The entire football world responded strongly to Pro Football Focus (PFF) naming Myles Garrett to the PFF first team all-pro list over T.J. Watt. It seems like everyone has an opinion on the subject. I figured they, like most sites, did it to get a lot of traffic by ticking off Steelers fans and delighting Steelers haters. Trust me, anything negative about the Steelers drives clicks, even on this site.
So when I saw PFF’s lead NFL analyst post a thread on Twitter defending their choice, I decided it might be worth my time looking into it. And this article is my response. Let’s get started.
1. The sack total is closer than it looks. Half sacks are screwing Garrett.— Sam Monson (@PFF_Sam) January 11, 2022
He sacked the QB 18 times. Remember, NFL divides up sacks not because they believe they're qualitatively worse plays, just to make the box score add up.
This is a legit argument, half sacks are a big factor in sack totals, and they are only half sacks because the NFL doesn’t want defenders on a team to have more sacks than their team does. So if you are going to give two players credit for sacking the quarterback, they each get half a sack.
And he’s right, Garrett had 4 half-sacks in 2021, and if you count them all as full sacks, his number goes from 16 to 18.
But he, surprisingly, leaves out that T.J. Watt was credited with 3 half-sacks, and those would bring his 22.5 sacks to 24 sacks.
So, in this case “closer than it looks” equates to a 6.5 sacks lead for T.J. Watt grossly overstating what should be only a 6 sack lead for T.J. Watt.
Every half-sack counts when you are justifying your
clickbait logical and unbiased decision.
The difference between Watt and Garrett's sacks could easily be 2 games vs Roethlisberger.— Sam Monson (@PFF_Sam) January 11, 2022
You can see these unrealized pressure plays on tape. We track them.
Winning rush-reps but the ball's gone before QB can feel pressure. Garrett has 56 of them. Watt 26.
So playing the Steelers screwed Myles Garrett because you can’t sack a quarterback that throws in 2.2 seconds.
Except that problem seems to be a Myles Garrett specific problem because the Steelers allowed 38 sacks on the season, a number that puts them right in the middle of the league. Funnier still, Mason Rudolph played in one game, all four quarters and a full overtime and was never sacked.
But I will agree that a large portion of T.J. Watt and Myles Garrett’s sack difference was from those two head-to-head games. That’s not really debatable.
In 16 games Ben Roethlisberger was sacked 38 times. Myles Garrett got one of those sacks. 2.6% of the times Ben Roethlisberger was sacked, it was Myles Garrett. In fact, the Browns sacked Roethlisberger twice in each game they faced him, those 2 sacks are pretty close to the 2.3 sacks the Steelers allowed per game over the course of the season.
So it wasn’t even the Browns struggling to sack Ben Roethlisberger, it was specifically Myles Garrett. And how did the Steelers manage to stymie the PFF 1st team all-pro? By putting rookie Dan Moore Jr. on him. PFF gave Moore Jr. a zero score for pass blocking, and Myles Garrett recorded one sack.
But he’s only half the equation, because while Garrett recorded 1 sack in their head-to-head matchup, T.J. Watt recorded 5.5 sacks (6 if you go by PFF counting). Those 5.5 sacks are 11.2% of the Browns total sacks given up.
That’s an impressive number, but even more impressive, Watt narrowly missed several other sacks because he was half a second later to the quarterback than a team mate. And those team mates got to Baker Mayfield. The Steelers recorded 13 sacks against the Browns in two games this season.
Clearly the Browns are a much worse pass blocking team than the Steelers then, right?
(Spoiler: this is my favorite part of this whole article.)
The Browns allowed 49 sacks in 2021, the Steelers allowed 38. In their head-to-head matchups, the Steelers sacked Baker Mayfield 13 times, while the Browns sacked Ben Roethlisberger 4 times.
If you subtract out their head-to-head games, the Steelers gave up 34 sacks to the Browns 36.
The reason the Browns look so much worse than the Steelers at protecting their quarterback is because they faced the Steelers twice, and in those two matchups the Steelers accounted for 26.5% of the Browns season sack total.
While PFF is right that most of the difference between T.J. Watt and Myles Garrett’s sack totals can be accounted for in the Browns and Steelers head-to-head matchups, that difference is because of T.J. Watt’s dominance against a division rival in two games that helped put the Steelers in the playoffs, and had Cleveland won them both they’d be in the playoffs instead.
I have a hard time saying the gap between two players is smaller than the stats say because one of them dominated the other, leading his team to a win that was necessary for making the playoffs. That sounds like an argument for T.J. Watt to me.
