On Tight Ends, Offensive Schemes, and Likely Lies

I've rehashed this Chip Kelly statement a couple of times now, but it remains one of the most interesting quotes so far this offseason:

"We are going to go three tight ends in a game. Now, if they go three linebackers, we spread them out and if they go DB's, we smash you. So, pick your poison. Simple game, isn't it? You guys thought coaching was hard. They bring little guys in, you run the ball. They bring big guys in, you throw the ball."

There's a lot to unpack in this short passage, including possible insight into Kelly's overall offensive philosophy: hit 'em where they ain't. It also confirms (as if the Zach Ertz pick didn't say this already) that Kelly wants to emphasize versatility with multiple tight end sets. However, at its most literal level, I think it's a lie: I don't believe the Eagles will base any sort of offense around having three tight ends in the game at once.

It's an easy fib to tell at this point in the offseason, especially when you need to explain away a crowded tight end position. Brent Celek has been an above-average starter for the last few years, a consistent threat in the passing game who isn't afraid of the dirty work. Kelly doesn't want to say that he's looking to replace the veteran leader. Meanwhile, the team signed James Casey to play an expanded H-back role and drafted Ertz just three picks into the second round. One might say there aren't enough snaps to go around if Kelly hadn't used some expert hand-waving.

Alright, so you say, what evidence is there that the team can't or won't use three tight ends at the same time? Plenty. Let's start with current stats. The New England Patriots are considered one of the most tight end-friendly teams in the NFL. Some, like Chris Brown, have even suggested that Kelly's pro offense will look more similar to the Patriots' than his own college spread. Like the Eagles this year, last year's Patriots squad ditched the fullback in favor of more tight ends/H-backs. However, they rarely used three such players at the same time. Looking at snap counts from Pro Football Focus, we can see exactly how often each Patriots tight end was on the field game-by-game. The players themselves aren't consistent due to injury, but in only 9 of 18 games of their games was the second-most active Patriots tight end on the field for even half of all offensive snaps. The third-most active tight end averaged only 15% playing time. 

However, even individual play counts don't properly convey how rare it was for the Patriots to line up with all three tight ends. According to the NFL's own game stats service, New England used three or more tight ends on the field together in only 6.4% of all snaps. That doesn't mean that Kelly couldn't play that combination more. It's the most cliched statement of the offseason that we don't really know what his offense will look like. But the Patriots' unwillingness to go to three tight end sets with any regularity underscores the difficulty with such a formation.

Kelly suggests that he can do anything out of it, but that's not really true. Keeping Celek, Casey, and Ertz in the game together means you only get two other skill players. One of them is obviously a running back like LeSean McCoy. Then you get one wide receiver -- presumably DeSean Jackson or Jeremy Maclin.  Between the 20s this lineup lacks any ability to stretch the field. The three tight ends are valuable, flexible pieces. They can pose match up problems against linebackers and safeties in the pass game and cornerbacks in the run game. But Ertz is the fastest, and he only runs a 4.67 forty. With only one wide receiver to deal with, the defense can bring an extra man down into the box to thwart both the short passing game and the run. It's not an ideal match up for the offense -- compared to a two tight end set that keeps the defense honest with down-the-field threats on the outside while adding flexibility in the slot that a player like Jason Avant can't provide.

The three tight end lineup is fantastic for one thing: the red zone. Bring a dangerous rushing threat together with a bunch of big targets and the Eagles might actually be efficient down by the goal line. But I remain skeptical that any offense will use the formation much beyond that. Two tight ends should quickly become a bread-and-butter package for this team, and with injuries and substitutions, having three "starters" on the roster isn't a bad thing at all. But let's not get carried away beyond that. If Ertz progresses quickly, the Eagles likely will be looking for a suitor for Celek 11 months from now.

PS: There's an unwanted player out in free agency who has a similar athletic profile as Casey, and might become an even more versatile weapon with some innovative coaching. It's too bad the Eagles have already denied interest.

