Been waiting for comprehensive study on this for a long time and @PFF_Eric did an excellent job.

Pressure rate and sacks allowed are primarily quarterback stats, **not** offensive line stats. https://www.pff.com/news/pro-z-quarterbacks-in-control-a-pff-data-study-of-who-controls-pressure-rates
As noted in the piece, we've had evidence suggesting this for awhile, but there's a lot of new pieces strengthening the argument. For example, the stability of QB time in pocket (which drives pressure rates) https://www.pff.com/news/pro-z-quarterbacks-in-control-a-pff-data-study-of-who-controls-pressure-rates
The takeaway: do not use sacks allowed or even something like FO's Adjusted Sack Rate as measures of offensive line play.

They are (primarily) quarterback stats.

Russell Wilson getting sacked 51 times tells us more about Russell Wilson than about Seattle's offensive line.
And for those asking what we should be using instead- https://twitter.com/benbbaldwin/status/1161616659758702594
Which offensive lines were good in pass protection in 2018?

By combining ESPN's Pass Block Win Rate (pressures allowed in <2.5 seconds) and PFF grades, we can get a pretty good picture.

Dominant OLs: Packers, Browns, Bears, Chiefs, Patriots, and Rams.
Weeeeeelllllllll........ https://twitter.com/BlakeTAtkinson/status/1161621164575350784
Finally, nerds & offensive linemen can be natural allies because the nerds argue that linemen should get more credit for a team's rushing production and little of the blame for their QBs getting sacked
It's not an accident that Bills, Seahawks, and Texans fans generally think their offensive lines are bad: those were the 3 QBs with the longest time to throw in 2018 (min 200 attempts) https://twitter.com/ThatKraz/status/1161632081061064705
Here's a big long thread for those of you just waking up on the West Coast.

Sacks & pressures are primarily quarterback stats.
You can follow @benbbaldwin.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: