Good news! The Cowboys’ defense—full of young, homegrown talent—single-handedly won them a game on Sunday. They forced turnovers and three-and-outs, scored points, and knocked out the opposing quarterback. They are a violent, bullying force with playmakers at all three levels of the unit, and they appear to just be getting started.
Bad news! The team would be in serious trouble if the defense hadn’t coalesced into this gameplan-wrecking death squad. Because, as you know, their previously unstoppable offense has been stopped. The side of the ball that we thought had all of the answers suddenly appears to have none.
Full disclaimer: there is no magic bullet here. No easy answer or explanation. Everything that ails this offense is not going to get better at the same time, in the event it gets better at all. But to properly analyze this funk, we must attempt to see what the biggest issues are and how they can be solved.
Ezekiel Elliott was stabbed by a pylon with a camera in it back in Week 5 against the Giants. Dak Prescott was injured on the final play of the game the following week at New England. Then came the bye week and The Cooper Rush Game as Prescott sat with an injury. Dak then returned to face the Denver Broncos and Vic Fangio, who claimed he had found the blueprint to slowing down the until-then prolific Cowboys offense. For those reasons, it makes sense to look at the Cowboys offensive season in two segments: Weeks 1-6 and Weeks 8-14 (understanding that Dak did not play in Week 8 against Minnesota).
Here are their ranks by Expected Points Added (EPA) per play.
Like I said, stark. Elite to way, way below average. League-wide, offensive production tends to hit a bit of a wall as the season progresses. In fact, all but 11 teams had a negative change in EPA/play from the first part of our sample to the second. But that doesn’t account for the change in Dallas’ relative rank. No team has had as much of a drop-off as Dallas did—not even close. The next closest team in EPA/play decrease from Weeks 1-6 to Weeks 8-14 is the Rams. Their rank went from 4th to 13th.
We are, of course, using a cutoff point here that is very specific to Dallas, and arbitrary for the rest of the league. It is, however, roughly the two halves of the season to this point. And no other offense has been as good and then subsequently as bad as the Cowboys’.
So, what’s the deal?
Again, no simple answers. But let’s start with something mentioned during the broadcast Sunday, which we have hit on before: how teams are defending Dallas, specifically the opposition’s blitz rate. Weeks 1-6, Dallas was the most blitzed team in the NFL (different services produce slightly different numbers when it comes to tracking data. I use Pro Football Focus, but in general, the differences are negligible). This made absolutely no sense because Prescott and the Dallas offense were among the best at making teams pay for their gambles.
Then suddenly, things changed. Beginning with their Week 9 matchup with Denver, Dallas has been blitzed at the second-lowest rate in the league. That sort of juxtaposition cannot be ignored.
What’s even more concerning, though, is in the few occasions where teams have elected to blitz, Dallas and Dak’s success rate has also fallen off of a cliff. So while I do think defenses bringing fewer rushers and playing more coverage has been a significant factor in slowing down the Cowboys’ offense, that clearly isn’t the entire story.
It’s pretty clear to anyone watching that Dak just doesn’t look right. Our expectations for him may be lofty, but they are not unreasonable. We are hoping for a return to the form that existed two months ago, not two years ago. I have no idea how much last season’s ankle injury or this season’s calf injury are affecting his ability to live up to that standard. I’m not going to pretend to be capable of breaking down his mechanics and exhibit how these injuries would contribute to passes sailing on him. What I do know is this: for whatever reason, Dak Prescott is now practically a nonfactor as a run threat. If we exclude scrambles, QB sneaks, and kneel-downs, Prescott has a total of four rushing attempts for the season. All but one came before the calf injury. Of course, Prescott’s game was never similar to that of Josh Allen or Lamar Jackson. But averaging one called QB run per three games? That is more in line with Matt Ryan or Sam Darnold this season. I know that is painful to read; it was painful to write, too.
This could be an organizational decision that was made before the season and before the calf injury to protect their most important investment. That’s defensible, but if he doesn’t keep it once or twice a game, the threat is no longer credible, and the ripple effects on the offense are severe (particularly the run game, which we’ll get to).