PPS: Support the Eagles Almanac Kickstarter if you haven't already!

Photo from Getty.

Geno Smith and Expected Draft Value

I hope everyone is reading Brent Cohen's work over at Eagles Rewind. He's playing with stats and coming to interesting conclusions, especially with his PVM (Positional Value Multiplier) prospect rankings that add a discerning adjustment to the standard draft preparation.

Brent's goal is to take the consensus draft prospect rankings and apply two main modes of adjustment. First is a risk factor that attempts to quantify how reliable different positions are. He shows that offensive tackles, for example, have a bust rate of only 26%, compared to 50% for quarterbacks and 66% for running backs. One can quibble with the metric (you might adjust for likelihood of Pro Bowl appearances instead of failure to start), but it's a good one to have. The second main adjustment is for positional value, i.e. how much one player is worth relative to another at a different position. Brent uses salary cap data to show a clear hierarchy, from centers and guards at the bottom to defensive ends and quarterbacks (far and away) at the top. Again, you can quibble with the exact numbers, but it's quite promising stuff.

The PVM rankings that emerge are interesting to look at. Certain players receive bumps -- especially quarterbacks. Others fall slightly, like guards and centers. The final result is not a radical departure from the original consensus, but that's not Brent's goal.

It is mine, though. I look at the numbers pulled, especially on salary cap figures, and it's obvious to me that there's a huge disparity in value between different positions. Brent's cap percentages are at right, and you can see that while the middle group is close, there are certain spots that vastly outweigh others. As I described above, the PVM rankings make an adjustment along these lines, but it doesn't come close to matching the discrepancies in value. In expected value terms, if you have two similarly-talented prospects -- one a center and one a defensive end, you should take the defensive end every time. In fact, if you take it literally, the center would have to be three times as good of a prospect as the end to warrant consideration. 

Maybe this is too extreme of an adjustment for your tastes. If you want something closer to the consensus prospect values, Brent's PVM should serve you well. But I think this more accurately represents the reality on the ground. Chance Warmack may become a Pro Bowl guard -- many scouts rate him as the best player in the draft. But there's a strong case to be made, based on this data, that he would have to be 1.5 times better than the next best cornerback, 1.8 times better than the next best defensive end, or 2.7 times (!!!) better than the next best quarterback.

Plenty of caveats apply here, including the risk factor that I mentioned above. And you may either disagree with the salary cap distinctions entirely or have a more refined model based on it. I also haven't run my own projections for using this data to provide an updated draft list.

But if you look at the draft through this qualitative lens, it's easy to make an affirmative case for a quarterback, even Geno Smith. At this point I don't expect the Eagles to draft him. Most scouts consider Smith the best quarterback prospect, but also suggest that he wouldn't be so in previous years. Average rankings seem to consider him about 20th overall. That doesn't seem like the type of player you want to use your top five pick on.

However, I've said before that you always take the quarterback if you think he's your guy and the data backs that up. A good starting quarterback is worth nearly 50% more than the next most valuable position, according to salary cap numbers (and one might argue that's understating it). It's importance towers over other concerns. That means that even if Smith has a worse chance than average to become a good starter, his expected value from that smaller probability is still high -- likely higher than many players who will be drafted ahead of him. Smith may not be the safe pick, but if the goal is to maximize value, he may be the right one.

Photo from Getty.

Pass Rushers Here, There, and Everywhere

In Chip Kelly's second press conference in Philadelphia, he told us that he liked the 3-4 defense best, but that scheme flexibility was of the major reasons he hired defensive coordinator Billy Davis:

What we're going to do is put our guys in the best position for them to make plays. I don't know if that's being a 3-4 team, a 4-3 team, a 5-2, a 6-1 team. I know we could add up at the numbers to see where we are. We're also not caught up in that. It's about making sure we play sound defense on first, second and third down. We could look drastically different on first and second downs than third downs. And that's going to be entirely personnel-driven for us. Could it be a 4-3 under defense? Yeah, or it could be a 3-4 under defense. 

