This week I’m excited to introduce a new data set into my quarterback rankings, courtesy of Pro Football Focus. I’d like to give a helmet knock to our good buddy Bryan Frye for working out some details behind the scenes to make this possible. [1]Note from Bryan: all I did was ask.
From this point forward, I will be using two metrics to rate quarterbacks: QBR and PFF offensive grades. This makes me giddy because both systems attempt to isolate the QB’s contribution from that of his teammates. That’s a significant step up from ANY/A, DYAR, and EPA which simply assign team offensive statistics to the QB taking the snaps. I can live with that for historical comparisons where we don’t have anything better, but in today’s world of robust data there’s no reason to settle for such a high degree of entanglement.
As neither QBR nor PFF grades account for workload, I needed to make an adjustment to prevent low usage QB’s from hogging the top of the rankings. After experimenting with a few ideas I settled on adding a z-score for play count (based on qualifiers only) and giving it half weight compared to the z-scores for the two metrics. It’s not perfect but it gets the job done without too many arbitrary decisions.
Let’s see how the new system looks for week 12:
# | QB Week 12 | Plays | PFF | QBR | zPlays | zPFF | zQBR | zAvg |
---|---|---|---|---|---|---|---|---|
1 | Aaron Rodgers | 48 | 85.8 | 64.6 | 0.93 | 1.86 | 0.91 | 1.29 |
2 | Joe Burrow | 28 | 91.9 | 76.6 | -1.38 | 2.34 | 1.48 | 1.25 |
3 | Carson Wentz | 53 | 79.6 | 61.4 | 1.50 | 1.38 | 0.76 | 1.16 |
4 | Josh Allen | 41 | 78.0 | 75.6 | 0.12 | 1.25 | 1.44 | 1.10 |
5 | Derek Carr | 50 | 64.5 | 78.8 | 1.16 | 0.20 | 1.59 | 0.95 |
6 | Dak Prescott | 54 | 74.2 | 46.5 | 1.62 | 0.95 | 0.05 | 0.73 |
7 | Teddy Bridgewater | 21 | 71.0 | 84.5 | -2.19 | 0.70 | 1.86 | 0.59 |
8 | Justin Herbert | 58 | 55.7 | 59.4 | 2.08 | -0.49 | 0.67 | 0.49 |
9 | Andy Dalton | 44 | 70.6 | 50.3 | 0.47 | 0.67 | 0.23 | 0.46 |
10 | Tom Brady | 41 | 73.7 | 42.7 | 0.12 | 0.92 | -0.13 | 0.34 |
11 | Jared Goff | 32 | 67.7 | 60.8 | -0.92 | 0.45 | 0.73 | 0.29 |
12 | Taylor Heinicke | 43 | 66.2 | 49.8 | 0.35 | 0.33 | 0.21 | 0.29 |
13 | Daniel Jones | 42 | 66.2 | 47.2 | 0.23 | 0.33 | 0.09 | 0.21 |
14 | Jimmy Garoppolo | 33 | 69.2 | 52.4 | -0.80 | 0.56 | 0.33 | 0.20 |
15 | Mac Jones | 37 | 51.9 | 72.9 | -0.34 | -0.79 | 1.31 | 0.14 |
16 | Tua Tagovailoa | 39 | 61.1 | 47.1 | -0.11 | -0.07 | 0.08 | -0.02 |
17 | Trevor Lawrence | 51 | 65.4 | 23.6 | 1.27 | 0.27 | -1.04 | -0.05 |
18 | Russell Wilson | 37 | 57.5 | 53.5 | -0.34 | -0.35 | 0.39 | -0.05 |
19 | Matthew Stafford | 42 | 58.8 | 45.0 | 0.23 | -0.25 | -0.02 | -0.06 |
20 | Baker Mayfield | 42 | 49.2 | 45.1 | 0.23 | -1.00 | -0.01 | -0.36 |
21 | Kirk Cousins | 36 | 57.9 | 32.8 | -0.46 | -0.32 | -0.60 | -0.46 |
