Patrick Causey, on Twitter @pcausey3
We are in the middle of self-scouting the Eagles, figuring out what has hurt this team during the first seven weeks and what they can do to improve their chances of winning the NFC East moving forward. Two weeks ago, I tried to diagnose the issues holding Sam Bradford back. Last week, I broke down why I think the quarterback has as much to do with drops as the wide receivers.
This week, I want to take a look at how Chip Kelly uses the talent at his disposal. There are two facets to this: playing time and play calling. Let’s look at this further, plus the changes that can be made to improve upon these areas.
But before we get started, a side note: I am cognizant that these have not been the most positive takes on the state of the Eagles. But that’s hard to avoid when your team is 3-4. As they say, you can only put so much lipstick on a pig. But I also want these to be productive articles. It is easy to point out the flaws, but harder to come up with the solutions. I have endeavored to point them out when I can, but always appreciate feedback and thoughts of your own. So don’t be afraid to speak up; leave comments on how you think the Eagles can fix these issues.
After acquiring Sam Bradford this offseason, many assumed that Bradford was a shoe-in to be the Eagles starter at quarterback. However, Chip Kelly insisted that Bradford would compete with Mark Sanchez for the starting job, as would every other player on the Eagles roster: “Everybody’s in competition and the best players play,” Kelly said.
A pure meritocracy sounds good in theory: a coach dividing up playing time based on production alone, regardless of a player’s past accomplishments, draft status or contract size.
But it’s not practical, not in an NFL where financial commitments and locker room chemistry must be taken into account. If you are tied financially to a person long term, you might give him a longer leash than a player to whom you have invested very little. And if you are a player that is respected in the locker room, a coach cannot unceremoniously bench you without having to deal with some repercussions.
Which explains why, with the exception of Marcus Smith, whose play has not justified his lofty draft status, Kelly has not stuck to this mantra of “best players play.” On several occasions, Kelly has stuck with a player despite evidence suggesting that an alternative would represent an improvement.
But has he gone too far?
Take last year, for instance. Darren Sproles was one of the Eagles best players, yet he curiously saw his snap count dwindle as the season progressed.
And how about Bradley Fletcher? Fletcher was a human piñata, both on and off the field, and his performance against the Dallas Cowboys late in the season — when he was burned for three touchdowns by Dez Bryant — was arguably the final nail in the proverbial coffin of the Eagles playoff hopes. But for reasons known only to the Eagles coaching staff, they continued to play Fletcher ahead of Nolan Carroll and Brandon Boykin.
Ditto with Trent Cole. I respect Trent Cole immensely; he was a classic lunch pale, blue collar worker that is the perfect embodiment of what this city stands for. But he clearly lost a step last year while Brandon Graham was wreaking havoc in limited playing time. Yet, it was not until Cole broke his hand in December that Kelly finally made the switch, simply because he had no other choice.
And here we are again, seven weeks into the 2015 season wondering why Kelly continues to rely on certain players when the backups are proving to be more effective.
Mathews v. Murray
The most obvious example this season is Kelly’s decision to stick with DeMarco Murray over Ryan Mathews, despite overwhelming evidence that Mathews is the better player at this point in his career and in this offense.
I covered this in-depth earlier this year, so I won’t rehash the same old material. But here are examples of the two biggest issues with Murray so you can see what I mean.
First, Murray has been inconsistent with his reads, leaving too many plays on the field as a result. Against the Saints, Murray failed to see and take advantage of an easy opportunity to gain yards down in the redzone.
As you can see, Murray has two lanes to attack. At a minimum, he should be able to get to the next level before having to beat the single high safety. Instead, Murray tried to bounce the play outside and fell before he was able to gain any yards:
Even when Murray has made the right reads, he has been a step too slow to exploit the hole. This loss of explosiveness and drop in production was to be expected. Any running back that has touched the ball at least 370 times in a single season had his production fall off a cliff the following year, to the tune of a 39.2% drop in production on average (for some, like Larry Johnson, Terrell Davis and Jamaal Anderson, it was much, much worse).
Murray touched the ball an absurd 497 times last season (392 regular season rushes, 57 catches; 44 carries in the playoffs, four catches). That is 33% more than the 370 bench mark for production decline. It should be a surprise to no one that Murray has lost a step this year.
Meanwhile, Mathews just makes plays whenever he is on the field. He is a decisive, imposing running back that fits the downhill running style that Chip Kelly wants in this offense. He also looks much more explosive in the run game, as evidenced by his 6.1 yards per carry, compared to Murray, who is averaging a paltry 3.5 ypc.
Compare virtually identical plays (save for minor formation changes), and you can see the difference in speed and decisiveness between the backs:
Here is Murray running ta staple of the Eagles offense, the outside zone.
Now watch Mathews, running the same play out of a slightly different formation:
Ignore the result for a moment, and focus on how much faster Mathews makes the right read and explodes through the running lane. To be sure, Mathews had an easier hole to attack, but its not like Murray didn’t have anything to work with. Look at this lane:
It is fair to wonder whether Kelly’s loyalty to Murray has cost the Eagles a loss or two, especially in a winnable game against the Carolina Panthers. After Mathews broke off a 63 yard touchdown in the third quarter, he did not receive a single carry the rest of the game.