Since saying that, Kelly and Howie Roseman have added new pieces to the mix. Isaac Sopoaga can immediately step in at nose tackle. Connor Barwin is a starting outside linebacker. These are important steps in the direction of a 3-4, or a 4-3 under defense (as discussed previously). But which is it? Does it matter? What's the direction to take this year, even if it's only a stopgap measure? The key to answering those questions is in the sudden surplus of pass rushers the Eagles seem to have between Barwin, Trent Cole, and Brandon Graham. Here are their numbers from Pro Football Focus, including the last two years for Barwin:

Before we get into ramifications for the defense, let's discuss each player in turn. Graham wasn't a full time starter last season, although only Andy Reid knows why. When he was on the field he was an absolute beast in every phase. Graham put up the highest Pass Rush Productivity mark of any defensive end in the league. He was also a top run defender -- ranking third out of the top 37 qualifying players in tackle and stop percentage.

Cole's performance was mixed in 2012. Despite being slightly undersized, he has always been solid against the run. That seems to have held up in his age 30 season. His run defense percentages were nearly as good as Graham's. But Cole definitely suffered a decline in his pass rush. From 2009-2011, he ranked seventh, third, and first respectively in PRP. This year he dropped to 18th. That still put him in the top half of 4-3 defensive ends, but only barely.

Finally we get to the newcomer Barwin. Truthfully, his numbers aren't pretty. As a 3-4 OLB, he was classified differently than Cole or Graham, but he didn't impress compared to his peers. Even during his 11 sack season in 2011, he was only middle of the road in total PRP, suggesting that the sacks may have been a lucky break despite less actual pressure. Barwin was also a subpar run stopper.

So can all three coexist at outside linebacker in a standard 3-4? That's going to be tough. Based on these stats, Graham needs to be on the field at all times, no matter what the role. Barwin wasn't signed to be a backup, either. He needs to be slotted in as pass rushing OLB. Where does that leave Cole? In a world where his contract wasn't so large, it probably leads to his trade. But if he's staying, it's silly to see him backing up a (worse) player like Barwin. And Cole's potentially-declining pass rush doesn't lend itself to a specialist role either.

What if the answer is relatively simple -- Cole goes back to playing more or less his old position of defensive end, but in the new 4-3 under scheme? He's no prototypical 5-technique defensive end, but Cole works there if you think more about the player and less about a rigid scheme. First, he's still great against the run, which is that player's primary job. Second, it allows him to keep his hand on the ground, rather than convert to a stand-up linebacker in his ninth season. Third, it keeps the team's best personnel in the game, and actually aids scheme flexibility. (Tinfoil theory: if Vinny Curry is bulking up to compete at DE rather than slimming down for OLB, that could be hinting at this move.)

As you can see from the numbers above, both Barwin and Cole played the vast majority of their snaps on the right side. Set them up on that side again, with Cole as the DE and Barwin as SAM. Across the formation, Sopoaga is the nose tackle on run downs, with Fletcher Cox is the penetrating 3-technique defensive tackle. Graham, hopefully your star, mans the "Predator" DE/OLB spot on the left -- where he lined up all of last year.

Because offensive alignments will shift, the defense won't necessarily have clear strongside-weakside assignments, but it will have flexible pass rushing duos. Graham + Cox and Cole + Barwin are dangerous pairings for an offense to handle. Whether Graham or Barwin are standing up or putting their hands on the ground doesn't really matter, they should both get one-on-one matchups on the outside when rushing the passer.

If Kelly was telling the truth about crafting the defense to fit his players, this is the direction he should head. Rather than forcing Cole to convert positions and sit on the bench, keep him where he's comfortable. In a year or two, after you've groomed young backups, you can revisit whether you need to replace Cole with a more traditional 3-4 end. Until then, live those "personnel-driven" words you preach.