22 | Tyrod Taylor | 34 | 62.5 | 22.1 | -0.69 | 0.04 | -1.11 | -0.56 |
23 | Ben Roethlisberger | 45 | 50.0 | 28.3 | 0.58 | -0.94 | -0.81 | -0.58 |
24 | Ryan Tannehill | 29 | 54.8 | 37.6 | -1.26 | -0.56 | -0.37 | -0.63 |
25 | Jalen Hurts | 39 | 56.3 | 17.3 | -0.11 | -0.44 | -1.33 | -0.73 |
26 | Lamar Jackson | 51 | 41.7 | 25.0 | 1.27 | -1.58 | -0.97 | -0.77 |
27 | Matt Ryan | 35 | 45.3 | 31.3 | -0.57 | -1.30 | -0.67 | -0.90 |
28 | Trevor Siemian | 36 | 50.8 | 14.2 | -0.46 | -0.87 | -1.48 | -1.03 |
29 | Zach Wilson | 30 | 44.1 | 9.8 | -1.15 | -1.40 | -1.69 | -1.46 |
30 | Cam Newton | 28 | 37.6 | 5.4 | -1.38 | -1.90 | -1.90 | -1.80 |
All things considered I think this passes the smell test. There are few instances in which PFF and QBR disagree sharply; the most notable is in regard to Derek Carr‘s Thanksgiving game in Dallas. QBR gives Carr credit for drawing four interference flags for 91 yards, all on third down. The PFF graders apparently weren’t as impressed by those throws. The crediting of quarterbacks for drawing penalties is a gray area that has no right answer, thus I’m glad we have representation on both sides of the debate.
Justin Herbert also sparked major disagreement between the metrics, with QBR liking his performance in Denver far more than PFF does. This one surprises me a bit considering Herbert threw a pick-six on a pass that bounced off his receiver’s hands. Normally PFF will absolve the QB of any wrongdoing on such plays while QBR will only give him a partial break.
One player PFF favored this week was Trevor Lawrence. His QBR was terrible to match his box score numbers but he actually graded above average against Atlanta. I didn’t see much of that game (who did?) so I can’t venture a guess as to why he graded relatively well. The Jags are a dysfunctional mess so it’s not hard to imagine that Lawrence’s teammates may have wasted some of his good plays.
Now let’s check out the season rankings:
# | QB 2021 Season | Plays | PFF | QBR | zPlays | zPFF | zQBR | zAvg | Diff |
---|---|---|---|---|---|---|---|---|---|
1 | Tom Brady | 511 | 91.5 | 63.6 | 1.00 | 1.75 | 1.21 | 1.38 | 0.54 |
2 | Justin Herbert | 534 | 85.8 | 65.9 | 1.26 | 1.17 | 1.40 | 1.28 | 0.23 |
3 | Kirk Cousins | 466 | 88.5 | 57.0 | 0.48 | 1.44 | 0.64 | 0.93 | 0.80 |
4 | Josh Allen | 522 | 82.0 | 60.6 | 1.12 | 0.78 | 0.95 | 0.92 | 0.17 |
5 | Aaron Rodgers | 432 | 82.7 | 64.9 | 0.09 | 0.85 | 1.32 | 0.89 | 0.47 |
6 | Kyler Murray | 336 | 85.6 | 63.5 | -1.00 | 1.15 | 1.20 | 0.74 | 0.05 |
7 | Matthew Stafford | 457 | 76.2 | 65.8 | 0.38 | 0.19 | 1.39 | 0.71 | 1.21 |
8 | Derek Carr | 496 | 79.1 | 57.2 | 0.82 | 0.48 | 0.66 | 0.62 | 0.18 |
9 | Ryan Tannehill | 494 | 81.4 | 53.4 | 0.80 | 0.72 | 0.34 | 0.58 | 0.38 |
10 | Mac Jones | 460 | 83.9 | 52.6 | 0.41 | 0.97 | 0.27 | 0.58 | 0.71 |
