- Joined
- Apr 7, 2013
- Messages
- 120,031
By Bob Sturm 1h ago
The end of the Jason Garrett era has been handled differently by different people. If you follow this particular writer, you know from our somewhat controversial “Jason Garrett Week” earlier this summer that I was pretty sure a change was due by the Spring of 2013.
Here, my opinion sort of matters. So I would tell you that most of my general feeling is this: He had plenty of positives, but the vanilla predictability on both sides of the ball seemed to transcend his coordinators. Is it truly fair? I’m not sure. But Scott Linehan and even Kellen Moore began to run together, while Rod Marinelli and Kris Richard both were as predictable as the sunrise. I wish we could believe that he allowed them all to do what they wanted and the way they believed it should be, but by the time we got to Richard, it felt he was being brought in specifically because he would call almost no blitzing or coverage adjustments to cross up an offense.
Now, being good at what you do in a simplistic approach is a great trait at certain levels of football. Some defenses are so loaded they don’t have to get cute. Seattle in its prime and even San Francisco last year were dominant enough that they did not require ambushes to win. They declared their plan and you still had a hard time beating them. It is great to have great players to make every scheme look great.
But most offenses are very good. Generally, that’s because their QBs are very good. They are information supercomputers. For that reason, predictable coverages provide a quick path to death in many cases. Maybe not rookie QBs, but definitely veterans. If you show Tom Brady, Drew Brees and Aaron Rodgers (Dallas played all three last year) the same coverages or defense and the same rushers in the spots on the field and the “down and distance” situations, you will be sorry.
And, man, the Cowboys got vanilla in a big way in the last few years, and I think one objective moving forward is to at least try to be multiple on defense.
So as we did with the offense, let’s examine where we left off on defense.
Understanding defensive coverages is not easy. I have spent years on that topic but never feel comfortable about my findings until I have watched a play several times from the All-22 angle. Even then, conclusions are an educated guess based on the team’s overall trends. But they are doing things meant to fool a quarterback and an opponent. Sometimes we see combo coverages where half the field is doing one thing and the other half is doing another. The defense is not meant to be easily deciphered. It is supposed to be complex; defenses want to confuse their opponents. They want an opposing quarterback’s head to spin. Also, they are human, so sometimes they’re also trying to figure out what a player is doing while he himself is making a mistake. You don’t know that by watching the film unless it is followed by teammates or coaches gesturing to him that he is supposed to be in one spot as opposed to another.
None of this, however, stops me from trying my best. I spent hours and hours looking at each pass play the Cowboys defense faced and recording each coverage — hundreds and hundreds every year. I have a few coaching/scouting friends to whom I send film, and they help me decide on the tricky coverages. Trust me, I am positive some of them are wrong, as this is my “best guess.” With their help here is what I found for this season:
2019 DALLAS COVERAGES
Due to the length of this study, I am going to simplify it quite a bit. As anyone who has ever played Madden — let alone high school and college football players — knows there are often up to a dozen variations of ways to run each coverage. Sure, a team may run Cover-1. But is it Cover-1-Robber/Bandit, Cover-1-Rabbit, Cover-1 with a blitz strong or Cover-1-Funnel?
We are going to place all the variations of Cover-1 under that same umbrella. It’s useful for us to examine which of the four or five base coverages a team will run. We can then extrapolate that to what the Cowboys like.
The Cowboys run two coverages most frequently: Cover-1 (man coverage with a single-high safety in center field) and Cover-3 (zone coverage with the corners dropping so that there are three players splitting the deep part of the secondary into thirds when they drop near the deep safety’s depth). They certainly run at least a few types of Cover 3, but again, we will put them together.
Both have one safety deep. But in an effort to confuse you (and me), the safeties can often start in a two-deep shell. This means that at the snap, they want the QB to see two deep safeties, which he will use to compute what he is facing. But then they quickly adjust into a single-high look. Again, everything a defense does is intended to fool you, so they don’t want this to be easily understood.
