Skip to main content
ABC News
Is Josh Allen Actually Good Now?

Josh Allen has had his share of doubters since entering the league. Perhaps his most outspoken critic — Jalen Ramsey — will be playing across the line of scrimmage from him this Sunday.

Ramsey, a starting cornerback for the Los Angeles Rams, infamously described Allen as “trash” after the Bills selected him with the seventh overall pick in the 2018 draft. At the end of Allen’s 2018 rookie season, there weren’t many who would argue with Ramsey’s assessment; in its annual QB Index, the NFL network ranked him 32nd among the 33 NFL quarterbacks with at least 500 snaps. The next year, in 2019, Allen improved to 18th despite completing just 58.8 percent of his passes and averaging 6.7 yards per attempt, numbers that didn’t do much to quell his skeptics.

This year — at least so far — things have been quite different. During his first two years in the league, Allen never threw for more than 266 yards in a game. In his first two games this season, Allen has already eclipsed 300 yards passing twice, the first Bills quarterback to have back-to-back 300-yard-plus games since Drew Bledsoe in 2002. Allen is just the fourth QB in NFL history to start the season with 700-plus passing yards, six-plus touchdowns and no interceptions.

Allen’s Total Quarterback Rating (QBR) is fourth-best in the NFL, behind only Aaron Rodgers, Russell Wilson and Patrick Mahomes. That’s heady company for a guy who ranked 24th in ESPN’s signature metric both of the past two seasons. His top-line season-to-date stats are just as impressive: Allen completed 70.4 percent of his passes for 727 yards against the New York Jets and Miami Dolphins. He passed for six touchdowns and zero interceptions despite facing the fourth-most blitzes of any quarterback through Week 2. And he’s been particularly good when forced to operate on the fly: On passes outside the pocket, Allen has the third-best QBR in the league (97.3).

Allen has also made strides in the small things coaches expect from a franchise QB. In Week 1 against the Jets, facing a third-and-5 early in the second quarter, Allen coaxed an offside penalty from the Jets defensive line with his hard count. The flag was enough to give the Bills the first down, but Allen wisely wanted more. He used the free play to take a deep shot downfield to John Brown, who narrowly missed catching the ball for a 30-yard gain.

The Bills have an analytically inclined front office1 and have surrounded Allen with talent. They traded four draft picks — including a 2020 first-rounder2 — to Minnesota for wide receiver Stefon Diggs in the offseason. Diggs already has 16 catches for 239 receiving yards on the year. The offensive line has done a good job of protecting its QB, posting the sixth-best Pass Block Win Rate in the league. And Bills offensive coordinator Brian Daboll has also done his part to give Allen the best chance for success, dialing up play-action — the most efficient play type in all of football — 34.8 percent of the time, good for fourth-most frequent in the league. It’s paid off: 54.4 percent of the Bills’ passing yards have come from those play-action calls, tops in the NFL.

There’s been a lot to like in Allen’s game this year. But how confident should we be that this season is the new normal, and not just a mirage? After all, we have two seasons of Allen playing below average and just two games of him playing close to elite. One way to make sense of the numbers is to divide them into two groups: stats we know are predictive of future performance and stats that are not.

For quarterbacks, the stats that tend to be stable over time include passing efficiency from a clean pocket, passing efficiency with no play-action, completion percentage over expected and the share of throws that are off target or negatively graded. Each of these aspects of a QB’s performance is indicative of their true skill and is less affected by circumstances and luck. Stats that aren’t predictive of future performance — and that tend to be highly affected by circumstances and luck — include passing under pressure, passing outside the pocket and passing efficiency with play-action.

Allen’s numbers are great in the unstable, less predictive metrics and middle of the road in their stable, more predictive counterparts. Allen’s QBR when not under pressure (79) is good for just 19th in the league. His off-target throw percentage (16.3) has improved from previous years3 but is still good for just 16th in the NFL. And his completion percentage over expected, an NFL Next Gen Stats metric that compares the difficulty of each pass to that of a league-average quarterback in a similar situation, is just 2.1, 16th in the league.

Given the Bills’ opponents so far, this should probably give us pause. Both the Jets and the Dolphins are rated as bottom-10 teams by our Elo rating system, while the Rams, the Bills’ upcoming Week 3 opponent, are a top-10 team.

Allen has raised his ceiling this year, performing at a level many would have considered improbable before this season. But if the math holds up, and Jalen Ramsey has anything to say about the outcome of this Sunday’s game, we should probably expect Allen to begin looking a little less like Mahomes and Wilson and a bit more like the player we’ve seen take the field the past two years.

Check out our latest NFL predictions.


  1. Analytics staffers on NFL teams ranked the Bills among the top five most analytically advanced in the NFL.

  2. Minnesota used the first-round pick to select LSU receiver Justin Jefferson.

  3. It was 24.4 percent in 2018, dead last in the league, and 23 percent in 2019, also last.

Josh Hermsmeyer was a football writer and analyst.