Any Given Sunday
The weekend's biggest upset goes under the Football Outsiders lens.

Any Given Sunday: Broncos over Texans

Denver Broncos QB Drew Lock
Photo: USA Today Sports Images

It was tonally kind of weird when I spent half of Any Given Sunday last week talking about the Texans run defense being bad. They had just won a football game! Against the New England Patriots! But it was something that has been an anchor for them since the bye week and it's something that stood out to me. In Week 14, they re-inserted Brennan Scarlett into the starting lineup and did alright at stopping the run -- they made it to -14.6% run defense DVOA, their best game in a month.

But with no pass rush without J.J. Watt or Jadeveon Clowney, the Texans have become extremely reliant on Romeo Crennel's game plans to have a stable pass defense. When it works, as it did against a Patriots team that couldn't deal with two weapons being taken away, it is OK. When it doesn't work, there's almost no floor for the defense. Well, we found that floor on Sunday as they allowed 112.0% passing DVOA to a rookie quarterback in his second start.

It wasn't that the Texans were bad -- it was that they were bad and so, so far away from being near most of these throws that Lock made. This was a surgical pantsing of Crennel's defense by Denver offensive coordinator Rich Scangrello. Denver was generating wide-open plays by manipulating Houston's initial reads. On this one above, the safety has to honor the outside fake. And once he bites, nobody is available to take Noah Fant coming off his block.

The Broncos were able to isolate Johnathan Joseph in the middle of the field on this play, and, by his rules, he had no choice but to stay as far off the receiver as he did because his deep help went with the play fake.

Interestingly, the biggest problem was play-action. The Texans were destroyed when Denver went to play-action despite being relatively solid coming into the game at defending play-action passes.

Play-Action Owning of Houston in Week 14
Weeks Dropbacks Yds/Dropback Avg EPA Comp%
Week 1-13 99 6.2 0.14 71.3%
Week 14 8 16.3 1.03 87.5%

But a lot of the play-action the Texans have played against is designed to suck linebackers in, right? Denver didn't always do that. They moved horizontally, and they understood that certain corners on the Houston defense, such as Joseph, don't deal very well with horizontal movement right now.

Ultimately, the Texans just aren't that good on a cumulative level. Their offense is good, but inconsistent and saddled with many games where it feels like they should be better. Their defense is bad! And I would argue that, despite our sanguine projections for their defense before this season, that it was fairly obvious that they would be bad. They played one of the easiest schedules in the NFL in 2018. This year, they've played much better quarterbacks and head coaches who have a clue, and they've done it with almost no non-J.J. Watt pass rush. With Watt on IR, they've got nothing. They've got so little that Texans fans overwhelm my mentions about Jacob Martin's playing time. They've got a 24.2% defensive DVOA since their bye in Week 10. We have them with just 6.1 estimated wins against eight actual wins, and our numbers have them at a disadvantage for the AFC South despite having a tiebreaker on the Titans if they just win one of their two remaining games with them.

That can fall on only one person: Bill O'Brien. It was his bizarre preseason exile of Jadeveon Clowney that left this defense as short-handed as it is. It was his constant tinkering that has the defense learning to integrate two new cornerbacks on the fly. Even if we wanted to blame somebody else, there's really nobody else to blame. Crennel didn't shop for these groceries.

The good news is that the Texans won't have a first-round pick to use as a remedy to this situation for two years. So get ready for some free-agency fireworks or moves that look even more desperate this offseason!

Where the Game Swung

Once again, let's see what our overlords at EdjSports have to say about where this game was lost by their Game-Winning Chance (GWC) metric. This'll be fast because, well, when it's 31-3 at halftime things are fast.

Denver-Houston

Kareem Jackson's fumble-six completely changed the perception of this game. Not only was it a return touchdown, but it also was a return touchdown that happened as the Texans were driving near the edge of the red zone:

That was worth a game-high 24.6% GWC. O'Brien after the game called it a "weird" play, saying that the back judge had signaled the ball down and saying that Jordan Akins had Jeremiah Attaochu "in the grasp" before it was sent over to Jackson. This is what happens when the NFL doesn't clean up their processes very well. Yes, we know that the Jaguars should have had a touchdown against the Patriots in the AFC Championship Game in 2017. We know that there are several examples of the Saints being screwed by plays blown dead before they should really be blown dead. At the same time, if a crew is torn on whether a play is up or not, well, that leads to a play like this that the Texans thought was dead:

Other plays of note:

  • Lock's first touchdown, a rocket to Noah Fant on third-and-12, was worth 9.2% GWC.
  • Lock's third-and-9 in-the-bucket completion to Tim Patrick to start the second quarter was worth 7.4% GWC.
  • Lock's second touchdown, to make the game 20-0, was worth 5.3% GWC.
  • Houston went for it in their own territory on fourth-and-1, down 28-3 at the two-minute warning. GWC liked the decision, saying it raised Houston's chances of winning by 1.8%. But when they failed, it cost the Texans 4.3% GWC.

By the (D)VOA

DVOA OFF DEF ST TOT
DEN 42.3% -3.9% 2.8% 49.0%
HOU -1.5% 48.8% 4.1% -46.2%
VOA OFF DEF ST TOT
DEN 42.7% 4.4% 2.8% 41.1%
HOU -7.0% 39.7% 4.1% -42.6%

I think I understand where this game was won and lost pretty clearly. I want to point out one extra fact about the Houston offense here, since we didn't talk about them much.