Photo from Getty.

The Rise and Fall of the Andy Reid Era

I know we've all moved on to the wonderful potential of the Chip Kelly era, but I wanted to take a moment to look back at the era that was, the 14 years of longstanding success and, ultimately, failure under Andy Reid.

Building on work I started in 2011, and inspired by previous work by Derek Sarley, I put together a few graphs that I hope will illustrate one aspect of the Reid era's rise and fall -- as well as the core precepts of team building that the new regime needs to learn from. Below is a graph showing the Eagles starters each year, colored by years experience with the team. For example, the yellow demarcates starters for whom this is their first season in Philadelphia -- both rookies stepping right into impact roles and free agents signed off the street. Check it out:

Hopefully the graph looks less like a jumble of colors and more like a series of patterns that you can follow. Look, for example, at those first year starters. It spiked in Reid's first season, when he drastically remade the roster. However, it fell off quickly and stayed low -- until the great free agent binge of 2011. Another piece to look at is the top of the graph. Green demarcates the culture-carriers, the players who have been with the team for eight or more seasons. Check out how after Reid took over, he eliminated any players with that distinction, only to have the group slowly reappear and grow from 2003 to 2008. 

Here's another, simpler version of the graph, in which I grouped some of the levels together:

You can see the broader trends easier when you break it down into just three categories. In Reid's first two seasons, he culled the longstanding veterans and added a bunch of new players -- rookies like Donovan McNabb and free agents like Jon Runyan. But during the celebrated run of four NFC Championship games, Reid maximized stability and growth and minimized turnover. You can also see clearly the mini-rebuild he embarked upon following the disastrous 2005 campaign. However, you'll note that while first and second-year starters spiked, the number of starters who'd been in Philly for seven years or more stayed flat. Reid had kept longstanding veterans like McNabb, Runyan, Brian Westbrook, Brian Dawkins, and Tra Thomas while rebooting with youngsters like Todd Herremans, Mike Patterson, Trent Cole, Jamaal Jackson, and Sean Considine.

What's fascinating is the third wave, another rebuild that started in 2010 but never really ended. Before the 2010 season, the Eagles had moved on from stalwarts like McNabb, Dawkins, Westbrook, Runyan, and Thomas. The problem is that as they eliminated those players, the replacements never came into their own. Some young players stepped in, like DeSean Jackson and LeSean McCoy. But guys like Nate Allen and Danny Watkins fizzled. Free agents were brought in by the boatload in 2011 to shore that up, but they caused more problems than they solved. In Reid's last two years, half of the starters were basically new to the team -- more than double the amount during the 2001-2004 run.

Here's another related graph, showing the number of starters in each year who have either been with the team for two years prior or were actually starters for two years prior:

The same pattern emerges. From 2002-2004 the team had 11 or more returning starters, and 15 or more starters who had been with the team for a few years. For example, in 2003 Jermane Mayberry missed most of the year due to injury. Bobbie Williams, in his third year with the team, stepped up to replace him. In 2004 Lito Sheppard and Sheldon Brown took over full time from Bobby Taylor and Troy Vincent. The new pair were relatively inexperienced, but it wasn't their first year on the job either. That makes a big difference. Compare that to 2011. The squad had three rookie starters, three sophomore starters, and four new free agent starters. That kind of discontinuity was a recipe for failure.

The lesson Kelly and company need to learn from the Reid era is that you can't build a Super Bowl team in a year. You have a lot of turnover when a new coach arrives, but the front office has to quickly shape the roster and find a core of young players who can grow together. By the time you're ready to compete, 2-3 years later, the team should be stable. And if you're going to rebuild, half-measures and plug-and-play starters won't do. There are no shortcuts to success.

Photo from Getty.