It is one thing to use these overall numbers, but, of course, they are useless unless you know when the Cowboys use each of them most. And now, the 2019 numbers:
2019 Dallas Defensive Coverages
| DOWN | COVER 1 | COVER 2 | COVER 3 | |
|---|---|---|---|---|
| 1st | 20% | 13% | 67% | |
| 2nd | 22% | 16% | 62% | |
| 3rd | 58% | 7% | 34% | |
| Overall | 29% | 13% | 54% | |
Here, you see how the coverages break down by down and distance. On first down, the Cowboys are in Cover-3 about two-thirds of the time. Cover-1 makes up 20 percent and Cover-2 clocks in around 13 percent, but we will explain that below. On both first and second downs, the idea is to get eight men in the box, and the best way is Cover 3. Force the opponent to rely on their passing game (which is amusing, given how much the NFL loves to pass now) and then play man coverage later on third down.
In fact, the Cowboys almost never play Cover-3 on third down when the game is close and less than 35 percent of the time even if you include games that aren’t close. They either get aggressive with Cover-1 or conservative with Cover-2.
Cover-3 is their base coverage of choice (54 percent of all snaps), but only in normal situations. If it is third down, they prefer to man up and even occasionally marry it with some complex pressure packages and looks in order to turn up the pressure.
Let’s run through the three basic coverages with the diagram for the basic looks from Bill Belichick’s 2001 Patriots playbook. I will concede they run a few more than three coverages, but they also seem to land on the very “vanilla” side of the ledger. There are some very high-risk and aggressive defenses out there, but Dallas hasn’t been one since Wade Phillips or maybe Rob Ryan.
Before we define each of them, here are the 2018 numbers and where the Cowboys were the last time we did this:
2018 Dallas Defensive Coverages
| DOWN | COVER 1 | COVER 2 | COVER 3 |
|---|---|---|---|
| 1st | 22% | 19% | 59% |
| 2nd | 25% | 21% | 54% |
| 3rd | 54% | 25% | 20% |
| Overall | 33% | 22% | 46% |
COVER-1
FREQUENCY: 21st in the NFL — 29 percent of all plays, 58 percent of all third downs.
DEFINITION: A deep safety in center field, everyone else in man coverage across the board. Often, there is still another free defender who can be used as a blitzer, a robber or a rat. The robber would generally be a safety sneaking down into the path of crossers and a rat would generally be a linebacker playing a shallow version of center field, looking to get his hands on the ball by reading the quarterback’s eyes.
WHEN THEY USE IT: Any time, any place, any down and distance — but especially when they face a big single down with high stakes. This is not “bend but don’t break.” This is “make a play.”
HOW WELL THEY USE IT: Generally, the Cowboys have really been good at this. It obviously causes issues as elite offenses pick matchups where their third- or fourth-best target can attack your third- or fourth-best coverage guy, but the Cowboys can execute this coverage well, and that is why they don’t mind blitzing as much as they used to.
COVER-2
FREQUENCY: 10th in the NFL, but just 13 percent of all snaps
DEFINITION: Two safeties deep, five shallow defenders across the board with variations where different shallow defenders drop deep to match up with different attacks from the offense. Tampa-2 is one variation we used to see much more often, where the Mike linebacker (usually Jaylon Smith) will split the safeties in deep center if someone heads down the seam. It seems to have almost disappeared in 2019. As you can tell, the Cowboys hardly ran it, but also ran it 10th-most in the league, telling you this is a dying coverage (for now).
WHEN THEY USE IT: This defense is not often used unless the Cowboys are trying to play conservatively with a lead or on third and long or just trying to run out the clock. It is their most careful bend-but-don’t-break look. They also roll it out quite a bit in the red zone, where the field is compressed.
HOW WELL THEY USE IT: Hit or miss. It seems like, overall, they preferred to use Cover 3 in 2019 in its place, so this could be a safety-dependent component, and I just think they wanted to be single-high.
