Heading into their Week 11 bye, the Cowboys’ coaching staff promised intense self-scouting—an analytical approach to discovering flaws in their approach to the game. It seems like they finally identified an inefficiency I’ve been urging them to locate for, well, years—a play-action pass deficiency.
During the bye week, I wrote this in my article on four ways to improve the offense:
Tip No. 3: Use play-action way more often
Want jaw-dropping evidence that the Cowboys don’t embrace analytics and are unwilling to adapt to new information? Last year, Romo ranked last in the NFL in play-action percentage, attempting a play-action pass on just 10.0 percent of his dropbacks despite totaling a 109.1 passer rating on those passes.
In 2013, Romo ranks last in the NFL in play-action percentage, attempting a play-action pass on just 10.3 percent of his dropbacks despite totaling a 121.2 passer rating on those passes.
What the hell?
All kinds of success on play-action, yet the rate has increased 0.3 percent points? With that sort of improvement, we’ll only need to wait just over 37 years until the Cowboys reach THE LEAGUE AVERAGE in play-action percentage.
Oh, but the Cowboys can’t run the ball, you say, so why use play-action? First, Romo’s ridiculous play-action success is reason enough to increase the rate. But more important, play-action efficiency isn’t correlated with rushing success.
Defenders play situations, not past rushing efficiency, so the Cowboys don’t need a strong running game for play-action to work. If they implemented more of a scientific approach to decision-making over the faith-based approach they currently utilize, they’d probably know that.
Better late than never.
Coming into Week 12, quarterback Tony Romo had been averaging 4.1 play-action dropbacks per game. Against the Giants, Romo showed play-action 15 times! He was actually sacked on three of those dropbacks, which is more of a fluke than a weakness of the play type.
Of the 12 play-action passes he got off, Romo completed eight for 111 yards, a touchdown and no picks. So if you’re counting at home, that’s 9.25 YPA on play-action passes and 5.35 YPA on straight dropbacks. Again, kudos to the coaches for (hopefully) recognizing an offensive inefficiency and correcting it, but you have to think a team with a greater emphasis on analytics wouldn’t have waited more than half of the season to do it.
Breaking Down the Play-Action
The most important sentence from the above excerpt is “Defenders play situations, not past rushing efficiency.” The Cowboys have historically run less play-action than anyone in the NFL because they haven’t been able to run the ball. They believed something to be true without researching it—that you need to run to set up play-action—then used that false “knowledge” as a foundation for building their offense.
When they took the time to study what was going on—and it’s really mind-blowing that it would take them 11 weeks to have someone do that—they realized, hey, maybe we could more frequently utilize this aspect of our play-calling that’s been extremely successful in the past.
To give you an example of how defenders play situations, let’s take a look at tight end Jason Witten’s first touchdown. On a 1st-and-10 at the Giants’ 20-yard line, the Cowboys lined up in “Gun Tight End Trips Left.”
Witten was lined up in-line to the field, and the Giants showed a Cover 2 look. The reason that Witten has been successful against New York this year is that the middle of the field—up the seam, in particular—is open with the safeties split out wide. That leaves the linebackers to trail Witten, which is made more difficult when Dallas shows run-action.
You can see both linebackers bit up on the Cowboys’ run-fake. Two yards into his route, Witten was already behind them. That wouldn’t happen without the play-action.
By the time the ball was in the air, the linebackers—both of whom were in coverage and tried to back up into their zones after they realized the play was a pass—were far out of position. This was a great read from Romo, who needed to deliver the football quickly to fit it into his tight end before the safeties converged on him.
On the Cowboys’ final drive, the ‘Boys ran another play-action pass to wide receiver Dez Bryant. On a 2nd-and-10 at the Giants’ 28-yard line. It was an awesome and unusually aggressive call from Dallas since they were already in field goal range—a sign that perhaps Jason Garrett is understanding that “field goal range” isn’t some magical land in which you can’t possibly miss a field goal.
It was smart because the Giants, who knew Dallas was in field goal range, were susceptible to run-action. On this play, the safety who was sneaking up prior to the snap continued a few steps toward the line when he saw the run-fake to Murray.
That opened up a throwing lane to Bryant—again, one that wouldn’t have been available without the play-action.
The pass was ultimately ruled incomplete—a call that should have automatically resulted in a review from the booth since, in my opinion, Bryant actually caught the ball. The ruling stood, but that doesn’t change the fact that it was an intelligent call for Dallas to make at the time.
It took too long for the Cowboys to recognize that they should be calling play-action, and there’s always a chance that this was just an outlier and they won’t really utilize it more moving forward. But if they did indeed notice that more play-action passes will greatly enhance offensive efficiency, we need to give them credit for taking the bye week to improve their decision-making.