Fun Stat Nerd Tidbit: Tennessee - Ranking Linebackers

In today's Tennessee profile, I mention the following:

With the rankings above, one can begin to zero in on where problems may have originated. Tennessee ranked 49th in Adj. Line Yards but 73rd in Rushing S&P+, suggesting their linebackers were perhaps giving away some of the gains made up front. Meanwhile, their Need for Blitzes figure was incredibly low, suggesting most of UT's mediocre pass rush came from the line. Only one linebacker managed even 5.0 TFL/sacks, which is a low number as well. In all, the linebackers left something to be desired, which is not altogether surprising given the unit's injury history and lack of depth.

One of the things I would love to do as time progresses is figure out the most accurate way to begin ranking different units. On offense, it's easy enough to begin figuring out -- measure running backs by adding up Adj. POE, measure quarterbacks by somewhat doing the same (with perhaps some receiver targets/catches data thrown in), measure offensive and defensive lines by looking at Adj. Line Yards and Adj. Sack Rates. But evaluating a unit like linebackers is really difficult for obvious reasons. They have a part to play in run defense, pass rushing and pass coverage. But what if we were to start with exactly what I mentioned above?

Adj. Line Yards vs Rushing S&P+

By comparing these two measures, we are, in theory, isolating how much the line was responsible for rushing success (or lack thereof) and how much everybody else was. And while the secondary obviously has a role to play in stopping the run, most of it comes from the front seven. So the difference between these two measures gives us a good hint for how the LBs held up.

Adj. Sack Rate vs Need For Blitzes

Here's a reminder of what the Need For Blitzes concept measures:

Need for Blitzes: This compares a team's standard downs sack rate to its passing downs sack rate.  The lower the number, the more they were able to generate pressure on standard downs, and the less need for blitzes.

You have to make generalizations in these stats, and the one we are making here is simple: teams aren't going to blitz nearly as much on standard downs (first-and-10, second-and-4) as passing downs (second-and-9, third-and-7), so a team's passing downs sack rate tells us much more about its linebackers' ability to attack the passer. Obviously if you're running a 3-4, you're much more likely to send players on standard downs, but ... again, it's about generalizations.

There is still an aspect of pass coverage that we have to account for, and I'm very much open to conversation about that one, but we can still get quite a bit of the way toward evaluating linebackers by using these two concepts. So, naturally, let's see what a rough rankings system would look like using this. As always, ratings are set up so that 100.0 is perfectly average, above 100 is good, and below 100 is bad.

Rest of the list after the jump.

This list needs further calibration (among other things, I'm pretty sure the Purdue linebackers are getting a rather hefty boost from defensive end Ryan Kerrigan's general badass tendencies), and it's rather confused by West Virginia's 3-3-5, but ... it's not bad. It certainly seemed to lock in on the poor units, anyway. Tennessee doesn't rank well, and it's hard to make the case that a team like Michigan or Minnesota should have ranked any higher (right, Maize n Brew?).

Anyway, consider this an experiment more than a declaration. Feedback is more than encouraged.