Dak Prescott is 19-8 as a starting quarterback the past two seasons. His 6.2 percent touchdown rate is second-best to only Joe Burrow since 2021. He is, in the purview of advanced stats like expected points added and completion percentage over expected, one of the most efficient passers in the NFL.
He also leads the league with 14 interceptions despite missing five games this season. His 95.8 passer rating is his lowest in five years. His 7.1 adjusted yards per attempt is nearing a career worst.
This makes Prescott’s Schrodinger’s quarterback. He is simultaneously good and bad, the outcome depending on your observation of him. He’s capable of roasting the Minnesota Vikings to a crisp or losing his stroke and coughing up a 28-14 fourth quarter lead to the Green Bay Packers in consecutive weeks.
But while conventional numbers aren’t keen on a quarterback who’s been the victim of some bad luck and deflected interceptions, the advanced numbers love him. Following last week’s Thursday Night Football win over the Tennessee Titans — a game in which Prescott was responsible for more than two touchdowns of added value — he’s vaulted into the top three quarterbacks, a hair behind division rival Jalen Hurts for second place behind Patrick Mahomes.
That’s enough to spark debate for weeks to come. And fortunately for us, we’ve got data on 32 other starting quarterbacks to pore through in order to parse out this week’s rankings.
We know the data is limited — but it does give us a pretty good idea of who has risen to the occasion this fall. Let’s see which quarterbacks are great and who truly stinks through seven weeks. These numbers are from the NFL’s Next Gen Stats model but compiled by the extremely useful RBSDM.com, run by The Athletic’s Ben Baldwin and Sebastian Carl.
Using expected points added (EPA, the value a quarterback adds on any given play compared to the average NFL result) along with completion percentage over expected (CPOE, the percent of his passes that are caught that aren’t expected to be in typical NFL situations) gives us a scatter plot of 33 quarterbacks (minimum 256 plays, or 16 per game) that looks like this:
The size of each dot represents the amount of plays they’ve been a part of. A place in the top right means you’re above average in both EPA and CPOE. A place in the bottom left suggests things have gone horribly wrong (i.e. Baker Mayfield).
There are a lot of players taking up the creamy middle ground and some strange outliers, making it tough to separate this year’s average quarterbacks into tiers. Here’s my crack at it, but full details follow in the text below.
