Who is the best QB in the NFL?

In 2014, Aaron Rodgers was 31 and on top of the NFL. After his second MVP season in four years, one that ended in a 12-4 record and an appearance in the NFC Championship game, there was little doubt he was the best quarterback in the league, if not the best ever.

And then, poof. Aside from a brief stretch at the end of 2016, we’ve rarely seen that player again.

The Packers are 5-1 and likely to make the playoffs, but their record is more a reflection of a revamped defense — now No. 4 in the league in takeaways with three wins when holding opponents to 16 points or fewer — than consistency on offense. Despite moving on from former coach Mike McCarthy, the Packers’ struggles in the passing game represent a continuation of the past few years, raising the question of whether we will ever see them reach the level of the early-2010s offense again.

The dropoff has been stark. From 2009 through 2014, Rodgers finished a close No. 2 to Peyton Manning among 60 qualified quarterbacks (minimum 500 plays) in Expected Points Added (EPA) per play, a measure of efficiency that takes into account situational factors such as down, distance and field position. However, since 2015, Rodgers’ per-play efficiency has fallen outside of the top 10, currently ranking No. 13 of 48 and behind players such as Carson Wentz, Matt Ryan and Kirk Cousins.

Let’s take a look at the story of Rodgers’ career to date as told by EPA per play. In short, EPA is a data-driven measure that captures the extent to which each play increases or decreases a team’s likelihood of scoring. For those unfamiliar with EPA, see this explainer here.



The figure above shows a rolling average of Rodgers’ 300 most recent plays, starting with his second season as a starter in 2009 (the first year for which EPA data from nflscrapR exist). To place Rodgers among other quarterbacks, I’ve added percentile lines denoting where each performance ranks among the entire sample of 300-play slices of quarterback performance in a given year. Because there are 32 teams, let’s think of the 90th percentile as a top-three performance, the 75th percentile as top eight and the 50th percentile as top half.

Starting in 2011, Rodgers was around or above the red-dashed 90th-percentile line of quarterback performance for four straight years, reaching far above the line in his 2011 and 2014 MVP seasons. Calling his performance during this period 90th percentile is selling him short: Rodgers’s 300-play peaks in 2011 and 2014 have been exceeded by only two quarterbacks during that time period: Manning in 2013 and Drew Brees in 2018 (prior to Brees’ late-season struggles).

However, Rodgers’ efficiency dropped steeply beginning in 2015. Since then, he has rarely reached the 90th percentile — or even the 75th percentile — of quarterback efficiency. Hovering between the 50th and 75th percentiles would place Rodgers in the Nos. 9-16 range of quarterbacks; in the upper half of the league but not in the top-three range we’re accustomed to. And aside from a return to form at the end of 2016, he has remained there for four-plus seasons.

Statistics are useful for telling us what has happened. It is inarguable Green Bay’s passing offense has declined from one of the most efficient in the league to something closer to league average.