/cdn.vox-cdn.com/uploads/chorus_image/image/23264741/152094843.0.jpg)
Chip Kelly's first ever player acquisition -- on the opening day of free agency -- was versatile tight end James Casey. He then spent a second round pick on TE Zach Ertz, who burned Kelly's Ducks for a key touchdown in their only loss last year, to Stanford.
Once the season began, though, he barely used either. According to Pro Football Focus, the Eagles have stuck with an "11 personnel" package -- 1 back, 3 wide receivers and only one tight end -- on 80% of their offensive snaps. That's second only to Green Bay's 84%; the league average is 51.62%. And most of the time, that TE is Brent Celek.
Why? The general hunch has been that it spreads the field more and makes it easier for for LeSean McCoy to run. Now, Mike Clay at Pro Football Focus has boiled down a vat of data and distilled some statistics that show more precisely how that works - and have successful the strategy is.
In a nutshell, the Eagles' 11 formation forces defenses into a nickel package, and that's exactly what Chip Kelly wants. Generally, defenses want to use their base defense on runs and bring in the nickel package in passing situations. But with two strong quarterbacks and, now, two potent wide receivers, defenses are forced into the nickel essentially every time the Eagles go with 11 personnel.
It sounds impossible that Chip Kelly could control the defense's choices that thoroughly, but Clay's numbers bear it out. Most teams face base defense 60-70% of the time on designed runs; for Minnesota, it's 80%. On dropbacks, that percentage typically goes down to 25-35%.
Even on designed runs, Philadelphia faces base defense only 16% of the time. Opponents go to nickel on fully 82% of the Eagles' run plays, barely less than on passing downs (85%). Detroit (64%), Denver (58%) and Green Bay (54%) are the only other teams facing nickel on more than 50% of runs.
Granted, the read option makes the definition of a "designed run" tricky, but that's precisely why they run the play. Compare Sunday's opponent, Washington, which also blurs the pass/run line with a great running quarterback and read/option plays. The Skins still face a base package on 76% of designed runs, nickel only 20%.
I've been as frustrated as anyone that Kelly has not made more use of his tight ends, and I got excited when we saw more 12 against the Raiders. I considered that evidence that Chip had shifted his focus from Michael Vick to Nick Foles, and the result of him finally listening to my sage advice was of course a Hall of Fame game for Foles.
Kelly has insisted all along that match ups with the defense drive his play calling, not his choices about what offense he wants to see. He won at Green Bay by going back to an 11-dominated attack. We may see more two-tight-end sets against Washington, because they are terrible at open field tackling and have some decent pass rushers, but again that's a match up question, not a change in the basic strategy.
Clay's numbers back up Chip's words, and document exactly how powerful his approach has been. Basically, he has achieved every coach's goal -- dictating the formations of both teams in a way that plays to his squad's strengths. His rushing-focused team faces one fewer opponent in the box as a general rule. Math wins, once again.
The exciting thing is that this offense is barely taking shape. As Clay notes, "the Eagles sit No. 4 in the NFL in offensive touchdowns per game in Year ½ of the Chip Kelly offense."
Kelly values versatile players, and drafted Zach Ertz in part because he has the speed and moves to (eventually) play a legitimate wide receiver in the NFL, as well as the size and blocking to play a traditional tight end. At that point, how will you even tell if the Eagles are in 11 or 12? If Ertz is in the slot, what position will he be playing?
Even now, facing nickel on almost every down, the Eagles lead the league in pass completions over 20 yards. With Ertz developing and Maclin potentially returning next year, they can get the advantages of two tight ends and keep running against a thinner box, without losing anything from their passing threat.
That offense would make this one look like a pencil sketch of the final result. Imagine an offensive package that can run anything from downhill plunges to "everybody go long" without a single substitution, based on post-snap reads. That's an offense that really would revolutionize the NFL.