Bottom line: Both guys had elite seasons— Sam Monson (@PFF_Sam) January 11, 2022
Their PFF grades are both fantastic
Both guys over 90 as rushers
Watt had more splash plays, Garrett had more winning pass-rush reps. On balance, the edge goes Garrett when tallying up ALL plays
I'm sure Steelers fans are now convinced
I’ll take these ones by number:
No. 3. Pass rush win rate is a cool thing, although, it is subjective, just like pressures. It’s pretty clear whether a quarterback is sacked, and PFF even has problems with half-sacks, but they are perfectly fine with one of their employees watching a play and assigning every single pass rusher a win or loss on every snap. I don’t know if they allow ties, I don’t know their qualifications for a win, so I can’t refute this at all. I can say that ESPN has a similar gap between T.J. Watt and Myles Garrett, with Garrett at 28% and Watt at 24%. Interestingly, ESPN has Garrett ranked No. 1 in pass rush win rate, while PFF has him ranked 3rd.
Although, to be fair, if PFF wants to discount Myles Garrett’s lack of sacks against the Steelers, I’d be interested in seeing how Myles Garrett would fare without the 46.7% win rate they gave him against the Steelers in week 8, the game they reference as cheating him out of sacks. Maybe the real difference between Myles Garrett and T.J. Watt for PFF is based on the fact that they viewed Garrett’s one sack, one QB hit game against Dan Moore Jr. as one of the best games a pass rusher had this season.
No. 4. Again, what constitutes a “clean-up play”. I haven’t broken down every T.J. Watt sack for what is a clean-up sack and what isn’t, let alone Myles Garrett. But I’ve seen every sack by both, and they both chase down quarterbacks that are already under pressure a lot. You get a lot of sacks by being the player that finishes pressure. I’d love to see the cutoff point where a sack goes from not a clean-up sack to a clean-up sack. I also find it interesting that while he says the difference is a lot, he also considered a .5 sack difference to be worth his No. 1 point and spent two tweets talking about Ben Roethlisberger getting rid of the ball fast...Is it too hard to actually type out the percentages here? It wouldn’t take that many characters and if it is that big of a difference, then let us see it. Not that I’m a big fan of any of these opinion-based stats, but I’d like to at least see the number.
No. 5. Absolutely agree with this point. Every game matters, and the Steelers went 0-4-1 when T.J. Watt missed half the game or more, every single one of those games mattered. I’m not going to rate a player higher because he wasn’t on the field. I know some people want to give Watt a bonus to his numbers because he missed time, but that’s not how it works. If we are going by that measure, maybe Tyson Alualu should be the team MVP, because the difference in the defense for the 57 snaps he was on the field and the rest of the season was crazy, in the worst possible way. I don’t read this as him directly penalizing Watt for missing games, but just rejecting arguments that his stats should be higher because he played less. You don’t get imaginary stats added on because you didn’t play.
Now I want to cover the last part, the comparison.
Watt had more splash plays, Garrett had more winning pass-rush reps. On balance, the edge goes Garrett when tallying up ALL plays.
Myles Garrett did rate higher in pass rush win rate than T.J. Watt. But that is it. When you tally up all plays, you don’t just count sacks and pressures and pass rush win rate, there’s more to the game of football than rushing the passer. Let’s look at some other numbers:
Passes defended: Watt 7, Garrett 3.
Tackles for a loss: Watt 21, Garrett 17.
Solo tackles: Watt 48, Garrett 33.
Fumbles forced: Watt 5, Garrett 1.
Fumbles recovered: Watt 3, Garrett 1.
I don’t have PFF’s run defense win rate (or whatever they have for run defense), and he didn’t list it in this thread, but I can see ESPN having T.J. Watt 7th in run stop rate, with Myles Garrett not making the top 10.
PFF can do all the mental gymnastics to downplay T.J. Watt’s advantage in actually sacking the quarterback all they want, but all they are really saying is that they gave Myles Garrett a higher pass rush win rate than T.J. Watt.
That’s their entire argument. They don’t get into run defense at all, despite their “ALL plays” focus, and discounting T.J. Watt’s game changing and season altering splash plays in favor of how often they were really impressed at how well Myles Garrett beat a rookie left tackle while having minimal impact on the actual outcome of the game is really all you need to know about how they choose their all-pro team.
But here’s a stat I think sums up each player’s value to their team better than anything else.
When T.J. Watt played 50% or more of the snaps, the Steelers were 9-3, when he didn’t they were 0-4-1.
When Myles Garrett played 50% or more of the snaps, the Browns were 3-8, when he played less than half the game on defense the Browns were 5-1.
Obviously Football is a team sport, but every game T.J. Watt wasn’t on the field for the Steelers it noticeably affected the entire defense, and the Steelers ability to win a game. You can’t make that same argument for Myles Garrett.