Jump to content
Sign in to follow this  
soulman

A Player-By-Player Examination of Chicago’s 2018 Defense

Recommended Posts

Posted (edited)

A Player-By-Player Examination of Chicago’s 2018 Defense

https://dabearsblog.com/2019/a-player-by-player-examination-of-chicagos-2018-defense

5fb7dabcc9ddb6eb415d87bdfbe6736d?s=16&d= Johnathan Wood | July 1st, 2019

 

data-entry.jpg?resize=736%2C300

As a unit, the Bears defense finished 3rd in the NFL in yards, 1st in points, and 1st in turnovers. Now I want to look at the impact each individual player had on the defense, as much as is possible. To do so, I’m using stats from The Quant Edge.


Defensive Line

Let’s start on the defensive line, where we can get a look at how often each player was on the field and how successful the defense was when they were in or out of the game. This can be measured through yards per carry (YPC), yards per pass attempt (YPA), and success rate, which is generally a measure of how effectively offenses stay ahead of the chains. Higher success rate

The table below shows data for defensive linemen who both played and missed 100 or more snaps, and is set up such that numbers for each category are in game/out of game for an easy comparison. Notable differences are highlighted in green (good) or red (bad).


DL-splits.png?resize=750%2C200


A few thoughts:

  • First, note that this data does not necessarily mean a player was good or bad, especially when we get to the smaller sample sizes (in terms of snaps played or snaps missed). But it can be really useful for players in the middle, who both played a lot of snaps and rotated out for plenty as well.
  • Speaking of those players, hello Eddie Goldman and Akiem Hicks! Look at those splits against the run. Those two make up a formidable duo up front for the Bears, and allow them to be stout against the run even in nickel looks when they only have two defensive linemen on the field. If you’re looking for more specifics here, Jack Soble of The Loop Sports did a great film breakdown of Goldman’s impact on the run game.
  • Roy Robertson-Harris had some flash plays this year, but the data suggests he didn’t really have a positive impact on the defense as a whole. That’s plenty understandable in the run game, because he’s not really a typical 3-4 defensive lineman and doesn’t 2-gap as well as the rest of these players.
  • Sometimes the differences between yards/play and success rate can be confusing. Let’s look at Jonathan Bullard as an example. In the run game, teams averaged a lower yards/carry when he was on the field, which is good, but had a higher success rate, which is bad. That tells us that the Bears gave up fewer long runs, but still let teams stay with or ahead of the chains a bit more when he played. That could mean he played in a lot of short-yardage situations, where a 1-2 yard run is a success but keeps the average gain low. On the flip side, teams averaged more yards/pass when he was on the field, but had a lower success rate. That means the Bears didn’t give up completions as much, but gave up more big plays.

 

Linebackers

Let’s move to the linebackers now, looking at the same stats in the same format as with the defensive line. I’ll note that players with less than 100 snaps were omitted, so Danny Trevathan is not included here. There’s no point comparing on/off field splits when the off sample size is so small.


LB-splits.png?resize=750%2C238


A few thoughts:

  • Hello Roquan Smith. I am very excited to see him quarterback the defense for the next decade.
  • It’s a small sample size, so take this with a grain of salt, but Isaiah Irving appears to not be great against the run. This makes sense for a smaller player who relies mostly on speed, not strength.
  • Otherwise we really don’t see much fluctuation here. That speaks to the quality of the defense as a whole; they kept up a high level of play even as guys rotated in and out and some of the best players like Khalil Mack weren’t on the field.

Secondary

Finally, let’s turn to the secondary. Same setup as before, and I’ll note Kyle Fuller and Adrian Amos weren’t included because they didn’t miss enough snaps.


db-splits.png?resize=750%2C224


A few thoughts:

