No, Ezekiel Elliott Didn’t Help the Cowboys’ Defense This Season
Ezekiel Elliott had a heck of a rookie season.
Selected as the fourth overall pick, he led the NFL in rushing yards, was second among running backs with 200 or more carries in Rushing Net Expected Points (NEP) per attempt, and was first among those backs in Success Rate -- the percentage of plays that positively impact NEP.
He has also awarded for his accomplishments, being named first-team AP All-Pro at running back and more recently being named to the 2016 PFWA All-Rookie team, along with being the overall PFWA Rookie of the Year. He’s even been mentioned as an MVP candidate for much of the season, though it appears unlikely he’ll come away with that award.
Despite all that, the Dallas Cowboys did not win a playoff game this season.
There’s not a direct correlation here, but the billing after the pick in April suggested Elliott was the one that would push Dallas to the next level. In some cases, they had a point, but in others, they missed it completely.
The Cowboys had the fourth overall pick, their first inside the top 10 since back-to-back selections in 2011 and 2012. It was Dallas’ first top-five pick since 2003.
With a healthy Tony Romo at quarterback, Dallas would have likely been nowhere near the top five of this past draft, so this was seen as a rare opportunity for a roster of a playoff contender to add a top talent.
The choice was between Elliott and Florida State cornerback, Jalen Ramsey, a decision that eventually went to the running back. In a pass-driven league, there’s an argument that a running back should not be taken with high pick because the value can be easily made up for in later rounds. The argument was that Elliott was special -- to a degree, that was correct.
Pro-Football-Reference uses a statistic called Approximate Value (AV) as an attempt to put a single number value on each player’s season. By AV, there was no more valuable rookie in 2016 than Elliott.
According to the draft pick value calculator from Chase Stuart at Football Perspective, a typical fourth overall pick can expect 25.8 AV over the first five years of a career. By that measure, Elliott is already 62 percent of the way to fulfilling the value. There’s opportunity cost in play -- the cost of not picking someone else in this spot or another running back later -- but Elliott has done his part so far in living up to the place of his selection.
The Cowboys were 29th in Adjusted NEP per play on offense in 2015, a ranking they were desperately trying to improve in 2016. That ranking, though, was mostly held down by a terrible passing game. Dallas ranked 30th in Adjusted Passing NEP per play with a rotation of Romo, Matt Cassel, Kellen Moore, and Brandon Weeden all getting starts at quarterback.
Meanwhile, they still ranked ninth in Adjusted Rushing NEP per play, mostly behind Darren McFadden.
The drastic improvement they were looking for came this past season, as the Cowboys finished the regular season ranked third in Adjusted NEP per play. Their run game dominated (first in Adjusted Rushing NEP per play), but the biggest improvement came from the passing game (fourth in Adjusted Passing NEP per play). That was engineered by rookie Dak Prescott, who was third among quarterbacks in Passing NEP per drop back.
Elliott's presence certainly helped, but the Cowboys would've looked more like the Buffalo Bills without the improved quarterback play. The Bills ranked second in Adjusted Rushing NEP per play, closely behind Dallas, but just 19th in Adjusted Passing NEP per play.
Overall, the Bills were just 11th in Adjusted NEP per play on offense, which wasn’t enough to pick up an average defense.
Helping the Defense
That brings us to one of the most polarizing arguments about the Elliott pick -- the proclamation that his presence on the ground would help keep the defense off the field. It wasn’t just an added benefit, it was one of the biggest selling points. This is an excerpt from Peter King at The MMQB, who chronicled the happenings in the Cowboys’ draft room (emphasis mine):
“Despite their rushing proficiency in 2015, the Cowboys ran 45 more offensive snaps in 2014 than last season. Elliott’s presence could lead to Dallas dominating in time of possession and offensive plays—both good for an average defense. Around the building, the oft-repeated thought is that Dallas can help its defense most by picking Elliott.”
Instead of selecting the top defensive player in the draft, the Cowboys wanted to help their defense by picking a running back. It was a strange argument at the time, but we now have results to see how this rationale turned out.
For this, we’ll use the drive and pace stats compiled by Jim Armstrong at Football Outsiders. Let’s start with the past two years of offense and where they ranked:
|2015||1,053||165||32.02 (13)||1.56 (27)||3:00 (3)|
|2016||1,082||167||35.63 (6)||2.54 (4)||3:03 (2)|
Dallas ran 29 more plays in 2016 with two more drives. While the yards and points per drive dramatically increased, the time of possession per drive increased by just three seconds. Even when the offense struggled in 2015, they were the slowest paced offense in the league (29.95 seconds per play).
This past season, they were marginally faster (29.48 seconds per play), which ranked 31st in the league. How did this impact the defense?
|2015||1,080||168||32.61 (21)||1.99 (17)||2:44 (20)|
|2016||1,080||166||33.04 (23)||1.89 (13)||2:45 (19)|
Despite facing two less drives, the Dallas defense still saw the same amount of plays. The unit gave up just under a half-yard more per drive, but cut down on points allowed. And with all this, the defense was on the field for one second more per drive.
Field position wasn’t even helped by the improved offense. In 2015, the defense had an average starting field position at the 28-yard line, which ranked 22nd. In 2016, the average starting field position for the defense was the 28.17-yard line, which was better only relative to the rest of the league, as it ranked 17th.
The defense did improve in efficiency, from 20th in Adjusted Defensive NEP per play in 2015 to 17th in 2016, but there’s no evidence any of the improvement came from a statistic that could be credited to the offense.
That defense is eventually what let the Cowboys down in the divisional round against the Green Bay Packers -- finding themselves down by as many as 15 points during the second half.
Despite a good game from Elliott -- 22 carries, 125 yards, and his 0.13 Rushing NEP was worth just above his full season average of 0.11 -- the Cowboys came up short in attempting to stop Aaron Rodgers. In fairness, the New York Giants, the second-ranked defense by Adjusted Defensive NEP per play, also struggled to stop Rodgers in the wild card round.
With as much as Elliott was able to do, it was other parts of the team that let the Cowboys down in the playoffs.
However, there is nothing about this season or the playoff game that's likely to change any minds from where Dallas was at the time of the draft. Those who loved the pick will look at his numbers -- both raw and advanced -- and point out one of the NFL's best running back performances.
Those against it will note that a defensive player like Ramsey could've helped the defense, while some like fifth-round pick Jordan Howard -- who ranked second in rushing yards and seventh among backs with 200-plus carries in Rushing NEP per play -- could've made a difference behind Dallas' offensive line and the overall team impact could've been greater.
It's easy to imagine the alternate draft play working for the Cowboys in 2016, but with the team happy about the pick when it happened, it's hard to think anything about this season made them sour on it.
One thing we can all agree on, though, is that Elliott did nothing to help the defense. That was silly.