Lights, Cameras, Revolution

I don't often link to articles not about the NFL or the Eagles, but this one deserves a read. Zach Lowe of Grantland gets an unprecedented inside look at the camera-tracking system used by the NBA's Toronto Raptors to analyze and compute optimal playing styles. I don't want to be that guy who always yells, "This is the future of sports analytics," but it sure is cool. Some NFL teams are also experimenting with cameras, GPS, and other similar systems.

Putting the Stopwatch on QB Release Times

Fantastic work by Jimmy Kempski (two links in two days). He timed every throw by Michael Vick and Nick Foles last year:

The Eagles have a very dynamic set of skill position players. When your QBs are holding the football for 2.87 seconds on average, you're essentially asking the QB to do a large portion of the work in your offense. That is not ideal when you have one aging QB that is in steep decline and another that was a rookie 3rd round pick. The quicker you can get the ball out of your QB's hands and into the hands of the guys who should be making plays, the better the offense will be.

Definitely check out his full breakdown. You can even tell when the Eagles decided it was time to change things up for Vick (as confirmed by All-22 tape).

The Past and Future at Defensive Tackle

On Monday, the Eagles parted ways with two veteran defensive tackles. The first, Cullen Jenkins, was a late-career mercenary hire in 2011, part of the free agent splurge that Joe Banner and company thought would put the team over the top. The other, Mike Patterson, was the longest-tenured Eagle, drafted with the team's first pick after the the Super Bowl. Neither cut was surprising; the new regime wants to get younger and less expensive. 

However, the moves do give us an opportunity to step back and assess where the Eagles were coming from after last season and where they're going in the middle of the defensive line. Let's look at some 2012 stats, courtesy of Pro Football Focus:

The first two stats, Tackle % and Stop %, are about run defense: per snap, how often did the defensive tackle get in on the play and how often did he cause a stop. As you can see, the young guys led the group. Out of the 52 DTs with the most snaps, Fletcher Cox made the eighth-most tackles per play and the 12th-most stops. And that was only his rookie season. Cedric Thornton wasn't too far behind, posting above average numbers in both categories. While Mike Patterson barely played, the two other veterans Jenkins and Derek Landri logged more run snaps but came up short. Both players ranked toward the bottom in tackles and stops.

Moving over to pass rushing, we see that Jenkins played more of those downs than anyone and was still effective in that role, with the 14th-best Pass Rush Productivity rating (Pressure per play). However, the rookie Cox actually edged him out. In (mostly) Jim Washburn's scheme, Thornton and Landri were also reasonably effective. Both were above average out of the 69 most active pass rushers.

Alright, so that's where the team stood at the end of the season. What does it mean, and where do they go from here? With Patterson's injury keeping him off the field and no dead money on his contract, he was an easy cut. Landri didn't impress on tape or in the stats department and will be allowed to walk as a free agent. Jenkins requires slightly more explanation. Some people have suggested that his experience in the 3-4 would be valuable in 2013. But I'm not sure how much value he would actually have in Billy Davis' defense (or any 3-4). Jenkins still shows flashes as a pass rusher, but he was a liability against the run last year. This isn't someone the Eagles can stick at 5-technique defensive end.

What do the Eagles need from their defensive tackles? With the edge pressure positions likely filled through 2012 holdovers (Brandon Graham, Trent Cole) and the draft, the team needs a dominant inside pass rusher, a lane-clogging nose tackle, and a strongside all-around 3-4 end type. The inside one-gapping defensive tackle is an easy choice. That's Cox's role all the way. From the start he's been a fantastic rusher and run defender, and he should only get better. The nose tackle is a tougher position to fill. Returnee Antonio Dixon has the physical skills to play there, but there will certainly be other candidates brought in.