Again, I don’t want to engage in armchair QB mechanics. But another area where Prescott has just looked off are on designed rollouts. This has been a tried-and-true, bread-and-butter staple for this offense. You know the play. Play-action, rollout, TE in the shallow flat, deeper receiver running the over from the backside. High-low coverage stresser; easy yards. Dallas is still calling them at basically the same rate across the season. But from Week 1-6, they were the fourth-most efficient team in the NFL on designed rollouts. From Week 9-14, they are 29th. The easiest, most reliable call in this playbook is no longer being executed at an average level, or even particularly close. It could be that there are more defenders in coverage, meaning the defense doesn’t have to pick its poison. It could be that Dak is not as comfortable throwing on the run and that defenses don’t really believe he’ll take off if they stick with the receivers. It’s probably some of both. Whatever the cause, the results are definitely not good enough.
When we were exalting this offense as virtually unstoppable, it was because they could beat you however you elected to be beaten. If defenses chose to play stacked boxes, the Cowboys could throw it 50 times using short, quick passes to get the ball in the hands of their playmakers. If the opposition decided to lighten things up and limit big plays, they could run the ball effectively all day long. They could grind out first downs on the ground while also producing explosive rushes. Few other offenses were capable of this dynamic versatility.
And then, just when they needed their ground game the most, it vanished. Health is clearly a factor. Elliott crashed into the pylon in Week 5, then suffered a knee injury in Week 9 in the Denver loss. He should not have played in the Saints game. He was in no shape to play against Washington, either, but the team had their hand forced by Tony Pollard’s foot injury.
It is impossible to fully isolate Elliott’s and Pollard’s performance from that of the offensive line or the play-calling. But one way we can attempt to do so is by looking at missed tackles forced, and yards after contact. This is where each player ranks out of the top 60 rushers by attempt for each segment of the season:
Pollard has been among the league’s best at forcing missed tackles and picking up yards after contact all season, regardless of who is playing where on the offensive line. Elliott has never been an elusive back, but early in the season, he was generating enough power to at least consistently pick up a few yards after initial contact. That power is gone. Nevertheless, the distribution of their carries has remained the same. This is a huge problem, and one that now appears much trickier to solve because a) Pollard’s injury, even if he plays, will likely limit his explosiveness and b) Elliott appears to be unbenchable. And Dallas may have missed their window to get Zeke a week or two off in an attempt to save what little he has left for the postseason.
So, what about the offensive line? This is probably the trickiest thing for an outside observer to evaluate. The Cowboys themselves have indicated they have no firm idea of what they want to do up front, from benching Connor Williams for Connor McGovern to starting Terrance Steele over an available La’el Collins and then reversing that decision. The one reliable statistic we can use for the offensive line is pressure rate allowed. First six weeks of the season? Sixth-best. Since then? Seventh-best. If we narrow that down to just the last four games since the Connor Exchange occurred? Still seventh-best.
Pressures allowed is not a perfect metric to assess offensive line play, but it’s a start. I’m hesitant to ever use Pro Football Focus grades, especially for interior play. But, for what it’s worth, the line ranked first in overall run-blocking grade for the first six games of the season. Since then, they’re 13th, and that drops to 17th since switching Williams for McGovern. The drop-off has not been catastrophic, but I think this matches up with the eye test showing some decline in creating space for the runners. It just got tougher to solve, too, as Tyron Smith will miss this week’s game, and likely the next.
Finally, an element that has gone largely underreported due to the other issues: drops. Through the first six weeks of the season, Cowboys pass catchers dropped just four passes, fourth-fewest in the league. Since then, they have 19, second-most in the NFL. Not all drops are created equal, so part of this could just be a function of the quarterback’s struggles with ball placement. CeeDee Lamb and Michael Gallup each have three, but so do Noah Brown and Cedric Wilson, who were thrust into roles a bit above their capabilities when Lamb recently missed a game-and-a-half and Amari Cooper missed two.
My colleague David Moore has described the issues with the Cowboys offense very topically as a “rolling blackout,” and really, that’s the best explanation I’ve heard. The offensive line has been in flux due to a suspension, injuries, and questionable lineup changes. The primary running back has been injured twice and won’t come off of the field, and now the more explosive backup has an injury. The quarterback coming off of an injury got injured, again. Wide receivers have been in and out of the lineup since Week 1. And in the background, opposing defenses have figured out how to pinpoint how these individual blackouts can make the Cowboys uncomfortable, and they’ve done just that.
Dallas can only hang on and hope that Smith, Prescott, and Pollard are close to full health by the time the postseason rolls around. Should that occur, the Cowboys can still be a threat late into January given where the defense is. If they aren’t, Dallas will get beat in short order, and the many looming decisions facing the offensive side of the roster this offseason will seem much more dire.
