn’t handle pressure well early in his career, going 94-of-261 with 10 touchdowns and nine interceptions on passes thrown while under duress. This year, he’s 26-of-44 on those throws, with six touchdowns and one pick. No other quarterback has thrown more than four touchdowns in 2020 under pressure.

As I prepare to go down with the S.S. Josh Allen Sucks, I wonder how I failed to realize that this 6-foot-5 iceberg would sink me. I analyze the quarterbacks entering the NFL every year, and generally I’m OK at it. Evaluating young quarterbacks is an inexact science, so I focus on two factors that seem immutable: Is a quarterback uniquely exceptional as an athlete, and can a quarterback throw the football accurately?

With Allen, the first answer was always an obvious yes. Allen is a ridiculous athlete, massive and mobile, with a monster arm and the ability to hurdle and juke defenders. I don’t know how many NFL players—quarterbacks or otherwise—could have opposing defenders pile onto him and just fight through them the way that Allen does.

But the second answer long appeared to be an obvious no. To some, accuracy issues might not seem like a big deal. Just teach the guy to throw better! NBA coaches have proved capable of turning players with broken shots into effective 3-point threats; pitchers who’ve struggled with accuracy issues in the minor leagues have improved their control as their careers have gone on. But quarterback accuracy, for the most part, does not seem like a teachable skill. Just ask Mike Leach, or scouts, or nerds. (Coaches, scouts, and nerds rarely agree on anything.) And Allen was spectacularly inaccurate at Wyoming. This was borne out by statistics and the tape, which showed Allen consistently missing his targets on easy passes.

The most damning statistic about Allen’s college career came from a 2018 article by Bill Connelly (now of ESPN, then of SB Nation). Connelly focused on passing success rate, a metric that measures how effective plays are, by down and distance, in helping an offense convert a first down. Among Connelly’s data set, which spanned from 2010 to 2017, no quarterback posted a higher success rate in the NFL than in college. This seemed like a bad sign for Allen, whose passing success rate at Wyoming was on par with that of Ryan Mallett at Arkansas. If a player’s college success rate represents his NFL ceiling, and Allen’s college success rate was the same as someone who served as a backup to Brian Hoyer on two separate teams, I thought he was destined to fail.

For the most part, Connelly’s analysis holds firm. To date, every QB in the 2018 draft class has a lower success rate in the pros than he did in college, including Allen, who still has a career success rate of 42 percent, below the 43.3 percent rate he posted in college. But so far this season, Allen has a 57 percent success rate—almost 14 percentage points higher than the rate he had at Wyoming.

Allen’s dramatic improvement means that one of two things must be true. The first is that it’s possible for quarterbacks to improve their ability to throw accurately well into their 20s, a lesson that would invalidate much of the current methodology behind how franchises draft quarterbacks. Many teams have come to accept that quarterback accuracy is an inherent trait that takes precedence over previously favored traits like size and arm strength. If they hadn’t, we never would’ve seen teams use no. 1 picks on players like the 5-foot-10 Kyler Murray and the 6-foot Baker Mayfield. If it turns out that quarterbacks can get better at throwing passes accurately after reaching the highest level, teams should be more willing to home in on prospects like Allen, who possess incredible physical traits but whose passing accuracy leaves plenty to be desired.

The other scenario is that Allen is a spectacular, incredible outlier. In this event, any teams that believe they should focus on finding mountain men like Allen in hopes of fixing their flaws will end up sinking valuable years and millions of dollars into doomed reclamation projects.

Bills fans probably want to believe that the first of these things is true, and that the media members like me who trashed Allen before the draft were entirely misguided in their thinking. And that may indeed be the case. Relying too heavily on statistical analysis in something as variable as quarterback scouting can lead to massive misses on players with singular traits. But Bills fans should probably hope that the second thing is more true, and that what’s happening with Allen is remarkable, irreplicable, and utterly unpredictable. Josh Allen seems to be special, and they got him. It’d be better for Buffalo if the rest of the league wasted valuable draft capital trying to find the next Josh Allen when in fact Josh Allen is the only person capable of doing the ridiculous things that Josh Allen is doing.