11 | Dak Prescott | 460 | 84.4 | 50.6 | 0.41 | 1.02 | 0.10 | 0.53 | 0.93 |
12 | Joe Burrow | 413 | 86.2 | 51.3 | -0.12 | 1.21 | 0.16 | 0.52 | 1.05 |
13 | Patrick Mahomes | 545 | 73.7 | 57.4 | 1.38 | -0.07 | 0.68 | 0.52 | 0.75 |
14 | Jalen Hurts | 514 | 80.0 | 48.0 | 1.03 | 0.57 | -0.12 | 0.39 | 0.70 |
15 | Carson Wentz | 490 | 72.3 | 56.7 | 0.76 | -0.21 | 0.62 | 0.31 | 0.83 |
16 | Matt Ryan | 450 | 78.0 | 50.5 | 0.30 | 0.37 | 0.09 | 0.24 | 0.28 |
17 | Lamar Jackson | 508 | 73.5 | 51.1 | 0.96 | -0.09 | 0.14 | 0.21 | 0.23 |
18 | Teddy Bridgewater | 415 | 78.7 | 50.1 | -0.10 | 0.44 | 0.05 | 0.18 | 0.39 |
19 | Jimmy Garoppolo | 338 | 72.3 | 58.6 | -0.98 | -0.21 | 0.78 | 0.03 | 0.99 |
20 | Tua Tagovailoa | 283 | 73.5 | 57.7 | -1.60 | -0.09 | 0.70 | -0.08 | 0.79 |
21 | Jacoby Brissett | 278 | 76.0 | 50.5 | -1.66 | 0.17 | 0.09 | -0.23 | 0.08 |
22 | Daniel Jones | 472 | 71.6 | 40.9 | 0.55 | -0.28 | -0.73 | -0.30 | 0.45 |
23 | Taylor Heinicke | 454 | 64.8 | 48.8 | 0.35 | -0.98 | -0.06 | -0.34 | 0.92 |
24 | Russell Wilson | 282 | 73.9 | 46.2 | -1.62 | -0.05 | -0.28 | -0.45 | 0.23 |
25 | Baker Mayfield | 389 | 65.3 | 37.9 | -0.40 | -0.93 | -0.98 | -0.84 | 0.06 |
26 | Ben Roethlisberger | 444 | 56.4 | 43.1 | 0.23 | -1.84 | -0.54 | -0.91 | 1.30 |
27 | Trevor Lawrence | 504 | 59.8 | 33.9 | 0.92 | -1.49 | -1.33 | -0.94 | 0.16 |
28 | Sam Darnold | 381 | 62.6 | 35.6 | -0.49 | -1.20 | -1.18 | -1.05 | 0.02 |
29 | Jared Goff | 433 | 60.6 | 29.3 | 0.11 | -1.41 | -1.72 | -1.23 | 0.31 |
30 | Justin Fields | 299 | 64.2 | 25.9 | -1.42 | -1.04 | -2.01 | -1.50 | 0.97 |
31 | Davis Mills | 252 | 59.3 | 30.5 | -1.96 | -1.54 | -1.62 | -1.65 | 0.07 |
32 | Zach Wilson | 247 | 56.3 | 23.5 | -2.02 | -1.85 | -2.21 | -2.03 | 0.36 |
Well, swapping out EPA for PFF grades definitely shuffled the top of the leaderboard. Tom Brady and Kirk Cousins (yes, the guy I dissed last week) rise to first and third place on the strength of their PFF evaluations. Matthew Stafford tumbles way down because PFF grades him as barely above average in spite of his still strong QBR and box score stats. As someone who long believed that Sean McVay’s scheme carried Jared Goff, I fully support the notion that the scheme is also making Stafford look better than he really is.
To highlight the contrast between PFF and QBR I added the ‘Diff’ column; this is the absolute value difference between the z-scores for the two metrics. Sort by Diff and you’ll see which QB generates the biggest disagreement. It turns out that Ben Roethlisberger divides the metrics even more strongly than Stafford does. QBR sees Big Ben as merely below average while PFF views him as being washed beyond repair.
I hope you guys enjoy the new rankings. Comments and questions are always encouraged!
References
↑1 | Note from Bryan: all I did was ask. |
---|