Week 3 Bayesian Quarterback Rankings

Tua keeps rising, maybe the pre-2022 data on Deshaun Watson is irrelevant, and Sam Howell emerging as a possible late-round success

The numbers are really starting to believe in Brock Purdy

I like to think of my Bayesian quarterback rankings as a check on perceptions, which sometimes are more correct than the cold, hard evidence, but more often not. For more details on the methodology and how I’ve leveraged Bayesian updating in the past, check out my rankings post leading into Week 2.

For those of you looking for the Adjusted Quarterback Efficiency (AQE) numbers, it might be a week (or two) before I have the data to produce them. I’m figuring it all out now and appreciate the patience.

One thing to think about, especially early in the season, is how much luck (or unlucky) plays can affect EPA per play. A decent way to add some signal is to consider PFF grading - accounts for turnover-worthy plays, etc - in addition to EPA. You’ll see the in plot below that Jordan Love was, by far, the highest difference between his efficiency in EPA and what you would have expected by his pedestrian PFF grade. There are also quarterbacks like Desmond Ridder, who haven’t been great by EPA, but even worse in grading, mostly due to dropped interceptions.

On the other side, quarterbacks like Trevor Lawrence aren’t going to get as big of a boost in these rankings as observers who agree with his fifth-highest PFF grade would expect. In the case of Lawrence, this is primarily driven by a hugely costly interception that wasn’t perfectly placed, but bounced off of the hands of his targeted receiver Tank Bigsby, plus a few great throws that were inches away from touchdowns, if the receivers could have kept their feet in.

WEEK 3 PROJECTED EFFICIENCY

These results are the ranking for the go-forward projections of quarterback efficiency this season. I also included the EPA per play rankings for each quarterback over the last six seasons (minimum 250 dropbacks in previous years, 30 in 2023) so you can see the evidence going into the projections.

Older data is decayed over time, so the 2023 EPA per play data matters more than those from pre-2020. That said, older data can’t be fully discounted, or else you miss bounce-back performers of great quarterback returning to form, like Aaron Rodgers in 2020 and 2021.

There are no changes, as of now, with the expected starters in Week 2. The biggest question mark at the time is Joe Burrow’s status for the next couple weeks.