His absence was magnified after the Eagles defense picked off Cam Newton following Mathews’ touchdown run, which set the Eagles up at the Panthers 18 yard line. Murray received three carries and gained a grand total of one yard on that drive. Mathews sat on the sideline for the entire series. The drive stalled, and the Eagles settled for a field goal to make the score 21-16, Panthers.
It’s unclear why Kelly has stuck with Murray so far. Perhaps Kelly is trying to prevent Mathews — who has an injury history himself — from breaking down as we get deeper into the season (entirely understandable). Perhaps Kelly sees the $7 million the Eagles owe Murray next year and wants to give Murray every shot to validate the contract (somewhat reasonable, but still not smart). Or perhaps Kelly is just loyal to a fault and/or unable to recognize when a backup deserves more playing time (indefensible).
Regardless of the reason, Kelly needs to make the switch. Mathews should get more carries moving forward, and a case could be made that Sproles needs to get more touches ahead of Murray as well. It might upset Murray, but with the team at 3-4 and in desperate need of a division win this weekend, hurt feelings are the least of the Eagles worries.
Cooper and Austin
A more subtle issue is the frequency at which he is relying on Riley Cooper and Miles Austin to make plays in the passing game. Look, I get it: none of the receivers have been lighting the world on fire. So it’s not like Cooper and Austin are getting playing time over a Julio Jones-esque player.
But Cooper and Austin have been especially poor, and Kelly’s continued reliance on them is questionable, at best.
Kelly had the stated goal of wanting to improve the Eagles depth this offseason, especially on offense. The idea behind this was simple: Kelly runs a lot of plays, fast, and wants to be able to rotate players in without having to adjust his play calling.
The strength by depth approach sounds good in theory, but it hasn’t worked out so far this year. Consider the following break down of the Eagles receivers this season:
|Name||Targets||% of Targets||Catch %|
A lot has been made about Matthews, Ertz and Sproles dropping the ball — and rightfully so. But if you compare their catch percentage to other receivers around the NFL, they actually are in pretty decent company:
Matthews has a 62% catch percentage, which is comparable to Julio Jones (65%), Odell Beckham (64%), Randall Cobb (64%), Jarvis Landry (64%), Demaryius Thomas (62%) and Calvin Johnson (63%).
Sproles (60%) and Ertz (57%), meanwhile, are among Allen Hurns (61%), Marques Colston (61%), Brandin Cooks (60%), Emanuel Sanders (58%), and DeAndre Hopkins (57%).
If we expect some regression to the mean for the drops on all three (which I think is fair, given that their current drop rate would be historically bad), those catch percentages would improve even more.
But Cooper and Austin? Their 50% and 52% catch rates, respectively, are among some of the worst in the NFL, ranking them along side the likes of Ted Ginn, Jr. (49%), Malcolm Floyd (50%), TY Hilton (50%), and Allen Robinson (49%). Despite their bad production, they still account for 16% of the team’s total targets.
So how can Kelly fix it? I would like to see the Eagles shift some of those targets to Matthews, Ertz and Sproles, and even Huff and Agholor. Yes, the latter two have been underwhelming so far, but it is too early to give up on them given their age and potential. And that is especially true with Agholor, who I suspect will see a resurgence in the second half of the season much like Jordan Matthews did last year.
But I think, no, I am quite certain, that the same improvement cannot be expected from Cooper or Austin. Cooper is a six year vet, Austin (who used to be very good), is in his ninth season. Expecting a resurgence from either of them at this point in their careers is unreasonable.
Kelly can offset the loss of Cooper’s run blocking skills by relying more heavily on the 12 personnel we have seen emerge in recent weeks (two tight ends, two wideouts, 1 running back).
Celek represents an upgrade over Cooper in the blocking category, so leaving Cooper on the sideline for someone like Huff or Agholor makes sense. And given Ertz’s versatility and strength as a pass catcher, there really isn’t much of a downgrade in the passing game.
I also think the Eagles can look to how the New England Patriots utilize former Eagle Dion Lewis for ways to get Sproles more involved. Lewis is averaging 86 yards per game through the air and on the ground, with the Patriots lining him up all over the field.
In the blowout win against the Miami Dolphins on Thursday night, the Patriots lined up Lewis out wide against a cornerback:
Here is a better look at the route:
The Eagles have split Sproles out wide a handful of times this year. But he is still being underutilized. There is no reason they cannot increase the frequency with which Sproles lines up as a receiver, and continue to look for ways to get Sproles involved in the screen game. He is a dynamic weapon that few defenses have answers for, but right now the best defense for Sproles seems to be Kelly’s unwillingness to use him.
And while having Mathews, Sproles, Ertz and Celek on the field at the same time might not be conventional, they are our best offensive weapons right now, so it makes sense to throw convention out the window in pursuit of more wins.