The Texans reached the 50-yard line or further on each of their first four drives, all of which started on their own 25. We divide the field into zones for DVOA, and we call one of those zones "FRONT" -- between the opponent's 39- and 21-yard lines. In the first half, the Texans ran six plays in Denver's FRONT zone and had a -209.2% DVOA.

Drew Lock Second Start Check-Up

One of the hardest things that I think we face as we look at quarterback evaluation going forward is dealing with the fact that a lot of NFL defensive game plans are built solely from NFL games themselves. Advance scouts don't always spend a lot of time on college tape or preseason games. Romeo Crennel, in the lead-up to this game, essentially said they weren't the same level of competition.

Since 2015, 66 first-year quarterbacks have started a first or second game. Many of them were very efficient, even the ones you might not expect to be. Nathan Peterman had a 100.8 quarterback rating in the snow against Indianapolis. Mason Rudolph's first start: 24-of-28, 229 yards, two touchdowns. Daniel Jones: 23-of-36, 336 yards, two touchdowns. Kyle Allen torched the Saints. Gardner Minshew's first two starts were great. Hell, David Blough came out and put 280 yards and two touchdowns on the Bears. Even Paxton Lynch completed 23-of-35 for 223 and a touchdown against the Falcons.

In that sense, there wasn't a lot of Lock tape, and there's a lot of people ready to declare Lock a long-term fixture for the Broncos. I do think he made some excellent throws in this game, and want to specifically point to this one:

This was Lock's first touchdown throw. Lock has an underneath defender to fit the ball over and a trailing safety, which means he has to put some mustard on the ball. He fits it right over the defender's shoulder, and in fact it hits Noah Fant so quickly that he has to recover the ball in his legs on the way down to actually catch the ball.

But as you saw above, a lot of Lock's yardage came on plays for which the Texans were in no way prepared. The Broncos were able to effectively use the information gap -- Houston had no idea what they were going to do -- to cut out to a big lead.

This isn't exactly a bold prediction, but my guess is that Lock will have a poor outing at some point in these next three weeks. There are still holes in his game, and there are still throws he is willing to chance a lot on, including the one big defensive play the Texans had -- Tashaun Gipson's interception.

Gipson read the quarterback's eyes the whole way, because Lock (ahem) locked on to his primary read from the second he turned his head. We can get very deep into quarterback critique woods, and this is not a great process that led to this result. But ultimately I'm not concerned about this in the second week of Lock's career. I'm concerned if he doesn't learn from it.

The learning process for quarterbacks starts as the NFL tape does. It is absolutely a good sign for Drew Lock that he played as well as he did -- nobody is saying it's not -- but these next three games are probably going to be even more instructive to the Broncos as far as whether they want to move forward with the idea that Lock is the quarterback of the future.

Comments

6 comments, Last at 11 Dec 2019, 12:22pm

1 Lock

“ This isn't exactly a bold prediction, but my guess is that Lock will have a poor outing at some point in these next three weeks.”

My guess is it will probably be this weekend in Kansas City. But that’s OK. This was at least the most enjoyable Denver game in the past 2 years. Hopefully Lock keeps getting better.

2 Groceries

"Even if we wanted to blame somebody else, there's really nobody else to blame. Crennel didn't shop for these groceries."

This won't prevent Crennel from getting sacked. Especially since it's O'Brien.

3 Dfew Lock

Drew Lock oaky looking rifjt now
True test will be vs Raiders Week 17.

Ues, if texans do not make playoffs R. Crennel could get canned. No general manager on thst team so maybe if wind blows certain way b. O'Brien will send Crennel out into it and get him blown far away.

Texans face Totans 2x in fisnl 3 weeks. Good chance meetinsg will be split. Each team pmays nfc soutj team between these matcjups. Saints at Tirans. other game is texans at Buccaneers.

4 Broncos twitter is…

Broncos twitter is simultaneously anointing Lock as Manning 2.0, and calling for Scangerello to be fired at the same time. I've unfollowed a lot of people in the last couple of weeks because homer-twitter sucks no matter what the team is.

5 Yes

Yesx, d o not follow many people who liek smae teams as me. Oen time criticized ex-Raider (p7nter who wrnt to Broncos) and said ir was good he waa gone and Raiders fan i followed jumped on me and led to whole big tfight  thing and i got atttacked by numerous other Raiders fans. I haf to unfollow guy and  basicallh rarely ever mention Raiders on twitter anymore

6 The fact is that no amount…

The fact is that no amount of creative play-calling/design can make an offense successful when that offense is quarterbacked by the corpse of Joe Flacco or Brandon Allen. This game featured two coordinators (Crennel, Scangarello) who are likely to be left holding the bag on the personell department's inability to field a roster talented enough to compete. Scangarello may yet be saved by Lock, but the more likely outcome is that Lock turns back into a pumpkin for the remainder of the year as teams get more tape on him. As a Broncos fan, I'm excited about the potential Lock's shown these last 2 weeks. But the list of quarterbacks I can think of who never looked bad during stretches of their first season of starting is Patrick Mahomes and Aaron Rodgers. Rodgers sat for 3 years, and Mahomes only played in week 17 of his rookie year.