As has been the case for nearly two months now, the steady decline of NFL offenses continued in week 13. Scoring has dropped from 24.8 points per game last year to 22.9 this year. The league is currently averaging 11.0 yards per completion; if this holds it will be the lowest in NFL history. There’s also been a marked shift in touchdown passes vs. interceptions. Early in the season there were 3.5 TD passes for every INT; that ratio is now below 2 to 1. For the first time in several years the NFL has found a nice equilibrium between offense and defense. It’ll be interesting to see if the competition committee devises rule changes to boost offense again in 2022.
Here are the week 13 rankings:
# | QB Week 13 | Plays | PFF | QBR | zPlays | zPFF | zQBR | zAvg |
---|---|---|---|---|---|---|---|---|
1 | Tom Brady | 52 | 84.9 | 82.3 | 1.07 | 1.49 | 1.61 | 1.44 |
2 | Tua Tagovailoa | 46 | 89.1 | 65.9 | 0.46 | 1.79 | 0.93 | 1.27 |
3 | Matthew Stafford | 41 | 80.3 | 62.1 | -0.06 | 1.15 | 0.77 | 0.79 |
4 | Justin Herbert | 42 | 71.9 | 79.9 | 0.04 | 0.53 | 1.51 | 0.73 |
5 | Kirk Cousins | 47 | 79.1 | 49.8 | 0.56 | 1.06 | 0.26 | 0.72 |
6 | Gardner Minshew | 33 | 72.1 | 78.3 | -0.88 | 0.55 | 1.44 | 0.53 |
7 | Ben Roethlisberger | 35 | 75.3 | 64.3 | -0.68 | 0.78 | 0.86 | 0.51 |
8 | Joe Burrow | 48 | 76.7 | 32.6 | 0.66 | 0.88 | -0.46 | 0.44 |
9 | Carson Wentz | 28 | 69.9 | 83.9 | -1.40 | 0.38 | 1.67 | 0.42 |
10 | Taylor Heinicke | 38 | 69.5 | 65.7 | -0.37 | 0.36 | 0.92 | 0.38 |
11 | Derek Carr | 45 | 76.3 | 33.1 | 0.35 | 0.85 | -0.44 | 0.37 |
12 | Dak Prescott | 45 | 67.2 | 38.2 | 0.35 | 0.19 | -0.23 | 0.10 |
13 | Zach Wilson | 44 | 67.0 | 38.0 | 0.25 | 0.17 | -0.23 | 0.07 |
14 | Jimmy Garoppolo | 35 | 73.6 | 32.9 | -0.68 | 0.66 | -0.45 | 0.06 |
15 | Russell Wilson | 46 | 66.5 | 30.8 | 0.46 | 0.14 | -0.53 | 0.00 |
16 | Jared Goff | 48 | 59.1 | 45.5 | 0.66 | -0.41 | 0.08 | -0.05 |
17 | Josh Allen | 42 | 70.9 | 20.4 | 0.04 | 0.46 | -0.97 | -0.05 |
18 | Andy Dalton | 50 | 66.2 | 11.5 | 0.87 | 0.11 | -1.34 | -0.17 |
19 | Taysom Hill | 56 | 57.0 | 28.2 | 1.48 | -0.56 | -0.64 | -0.18 |
20 | Kyler Murray | 29 | 51.7 | 85.4 | -1.29 | -0.95 | 1.74 | -0.21 |
21 | Matt Ryan | 51 | 59.8 | 23.8 | 0.97 | -0.36 | -0.82 | -0.23 |
22 | Patrick Mahomes | 34 | 61.4 | 44.3 | -0.78 | -0.24 | 0.03 | -0.27 |
23 | Lamar Jackson | 59 | 44.2 | 42.5 | 1.79 | -1.50 | -0.05 | -0.41 |
24 | Trevor Lawrence | 37 | 56.7 | 29.1 | -0.47 | -0.58 | -0.60 | -0.57 |
25 | Teddy Bridgewater | 48 | 44.1 | 25.5 | 0.66 | -1.51 | -0.75 | -0.85 |
26 | Mike Glennon | 47 | 40.9 | 18.5 | 0.56 | -1.74 | -1.04 | -1.07 |
27 | Tyrod Taylor | 20 | 42.5 | 2.7 | -2.22 | -1.63 | -1.70 | -1.77 |
28 | Davis Mills | 18 | 36.4 | 6.2 | -2.42 | -2.07 | -1.56 | -1.99 |
This week is notable for some major disagreement between PFF grades and QBR scores. The correlation between them is just 0.53 which is amazingly low for two metrics that are essentially trying to measure the same thing. For a pair of quarterbacks, their score divergence can mostly be attributed to something rather silly – the NFL’s antiquated method for assigning fumbles.