  • The red for Eddie Jackson really illustrates what can be the limits of this approach. He basically didn’t miss any snaps until his ankle injury, so this isn’t rotation data, but rather highlighting the games he missed. You can see that his stats are basically inverse of Deon Bush’s, as Bush filled in for Jackson for the last 3 games (including playoffs). This to me doesn’t say that Eddie Jackson was bad in 2018 (because he wasn’t), but more that the defense as a whole was still really good in the games that Jackson missed.
  • We see some of the same with Bryce Callahan and Sherrick McManis, though unlike Jackson, Callahan was off the field a decent amount even when healthy. In general, McManis seemed to struggle in run defense a bit, but he was very good in pass coverage last year. The job that both McManis and Bush did filling in during 2018 should make the Bears feel pretty good about their depth in the secondary.
  • Kevin Toliver is another player who saw temporary starter snaps in relief of an injured player, this time Prince Amukamara. The sample size is pretty small there, but generally he held up ok in pass coverage (though gave up a few big plays, more on that in the next article) and seemed to struggle a bit against the run.
  • The really high passing average when Amukamara is out is basically the Toliver split, as most of Toliver’s snaps came in the 168 that Amukamara missed.
  • Quick stats on the new Bears: The Jets’ defense was worse against both the run and pass with Buster Skrine in the game in 2018. This difference was more pronounced in the run game (4.0 vs. 5.0 yards/carry, 38% vs. 51% success) than passing (7.5 vs. 7.9 yards/attempt and 47% vs. 50% success). The run defense might be largely because Skrine played almost exclusively at nickel back, and teams by and large stop the run better in base packages than nickel looks.
  • The other new defensive addition is safety HaHa Clinton-Dix. Washington’s defense was worse against the run in yards/carry (4.8 vs. 3.9) but better in success rate (49% vs. 51% success) with him on the field in 2018, and the pass defense was worse when he played (8.5 vs. 7.6 yards/attempt, 57% vs. 46% success). Green Bay’s defense, meanwhile, was worse against the run (4.4 vs. 4.2 yards/carry and 48% vs. 45% success) and similar through the air (8.2 vs. 7.8 yards/attempt, 48% vs. 55% success) when he played; Clinton-Dix was traded from Green Bay to Washington mid-season and played more than 400 snaps for each team.
  • In both cases, it looks like the new Bears’ defenders made their teams’ defense worse in 2018, but it’s important to note that the rest of the defense around them matters a lot. Chicago will be a new situation with new roles and responsibilities for each player, and we don’t know exactly how their performance will translate from a different situation. Still, it’s reasonable to have some questions about how effective Skrine and Clinton-Dix will be based on this data.
Edited by soulman

Share this post


Link to post
Share on other sites

Just a quick observance.

Although we see RRH's impact as a pass rusher at times his overall impact indicates something else.  He shows some weakness as a run defender.

Given the emergence of Nichols is could make sense to have him drop 15-20lbs and convert him back into and Edge/OLB as a 2020 replacement for Lynch.

Just thinkin' out loud.

Share this post


Link to post
Share on other sites

Interesting analysis.  Roquan Smith is a freaking animal!  I can't wait to see what he looks like with a full off-season under his belt.

Share this post


Link to post
Share on other sites

Why was Trevathan not done, it mentioned leaving players off with less than 100 snaps and he had almost 1000. 

Share this post


Link to post
Share on other sites
1 hour ago, blkwdw13 said:

Why was Trevathan not done, it mentioned leaving players off with less than 100 snaps and he had almost 1000. 

Less than 100 snaps missed. Same reason for no Fuller or Amos. 

Share this post


Link to post
Share on other sites
11 hours ago, AZBearsFan said:

Less than 100 snaps missed. Same reason for no Fuller or Amos. 

Seems dumb to leave off the biggest sample sizes. 

Share this post


Link to post
Share on other sites
39 minutes ago, blkwdw13 said:

Seems dumb to leave off the biggest sample sizes. 

The sample size of snaps NOT taken is too small.  The stats compare defensive performance on the field vs off, if the player is never off the field theres nothing to compare.  It's trying to be like +/- in basketball and hockey.

Share this post


Link to post
Share on other sites

Not sure how valuable this data is as a whole given Eddie Jackson is all red and he was arguably our most valuable overall defender last year and Bilal Nichols was clearly better than Bullard to naked eye.

Sometimes stats are lies and sometimes they are damn lies.

 

 

Share this post


Link to post
Share on other sites
On 7/1/2019 at 11:30 AM, soulman said:

The red for Eddie Jackson really illustrates what can be the limits of this approach. He basically didn’t miss any snaps until his ankle injury, so this isn’t rotation data, but rather highlighting the games he missed. You can see that his stats are basically inverse of Deon Bush’s, as Bush filled in for Jackson for the last 3 games (including playoffs). This to me doesn’t say that Eddie Jackson was bad in 2018 (because he wasn’t), but more that the defense as a whole was still really good in the games that Jackson missed.

He kind of covered that here.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  



×