Kelly has invested $11.61 Million in the running back position, good for third highest in the NFL. It was a curious decision given how much the NFL has devalued the running back position. The Eagles currently spend more money on running backs than the Denver Broncos, New England Patriots, Atlanta Falcons and Arizona Cardinals combined, who are a collective 26-4 this year. The teams in the top four of money spent on running backs? The Minnesota Vikings, Chicago Bears, Philadelphia Eagles and Houston Texans, who are a combined 13-16.
While we can debate the merits of Kelly’s investment in the running back position, there really can be no debate that Kelly has under utilized those backs throughout the season. Kelly has actually called the least amount of run plays in his time with the Eagles, as you can see in this chart, which breaks down the run to pass ratio (rankings in parenthesis):
|2013||53 (27)||47 (6)|
|2014||57.82% (21)||42 (12)|
|2015||60.17% (15)||39.83% (18)|
Earlier in the year, the Eagles’ offensive line could not run block to save their lives, so Kelly had no choice but to abandon the run. But since the Jets game, the run blocking has improved considerably. So it is unclear why Kelly continues to abandon the run game, especially given how poorly both the receivers and quarterback have played.
Take the Carolina game, for instance. With the exception of the tail end of the fourth quarter, the Eagles were not in a position where they needed to abandon the run. Yet, Kelly kept dialing up the pass, calling 51 pass plays (46 passes, 5 sacks) to just 30 runs.
Why is balance so important? Well, for starters, the Eagles have a much better win percentage when they have a more balanced attack. The Eagles are 12-2 when they run more than pass. But when the inverse is true? The Eagles are 10-15.
I understand that correlation does not necessarily equal causation; but it is hard to ignore this sample size. 12-2 and 10-15 are not statistical aberrations. They are large enough sample sizes from which to draw the conclusion that the Eagles are much better when they take a balanced approach.
The other reason is that it will take pressure off of Sam Bradford and the wide receivers, and open things up in the passing game. Look at the difference between the Eagles offensive production in 2013, when they asked Nick Foles to throw the ball on average only 31 times a game, compared to 2014, when Foles was throwing it 39 times a game.
Or look at how Tony Romo enjoyed the best year of his career last season, when the Cowboys ran the ball 49% of the time, good for third most in the NFL. Romo was 12-3, completed 69% of his passes, threw 34 touchdowns, 9 interceptions, and had a 113.2 quarterback rating. That was career best marks in completion percentage and quarterback rating, and the second best marks in his career for touchdowns and interceptions.
Simply put, the Eagles invested heavily in the running back position. It is time they start using it.
So in short, the Eagles need to consider making the following changes.
- Give Mathews more carries than Murray.
- Stop relying on Cooper and Austin so much in the pass game. Divide their targets among Matthews, Ertz, Sproles, and even Huff/Agholor.
- Run the ball more. The team wins more often and it opens up the passing game.
Great analysis – for all the ballyhooed notion that Chip Kelly is an offensive genius – he is far from that – I really do not understand the media’s love for him – the above are plain and simple things but of course Coach Kelly thinks he is way smarter than anyone else. Investing in Demarco Murray was idiocy from the start – Kelly keeps talking up Sproles and does nothing with him – least he can do is look at the Patriots and see they utilize their personnel
Thanks much, Max. Glad you took the time to read and enjoyed.
Perhaps genius is too strong of a word, but I do think Kelly is a very good offensive mind. One thing the tape has repeatedly showed this year is that, at least for the most part, the scheme is working. Receivers get open, but the QB misses him or the offensive line doesn’t give the QB time to throw. The offensive line run blocks well, but the RB misses the lane (or vice versa). So I would not write off Chip Kelly yet.
With that said, he definitely needs to do a better job at utilizing the talent at his disposal. This is still a players league, so it is time he starts relying on the best ones to make the plays.
I’d also like to see an improvement in game-planning. It’s pretty obvious, but the team is much better in the second half than first half of games. CK is good at making in-game adjustments, but the team can look so bad in the first half.
The alternative explanation is that CK relies on tempo so much that it takes a half for the other team to wear down. Under this reasoning, it’s not actually game planning that’s the problem. I buy this to a certain extent, but not fully. If the Eagles get a couple first downs in a row, they’ll start to roll. Tempo teams work by momentum (Collinsworth said that 2 first downs are the magic number for tempo teams, given his time in CIN). The early 3 and outs are killer, and it’s hard for me to buy that this is ok under whatever game plan has been developed.
You touched on an issue I have thought about but have not had the chance to write about yet. I have wondered whether it is game planning or whether the offense just works better when defenses are tired. The more I see the team play, the more I am starting to worry it is the latter. Look at the second half of the Falcons game, when the offense started moving the ball at will because Atlanta got gassed. The same could be seen against the Jets.
If that is the case — and I can’t say definitively one way or the other — that is extremely problematic. You cannot rely on an offensive scheme that only works when the defense is tired. It leaves you with only half the game to make an impact. Will be something worth watching going forward.
Very good points
Much appreciated Mark.
Pingback: Inside the Huddle Part 1: Run Baby Run | Eagles Rewind
Pingback: The Eagles are Bad and Chip Kelly (the GM and Head Coach) is to Blame | Eagles Rewind
Pingback: Chip Kelly, the Learning Curve, and the Quarterback | Eagles Rewind