clock menu more-arrow no yes

Filed under:

Reassessing The Ezekiel Elliott Impact: Once More Into The Breach

New, comments

Thanks to some inquisitive effort by BTB posters, we have some new light on just how the Dallas Cowboys expect Ezekiel Elliott to help the defense.

Matthew Emmons-USA TODAY Sports

The best defense is a good Oh-fense... you know said that? Mel, the cook on 'Alice'

Kudos to LonghornRealtor for sticking to his guns and asking the right questions about how the offense can help the defense. Kudos to NepaCowboy for nailing the crux of the issue and steering me in the right direction.

Many of you have (I can't remember everyone who has pointed this out so apologies to all I failed to mention) maintained that the time of possession and plays per game aren't all that different between 2014 and 2015. Your, not unreasonable, conclusion has been that the thing with offenses helping defenses is a mirage. But I think we've been looking in the wrong spot.

Here are the averages for the Dallas Cowboys offensive drives in 2014 and 2015:

Year Plays Yards ToP Start Score
2014 6 35.7 3:01 28.2 2.3
2015 6 31.8 2:59 24.9 -2.9

As you can see, there really isn't a significant difference between the time of possession or plays per drive in ether year.

Here are the averages for the Dallas Cowboys opponent's drives, in 2014 and 2015:

Year Plays Yards ToP Start Score
2014 5.7 32.4 2:32 27.2 -5.8
2015 6 32.7 2:43 28.3 0.7

Once again, you'll notice that the differences aren't very drastic... until you get to the last column. Now, if you didn't, go back and take a look at the last column of the first chart. Once again, you see a difference.

The offense doesn't protect the defense as much by running the clock or limiting opponents' plays, but by scoring points. This forces the opponent to play from behind. And that plays to the one big difference between a 2014 division champion and a 2015 cellar dweller:

Turnovers.

A team that is almost two field goals behind on average is going to have to press a LOT harder on offense than a team with a lead. That really is a staggering statistic when you think about it. On a typical drive in 2014, Dallas's opponents were down by nearly a touchdown.

Let's look into the statistics a little more deeply to see if this scoring thing is how the offense truly protects the defense.

In 2014, the Cowboys scored touchdowns on 31% of their possessions, and had a scoring opportunity on 48% (this includes missed and blocked FGs). In 2015, they scored touchdowns on just 14.5% of their possessions and overall had scoring opportunities on 33.8% of their drives. To put it clearly, Dallas scored a touchdown in 2014 nearly as often as it scored at ALL in 2015. That alone is huge.

But more staggering is when you break the statistics down by half. In the first half of games, the 2015 Cowboys scored touchdowns on just 13.1% of drives, though they did increase their scoring opportunities to 35.8% of drives. Still, they were clearly less effective in the first half during 2015. In 2014, the opposite was true. Their already terrific touchdown percentage actually went up almost a full percentage (to 31.8%) and their overall scoring increased to 48.3%. So the 2014 Dallas offense was actually more effective in the first half of games.

And the overall first half drive stats bear that out:

Year Plays Yards ToP Start Score
2014 6.6 38.8 3:18 26.1 0.0
2015 6 30.9 3:04 21.9 -1.5

Half an extra play per drive. 14 seconds time of possession per drive. Not letting the team get behind early. All significant differences in the way the first half of games went for the 2014 Cowboys. And the total time of first half possession was dramatically different: 258:05 and 500 plays in 2015 versus 281:20 and 559 plays. That's an average difference of 1:27 time of possession and 3.7 plays per game in the first half. That's a huge difference.

But back on point, my true contention here is that defenses in general and this defense in particular plays better with a lead, because it puts better pressure on the other team and forces more mistakes. Does that bear out?

In 2015, the Cowboys obtained three first half turnovers or on 3.5% of drives. In 2014 they had eight, for 10.1% of drives. So it does seem that the defense was indeed better at taking the ball away in general in 2014. But in the second half of 2015 games Dallas only had 7 takeaways for 8.6% of drives versus 22 takeaways for a whopping 23.4% of drives in the second half of games played in 2014. For those scoring at home, the Cowboys averaged about 1.4 takeaways in the second half of 2014 games versus less than 0.5 in the second half of 2015 games. Add in sacks now, where Dallas had 10 sacks in the second half of 2015 games, and 17 in the second half of 2014 games, despite having fewer overall sacks in 2014 and the pattern looks pretty clear.

And so there it is, the statistical evidence that a more potent offense, and particularly a better scoring offense, does in fact protect the defense and enable it to perform better. Those who had the instinct that an improved offense would allow the defense to play up to its potential had the right idea, they were just looking in the wrong place. It's not about overall plays per game or overall time of possession. It's about effective drives early in the game that rest your defense and force your opponents to press to keep up.

Certainly, the most essential part of finding is knowing where to look.