Towards a better understanding of pressure rates and offensive line play

This all got started by an interesting comment by skepticismissurvival on the NFL subreddit, who made an observation that, like most good observations, seems obvious in hindsight: the pressure rate allowed by an offensive line is not a great measure of pass protection because it depends in part on how long their QB holds the ball. There is wide variation in release time, which varied from Alex Smith at 2.38 seconds to Tyrod Taylor at 3.12 seconds in 2016. We should not expect the Chiefs and Bills to allow similar pressure rates because the Bills are protecting a QB who holds the ball longer. This post uses the following data:

Here is a raw plot of time to throw against pressure rate, along with a fitted linear trend line and 95% confidence interval.

We can see that there is a relationship between time to throw and pressure rate (note the upward-sloping trend line), but there are some big outliers. The biggest outlier is Goff, who is so much of an outlier that removing him increases the R^2 from .25 to .39, a massive change.

We also see that Alex Smith and Tyrod Taylor are both pressured almost exactly as often as predicted by the best fit line. In other words, assuming we can think of this as a measure of pass protection, the Chiefs and Bills lines protect their QBs almost as well as each other given their QBs’ release times.

By running a regression of pressure rate on time to throw and plotting the residuals, we can see which quarterbacks are pressured more often (above the red line) or less often (below the red line) than expected, given how long they hold the ball.

What I like about this is that the teams that are typically thought to have good offensive lines have their QBs fall in the “good” region (Raiders, Steelers, Titans, Saints) and the teams with bad offensive lines fall in the “bad” region (the entire NFC West), meaning that this might be a promising way to measure a team’s pass protection.

By this measure, here are the best 5 offensive lines (measured by the distance from the QB to the red line in the graph above):

  1. Steelers (best)
  2. Titans (2nd best)
  3. Bears
  4. Raiders
  5. Redskins

And here are the worst 5 offensive lines:

  1. Rams (worst)
  2. Seahawks (2nd worst)
  3. Cardinals
  4. 49ers
  5. Colts

Here is the full list. Negative numbers are good (pressured less frequently than expected given time to release):

  • evo34

    I like the effort, but pressure rate is also a function of how the QB moves to avoid pressure. E.g., Ben has a low rate bc he is very good at knowing when/where to slide to create room.

    • Ben B

      Yep, agreed. Brees is another example at someone who is great at moving in the pocket. No one stat will be perfect but I think this is more informative than looking at unadjusted pressure rates.

      • evo34

        True. As an aside, I think part of Wilson’s high pressure rate is the fact that he loves to bail on a play within a couple seconds, run backwards and spin 360 to end up awkwardly scrambling to his left well behind the line of scrimmage. He has the legs and arm to pull it off occasionally, but the move usually gets him into more trouble than his (admittedly terrible) line creates. I’m a huge fan of his talent, but I hope he either gets a better line or finds less risky ways to deal with pressure.