RANT TIME
Among the things that don’t affect game outcomes and mean nothing in the grand scheme of life, the NFL’s administrative nonsense around QB fumbles really grinds my gears. As FP readers likely know, all aborted snaps and botched handoffs are charged to the QB whether the mistake was his fault or not. These plays are also classified as QB rushes even though the QB wasn’t rushing. Let’s take a look at a couple examples of why this matters…
On Sunday the Falcons’ offense botched three snaps. We can debate how much responsibility Matt Ryan bears for those miscues but they were most certainly not all his fault. As is the case with most aborted snaps, none of Atlanta’s led to turnovers and the loss of yardage was minimal. These plays are often no worse than running into a wall for a two yard loss.
If you look at the box score it lists Matt Ryan with three fumbles. Someone who didn’t know the context would think Ryan spent the game waving the ball around like he was trying to hail a taxi. But these weren’t real fumbles, at least not in the way that football fans generally think of them.
In Monday night’s game the Bills fumbled a handoff between Josh Allen and Matt Breida. It was probably 80% Breida’s fault but of course Allen gets charged with a lost fumble. Given the way Allen plays, a box score observer would naturally assume that he must have coughed up the ball on one of his crazy scramble attempts. But that wasn’t the case at all.
Back to the actual point of this rant, let’s dissect how these fumble rulings affect QBR scores (they don’t affect PFF as far as I know). We know that QBR places extreme weight on QB runs and we know that aborted snaps are treated as QB runs. This becomes a confluence of stupidity that defies any semblance of logic. According to QBR, Matt Ryan wasn’t just the worst rushing QB this week, he was more than twice as bad as the second worst. That’s entirely a product of the aborted snaps and the way they’re credited.
The situation with Josh Allen is even more ridiculous. He had a 21 yard scramble in a high leverage situation in addition to a couple of other successful runs. Normally the stuff QBR absolutely drools over. Instead, Allen is credited with negative rush EPA despite his obviously positive performance. Why? You guessed it, because he got charged with a massive EPA loss for Breida’s fumble. According to PFF Allen had the second highest rushing grade of any QB this week. Yet QBR has him third from the bottom.
Both Ryan and Allen were dealt QBR scores in the 20’s despite average overall performances. That’s what happens when you take a few plays with wonky accounting and compound the error by putting extreme weight on the subset of plays that happen to include the wonky ones. I don’t think it’s an exaggeration to infer that Ryan and Allen would’ve had QBR’s around 50 if we simply removed the questionably assigned fumbles.
One of my offseason projects will be an attempt to reverse engineer the rushing component of QBR and make it more reasonable. It’s a shame because the other components (passing, sacks, penalties) are expertly calibrated and tell a very accurate picture of how well the quarterback performed. If the rushing component were fixed, QBR would be a genuinely fantastic metric.
In the meantime I’ve decided to tweak the weights in my ranking formula to minimize some of the noise from QBR’s rushing roulette. From now on PFF grades will be weighted 1.25 units and QBR will be downgraded to 0.75 units.
With that off my chest, here are the season rankings:
# | QB 2021 Season | Plays | PFF | QBR | zPlays | zPFF | zQBR | zAvg | Diff |
---|---|---|---|---|---|---|---|---|---|
1 | Tom Brady | 563 | 92.1 | 66.2 | 1.17 | 1.81 | 1.45 | 1.58 | 0.36 |
2 | Justin Herbert | 576 | 86.0 | 67.3 | 1.31 | 1.19 | 1.55 | 1.32 | 0.36 |
3 | Kirk Cousins | 513 | 89.6 | 56.5 | 0.64 | 1.56 | 0.62 | 1.09 | 0.93 |
4 | Josh Allen | 564 | 82.2 | 57.3 | 1.18 | 0.80 | 0.69 | 0.84 | 0.11 |
5 | Aaron Rodgers | 432 | 82.7 | 65.2 | -0.23 | 0.85 | 1.37 | 0.79 | 0.52 |
6 | Matthew Stafford | 498 | 78.4 | 65.9 | 0.47 | 0.41 | 1.43 | 0.73 | 1.02 |
7 | Kyler Murray | 365 | 82.7 | 65.6 | -0.96 | 0.85 | 1.40 | 0.65 | 0.55 |
8 | Joe Burrow | 461 | 87.1 | 48.7 | 0.08 | 1.30 | -0.04 | 0.65 | 1.34 |
9 | Derek Carr | 541 | 80.4 | 55.0 | 0.94 | 0.61 | 0.49 | 0.64 | 0.12 |
10 | Dak Prescott | 505 | 83.7 | 49.7 | 0.55 | 0.95 | 0.04 | 0.60 | 0.91 |
11 | Ryan Tannehill | 494 | 81.4 | 53.8 | 0.43 | 0.72 | 0.39 | 0.56 | 0.32 |
12 | Mac Jones | 466 | 83.0 | 51.4 | 0.13 | 0.88 | 0.19 | 0.52 | 0.69 |
13 | Patrick Mahomes | 579 | 73.2 | 56.5 | 1.35 | -0.13 | 0.62 | 0.39 | 0.75 |
14 | Jalen Hurts | 514 | 80.0 | 48.0 | 0.65 | 0.57 | -0.10 | 0.38 | 0.68 |
15 | Carson Wentz | 518 | 73.0 | 58.5 | 0.69 | -0.15 | 0.79 | 0.30 | 0.94 |
16 | Tua Tagovailoa | 329 | 78.8 | 58.2 | -1.34 | 0.45 | 0.77 | 0.19 | 0.32 |
17 | Matt Ryan | 501 | 76.5 | 48.0 | 0.51 | 0.21 | -0.10 | 0.18 | 0.32 |
18 | Lamar Jackson | 567 | 70.1 | 50.3 | 1.22 | -0.44 | 0.09 | 0.05 | 0.54 |
19 | Jimmy Garoppolo | 373 | 73.7 | 56.3 | -0.87 | -0.07 | 0.61 | -0.03 | 0.68 |
20 | Teddy Bridgewater | 463 | 74.3 | 47.4 | 0.10 | -0.01 | -0.16 | -0.03 | 0.14 |
21 | Jacoby Brissett | 278 | 76.0 | 50.4 | -1.89 | 0.16 | 0.10 | -0.27 | 0.06 |
22 | Daniel Jones | 472 | 71.6 | 41.0 | 0.20 | -0.29 | -0.70 | -0.32 | 0.41 |
23 | Taylor Heinicke | 492 | 65.9 | 50.1 | 0.41 | -0.87 | 0.08 | -0.33 | 0.95 |
24 | Russell Wilson | 328 | 73.9 | 44.5 | -1.35 | -0.05 | -0.40 | -0.42 | 0.35 |
25 | Ben Roethlisberger | 479 | 58.8 | 44.3 | 0.27 | -1.60 | -0.42 | -0.87 | 1.18 |
26 | Baker Mayfield | 389 | 65.3 | 37.9 | -0.70 | -0.94 | -0.97 | -0.90 | 0.03 |
27 | Trevor Lawrence | 541 | 59.4 | 33.7 | 0.94 | -1.54 | -1.33 | -0.98 | 0.22 |
28 | Sam Darnold | 381 | 62.6 | 35.2 | -0.78 | -1.21 | -1.20 | -1.12 | 0.02 |
29 | Jared Goff | 481 | 60.3 | 30.8 | 0.29 | -1.45 | -1.57 | -1.14 | 0.12 |
30 | Justin Fields | 299 | 64.2 | 25.8 | -1.67 | -1.05 | -2.00 | -1.46 | 0.95 |
31 | Zach Wilson | 291 | 58.5 | 25.0 | -1.75 | -1.63 | -2.07 | -1.79 | 0.43 |
32 | Davis Mills | 270 | 56.2 | 30.4 | -1.98 | -1.87 | -1.61 | -1.81 | 0.26 |
What are your thoughts?