COVER-3
FREQUENCY: Second-most in the NFL, 54 percent of all snaps.
DEFINITION: Corners drop deep and play the outside thirds of the field, funneling the receivers inside to the deep safety by playing outside leverage on them. There are four shallow defenders in front of the three deep zones. This can often start out of a 2-shell (a two-deep look in which the strong side safety will rotate into the strong-side flat). The complete premise behind this defense is to stop the run with the eighth man in the box and still play a reasonably conservative secondary behind it.
WHEN THEY USE IT: The Cowboys use this as their standard coverage on first and second downs. In fact, the defense is in Cover-3 65 percent of the time on first and second downs. Only the Chargers ran this coverage more in 2019.
HOW WELL THEY USE IT: I believe this has become the NFL’s prevailing base coverage, so there are many ways to attack it that a team must be prepared to see. That said, the Cowboys were a very difficult team to attack through the air in 2018, and although more susceptible in 2019, their usage of Cover-3 has been a big part of their work.
The other thing we can track in deployment is blitzing. You should know that Cover 3 is not the same blitzing coverage as Cover 1. When you are in man coverage, you have more available blitzers. When you are playing zone, every man you rush means less of the field is covered, right?
Let’s check that theory:
As you can see, the Cowboys did not blitz much, but they did blitz on third down at least about league average, and over it on third-and-long. This sample is much smaller, but overall, you could see they wanted more of that.
Earlier this spring, I spent time writing about the new coaches and what new DC Mike Nolan and new head coach Mike McCarthy could mean to building a defense that properly compliments the offense:
The Cowboys decided they needed a change. I do not believe they seriously considered Kris Richard or Rod Marinelli for future positions here. They tried their ideas repeatedly and found that no takeaways were a constant, and this constant was tilting the field against them. When that happens, you are asking more of your offense — not less — by starting every field position deeper than your opponents every week. And, no matter how careful your offense is with the football, they still lose the turnover battle because the defense can’t get it. They were determined to change in 2020 long before they hired the new help.
I want to be really clear when we discuss blitz rates and other aggressive defensive tactics. There are bad teams that blitz all of the time. There are good teams that never blitz. Never ever confuse attacking as shorthand for dominance. Blitzing can often be an act of desperation. I constantly repeat a coaching quote I learned years ago: “We blitz as a weapon, not as a necessity.” If you have to blitz, you are in trouble. But if you believe the best way to defeat an opponent is to terrify them, then it can really be something special.
You already know some of this intuitively if you closely follow the Cowboys, but since the Rob Ryan era (2011-2012), Dallas’ defensive philosophical shift has been based on two basic premises: They don’t blitz, and they very seldom play man coverage. I am speaking in absolutes, and when we describe seven years and 7,500 plays, I am only speaking with regards to the majority of the time. But that is basically what they have done, mostly under Rod Marinelli throughout.
To demonstrate, I made you a fun little visual aid here with four different lines. The red line is the league-average blitz rate, which is usually around 30 percent or so over the last decade (but it obviously fluctuates).
Now, look at the blue line below, and you will find the Cowboys’ annual blitz rate over the last decade. You will quickly see that it almost is never above the red line. When it was, Rob Ryan was in charge, and the results were often bad.
McCarthy, in particular, is never even close to the red line. His teams always brought pressure and also were always well above Dallas in takeaways. Interestingly enough, Green Bay this year dropped their blitz rate enormously without him (with the same defensive coordinator, Mike Pettine), and they experienced their best defensive season in years. Of course, they also added a boatload of defensive talent since McCarthy was fired, so I don’t know how relevant that shift in play-calling truly is.
Soon, we will know if I am correct. I think a high-scoring offense should be paired with a high-aggression defense to play “complementary football,” and I assume the Cowboys feel the same way. Without real exposure to training camp or preseason games, we will need to see in September.
But I do think we will see a real difference in 2020.
