2017 By The Numbers: Chiefs OLBs in Coverage

Sports fans don’t fully agree on anything. Regardless of the team you support or the stance that someone is taking, there will always be dissenters to that opinion.

As a Kansas City Chiefs fan, it’s especially difficult to find something close to a universal agreement. Well, outside of the fact that people want to see the Chiefs’ 3-4 outside linebackers in coverage much less than they have in Bob Sutton’s tenure with the Kansas City Chiefs.

We all know the reasons why we want those OLBs rushing the passer more than they currently are. At times, the pass rush seems stagnant and the quarterback has all day to pick apart the defense and find an open man. We also know that the OLB’s in today’s game have some coverage responsibilities and understand the need for them to drop into coverage at times. But most agree that the pass rusher the Chiefs are paying a lot of money for should be, y’know, rushing the passer.

And, well, yeah. He should.

This post isn’t here to try to condone any of the coverage calls that Justin Houston, or any of the Chiefs OLBs, were put into by Sutton. It’s also not here to tell you who did what correctly or try to dictate how you should feel about the 2017 Chiefs defense.

I’m going to spend this post and a few more over the next couple weeks delving into the where, how and a little bit of the why behind some of the actions of the 2017 Chiefs defense. The goal of all of this is to provide information regarding several oft-debated topics that goes beyond what’s typically provided to the fans.

Maybe along the way, we can defeat some perpetuated narratives that aren’t true anymore. And just maybe we end up taking the overall level of discussion and debate to a higher knowledge level.

Cool? Let’s get into it.

---

I have spent a good bit of my life for the past couple weeks watching every passing snap that the 2017 Chiefs had to defend. Some of it was not pretty. Some of it was downright frustrating to relive. And most of all, I’m sick of watching Derek Carr throw passes at the Chiefs defense (111 passing snaps and you couldn’t be bothered to throw with a little more variety?).

While watching these snaps, I charted coverage drops for both OLB positions. While charting, I kept track of down, distance, point differential, quarter, offensive personnel, offensive formations/alignments, defensive formations, coverage responsibilities, number of defenders rushing the passer, if a blitz was shown and whether or not a blitz was executed for every snap. What I ended up with was a large block of data that shows some tendencies that Sutton had in the 2017 season.

Before we really start hitting the numbers, I wanted to know a little about the perception of last year’s defense. I asked on Twitter what everyone’s three “best” performances for last year’s defense were. I didn’t want people to cite stats, I simply wanted a gut feeling from fans on what they believe were the best performances. By and large, most of the votes were cast for the Week 1 New England Patriots game, the Week 2 Philadelphia Eagles game, the Week 12 Buffalo Bills game, the Week 14 Oakland Raiders game, and the Week 15 Los Angeles Chargers game. I’ll keep referencing these five games as the “good” performances over the next few weeks as we get into other aspects of the defense.

First and foremost, let’s hit the stuff that most clicked over here for: the game-by-game and overall totals.

The numbers that are going to jump out at everyone are the totals of 25.67 percent for the LOLB (typically Justin Houston’s spot) and the overall snaps with an OLB in coverage of 34.70 percent.

Yes, that’s correct. Sutton put at least one of his OLBs in coverage over one-third of the time. You can let that settle in, take a break and come back when you’ve shaken that off. This post will be right here when you get back.

Now that we’re over some of the shock value of these numbers, let’s take a look at the individual games themselves. If we look at the “good” defensive games from 2017, we see that three of the five “good” games exceed the season average for coverage drops by OLBs in them.

The New England game, specifically, has over 50 percent of the snaps in the game with one of the OLBs in coverage. Only the Week 14 Oakland game is very much under the season average out of those “good” games, at 24.49 percent of the snaps with an OLB in coverage. Meanwhile, Weeks 7-9 (Oakland, Denver, and Dallas) and Week 13 (New York Jets) are the lowest four games in OLB coverage snaps. Those games weren’t even really considered in the “good” defensive games category.

You might notice that I’ve dedicated one column on the table to “Average Point Differential.”

That column shows a general temperature of the game on defense for Sutton as he’s making his calls. If the Chiefs are up big and have multiple coverage snaps while they’re up big, that number will increase. As you can see, there are several games that were pretty comfortable wins for the Chiefs with high Average Point Differential numbers for those coverage snaps. Those tended to be games where the defense was in sub packages more often against teams throwing to catch up...

...which brings us to our next set of data:

As you can see, as the Chiefs gained a bigger lead, they had a tendency to drop their OLBs in coverage less often. When losing, they tended to play closer to how they did with a slight lead. The value that jumps out the most to me is that of the LOLB when the game is tied, which goes hand in hand with the splits by quarter, shown below.

Early in the game when the score is tied, Sutton likes to drop his OLBs in coverage. The first quarter percentages line up very closely with those of a tie ball game. As the game progresses and the score potential shifts, the amount that the OLBs drop into coverage decreases, especially inside the two-minute warning at the end of the second and fourth quarters, typically when the Chiefs are in a sub package against a hurry-up offense.