The question in my mind is Thornton. Jim Washburn liked him because he, like Cox, had inside pass-rushing skills. But breaking down the numbers, Thornton was actually a better run defender in his first extended NFL action. At 6'4", 309 lbs, he's a physically-gifted player. You could slot him in as Fletcher Cox's backup easily and hope that enables more pressure. But why not give him a shot at the 3-4 style strongside end? That position values run-stopping first, and Thornton can do that already. When the nose tackle comes off the field on passing downs, he can slide inside and provide interior pressure. With the old guys out of the way, that seems like a move that could maximize current talent in 2013 and beyond.

Photo from Getty.

I Come To Bury DRC, Not To Praise Him

Good friends, sweet friends, let me not stir you up
To such a sudden flood of mutiny.
They that have done this deed are honorable:
What private griefs they have, alas, I know not,
That made them do it: they are wise and honorable,
And will, no doubt, with reasons answer you.

Like Asante Samuel before him, Dominique Rodgers-Cromartie drew a lot of righteous anger during his stay in Philadelphia. Despite natural athletic talents that rival any cornerback in the NFL, Rodgers-Cromartie floats through the game of football. He was an awful tackler even when he tried, and he rarely put in the effort anyway. He gave up many more penalties and long plays than any "Pro Bowl" corner should. In short, despite those tremendous outfits, he was nearly impossible to root for.

To that end, I have little quarrel with the Eagles deciding to let Rodgers-Cromartie walk without applying the franchise tag, as Geoff Mosher reported. If Chip Kelly doesn't want him, nor Howie Roseman (who acquired him in 2011), nor defensive coordinator Billy Davis (who coached DRC in Arizona), who are we to argue? It's not like the cornerback stood out last year. Here are the coverage numbers, from Pro Football Focus. The ranks show their place among the 77 cornerbacks who played at least half of their team's coverage snaps in 2012:

Looking at DRC's stats, you can tell he was no star. Of course, none of the Eagles corners were. Brandon Boykin held his own in the slot, proving himself admirable, if not spectacular in his rookie season. Nnamdi Asomugha continued his long streak of scaring quarterbacks off, posting the fourth-lowest targets per snap in the league. Of course, when they actually did throw his way, Asomugha was awful. He ranked 63rd in completion percentage and dead last in yards per target.

Rodgers-Cromartie doesn't deserve any breaks either. True, his target and completion rates were above average, but his performance declined dramatically over the course of the season as he seemed to lose interest. See the table at right, as DRC's quarterback rating against dropped after a good start to the season through week six.

This is probably the point where we should bring up the event that occured after week six: the firing of Juan Castillo. Remember what the cornerback said when Castillo was kicked out the door?

Most of the players have had fond words regarding Juan Castillo the man, but few have stood up for him as a coach quite like DRC did on Thursday. Rodgers-Cromartie revealed that it was Castillo that transformed him into a press corner, working overtime with him to make sure the 26-year-old went through the change as smoothly as possible.

“No question. He changed my game. He turned me into a press corner,” said Rodgers-Cromartie. “He’s one of the guys that took the time with you, to make you understand the weakness in your game and keep you after practice and make you work on it. This is the NFL, you’re accountable for yourself. Not too many people are going to say, ‘Hey, you need to do this and make sure you do it.’ They tend to let you do it on your own.”

Rodgers-Cromartie is capable of playing at a high level in spurts, but to do so consistently he needs special attention from coaches. That's not something we expect Kelly and his team of 50 assistant coaches (give or take) to have time to do, right? If DRC's performance goes downhill after his favorite coach is fired and the rest of the defense is falling apart, surely we don't want him around any more.

Time to move on to the team's other alternatives. Restructure Asomugha's contract, sort through the free agent bin and draft heap to find another adequate starter. Moving Boykin out of his natural slot position could work. Or perhaps one of the gang of invisible backups is will start showcasing Pro Bowl talent. Whatever the result, letting a young, capable, and inconsistent starting cornerback go for nothing is definitely the right move. There have to be better ways to spend the Eagles' plentiful salary cap space. DRC, don't let the door hit you on the way out.

Photo from Getty.