Which brings us to the most interesting aspect of 2017’s OLB Coverage splits for me (and also the final table of the post, I swear), and that’s looking at it by formation.

A whopping 80.31 percent of the “base” 3-4 defense dropped an OLB when facing the pass. That is a STAGGERING amount of coverage snaps. By comparison, the Chiefs actual “base” 2-3-6 defense (2 DL, 2 OLB’s, 1 ILB, 3 CB’s, 3 S) saw less OLB coverage snaps, and that formation saw nearly 3 times the overall coverage snaps in 2017. Basically, if the Chiefs were in a 3-4 defense and the offense was going to pass, you’d have really good odds that they weren’t going to rush both of the EDGE rushers from that defense.

So why is that coverage number so high? In order to do that, we need to look at how Sutton substitutes against various personnel. The 2-3-6 defense almost exclusively (96.8 percent of the time) took the field against 11-personnel on coverage snaps, which is one running back, one tight end and three wide receivers.

There were times when Sutton would trot out a 3-3-5 or 2-4-5 defense, taking off a linebacker or a defensive lineman for a safety or a cornerback against 11 personnel, but for the most part, it was leaning heavily on the 2-3-6. Conversely, Sutton would make wholesale changes any time the third wide receiver came off the field for an additional tight end or an additional running back. This was especially crucial against 12-personnel (one running back, two tight ends, two wide receivers), which saw an OLB drop into coverage 85 percent of the time.

Easily the most glaring coverage drops for me were these 12-personnel settings, so let’s look closer at that. The opposing offense was able to force the Chiefs into their base defense regularly during the 2017 season due to run deficiencies, which meant that they could manipulate their offensive formations to force the Chiefs OLBs into coverage.

The GIF starts with the Buffalo Bills motioning a tight end out wide. Houston recognizes the TE’s motion and switches with Marcus Peters, so there’s a better matchup on TE Charles Clay in the slot. On the snap, Houston is in man coverage with no help on a vertical route against the TE, and frankly does pretty well. It would require quite the throw to beat him.

Buffalo specifically targeted Houston with this play, looking to drag him out of the pass rush, either with Clay (who likely would give him trouble) or by forcing him out wide to cover the motion TE. If the TE had stayed on the opposite side of the formation (where he started), OLB Frank Zombo would have been tasked with the coverage responsibilities on that TE.

So why doesn’t Sutton switch things up and have an inside linebacker or the safety come into the box and help cover the TE in this scenario so that Houston can rush? First off, Sutton does NOT like to run Cover 0 (no deep safeties), and with the cornerbacks the Chiefs had last year, I can’t say that I blame him.

Second of all, as good as Houston is in coverage, he’s not as comfortable dropping and becoming the hook (short middle zone) as the two ILBs. Those ILBs also need to be alert to the running back out of the backfield and the run threat of quarterback Tyrod Taylor, which is a situation that the Chiefs don’t necessarily want to force their OLBs into in this scenario.

This play is a very similar look to the previous play, except now Houston is stuck on Vernon Davis, a better athlete at TE who runs right by him for a big gain on the vertical route. Ron Parker has to respect the verticals on the opposite side of the field and can’t cheat to help Houston, resulting in a big gain on a forced matchup through 12-personnel.

What we learned

Sutton found himself last year in a predicament where he needed to have the 3-4 defense on the field to feel comfortable stopping the run. When offenses brought run-heavy personnel on the field, he made wholesale changes to the coverage responsibilities, tasking the OLBs with more than their fair share. He tried multiple times to keep the 2-4-5 defense on the field, but as you’ll see in an upcoming post, he simply wasn’t able to get results with it against the run.

Meanwhile, the 2-4-5 really is a comfortable spot for Sutton, if he has the personnel to run it. As you can see above, he dropped the OLBs in coverage in the 2-4-5 the least of any significantly-run formation in 2017. This is because it can bring a third safety onto the field if the offense switches to 12-personnel, or it can bring a nickel CB on the field if the offense brings out 11-personnel, all while keeping your front six run defenders on the field and not asking as much from your OLBs in coverage.

As Reggie Ragland came into the lineup more and more toward the end of the year, Sutton really did try to flex into that formation more, but the pieces just didn’t quite fit to make it work in the run game.

So that’s...a LOT of data about OLBs in coverage. I’ve got a bunch more, and if there are specific situations you’re curious about (e.g. “How often did they drop into coverage on 3rd and 10 in the 4th quarter when they were protecting a 3 point lead?”), do NOT hesitate to ask. The hope is that I can offer some clarity on some of the questions that the community might have that they may not KNOW they have yet.

I’ll see you next week to talk about 2017’s blitz and pass rush data!

In the meantime, I’m going to start a petition to delete all Derek Carr data from the NFL’s website.