Old Mother Hubbard: The Defensive Ends
>> 3.07.2011
The Old Mother Hubbarding of the DTs got a bit of attention; I was even linked by ProFootballTalk. I’ve been a dedicated PFT reader for, like, ever, and secretly I’m over the moon about that—but I’ve decided to act like I’ve been there before. So.
The Pro Football Focus data I’m using is widely misunderstood, so let me explain a couple of key points. First, the PFF data doesn’t measure pure talent. The final grade doesn’t mean that one player is “better” than another, just that they turned in a more effective performance, on a down-by-down basis. Second, PFF weighs penalties in the final grade—so Sammie Hill, who didn’t commit a penalty, graded out higher than Corey Williams, who was more effective, but flagged a ridiculous number of times.
Second, it’s my observation that PFF’s system rewards consistency. A player that repeatedly makes a positive impact (and is rarely a liability) will have a much higher final grade than a player who “swings for the fences,” makes a few spectacular plays, and is otherwise often ineffective . . . even if those few spectacular plays change the outcome of the game. Watching at home on TV, at full speed, we tend not to notice line play unless the results are spectacular—so a player who is always a little more effective than you think (like Sammie Hill) is going to grade out surprisingly well. Meanwhile, the home run hitters (like Ndamukong Suh) are going to grade out surprisingly poorly, since every time you notice them, they’re doing something awesome.
Sometimes, this leads to counterintuitive results. However, a lot of what’s really happening on the field is counterintuitive from a fan perspective. I’ve said before that watching film is the “red pill of fandom;” if you start doing it you learn the truth about what’s happening on the field . . . and you realize just how wrong most of the national football groupthink is. PFF is a group of dedicated football fanatics (like you and I) who spend outrageous amounts of time popping red pills like candy. If you and were committed enough do to what they do, we’d likely get the results they’re getting; that’s why I’m using their data. So, with all that in mind . . . to Wonderland.
Just as before, I’ve included the top-graded 4-3 DE, the Eagles’ Trent Cole, and the bottom-graded 4-3 DE, Kentwan Balmer of the Seahawks. The thick black line represents the average for 4-3 DEs who saw at least 25% of their team’s available snaps.
Just a week ago, I wrote that Cliff Avril was one of the two Lions who had the most to prove in 2010; boy, did he ever prove it. Avril was massively improved over 2009, and finished as PFF’s 11th-best graded 4-3 DT. He was 7th-best in the pure pass rush grading, and 4th-best in coverage. His athleticism, and linebacker pedigree, shows through in those grades. The question mark with Avril is his ability to anchor against the run; he grades out below-average there, but was far from a liability.
Avril was the Lions’ best defensive end, by a long shot, and a true difference-maker for the defense. He saw an average of number of snaps, fewer than most of the other top-ten-type—but that was due to injury, not lack of use. If he was healthy, he was in there. By my calculations, Avril was behind only Ray Edwards and Dwight Freeney in pressures-per-snap—he had a truly outstanding year applying heat to the quarterback. When Kyle Vanden Bosch was hurt, Avril not only didn’t wilt under the pressure, he took the reigns and ran with them, handing in better grades as the season went on. Food for thought: he turns 25 next month.
Interestingly, Avril only had two QB hits all season, far fewer than most with his kind of sack and pressure numbers. My guess is that he’s relying on pure speed to get around right tackles—then either getting there and flushing/sacking the QB, or not. I wonder if pass rushers the QB can see coming don’t get as many hits . . . an interesting project for another time.
Bottom Line: Cliff Avril was pigeonholed by most as a 3-4 ROLB, and I’m certain he could shine in that role. But he’s developed into the fast, athletic 4-3 rush end Rod Marinelli thought he could be. Avril will never be a 270-plus-pound, two-way monster—but he’s already an impact defender, top-flite pass rusher, and a huge part of this defense going forward. Signing this RFA to a long-term deal must be a top priority.
The centerpiece of the Lions’ free-agent additions, Kyle Vanden Bosch had a lot to prove, too: namely, that he could still get to the quarterback like he used to. In his first game, it seemed like he answered those questions in style; I said it was “one of the most amazing individual performances I’ve seen.” KVB poured his heart and soul out into that game, trying to will the Lions to victory. He was sideline-to-sideline, or as much so as a defensive end can be. He put up a truly dominant PFF grade that week, too: a +6.1.
Unfortunately, KVB was inconsistent the rest of the season. He turned in very solid games against the Giants and Redskins, but was negatively, or neutrally, graded in every other game. He was graded well below-average in pass rush; negative in five of eleven games. He was the 46th-best pass rusher out of 65, and had less than half the pressures Avril did—yet, KVB had five sacks and ten quarterback hits in just eleven games. His sack-and-hit per-snap rate was 25th-best in the NFL. This suggests that he was “saving it up” for critical moments; generally just below average except for several plays a game.
However, all of this does an injustice to KVB’s contribution. He was the bell cow for the defensive line, the tone-setter; he pushed every single Lion on the line to practice at full speed, to workout like you practice, to push, to go to the limit and then realize it’s not really the limit and keep going. Avril, Suh, Hill, all the young linemen repeatedly pointed to KVB’s leadership as a major factor in their progression; he was a catalyst, and without him I doubt the others take the strides they did.
I had really good seats to the Jets game, and seeing KVB in action live was something else. He never missed an opportunity—before, during, or after the snap—to remind the Jets that he was there. That he was fighting. That they needed to keep their head on a swivel. That they needed to watch their ass. Vanden Bosch and the Lions out-Jetted the Jets on that day, mostly thanks to KVB and his leadership. PFF graded him at a –2.1 on the day, but I know he had a positive impact.
Bottom Line: Kyle Vanden Bosch is the ultimate leader, a consummate professional—and as a player, the yang to Cliff Avril’s yin. Had he stayed healthy, he would have had seven sacks and fifteen QB hits, more than acceptable standalone production, besides the undeniable halo effect. Unfortunately, he’s 32, and recovering from a major neck injury. The Lions need to find a starting, impact, two-way end to replace him by the 2012 season.
The most surprising line in the chart above is the red one, the one representing Lawrence Jackson. Jackson was the 25th overall pick in 2008, a four-year starter at USC. Lo-Jack is a 6’-4”, 270-pound two-way defensive end, the prototype end for this defense. He was the victim—and the Lions, again, the beneficiary—of “One Man’s Trash” syndrome. The Seahawks let a first-round pick two years into his career go in exchange for a sixth rounder, because he no longer fit their system. To replace Lo-Jack, the Seahawks traded their own sixth-rounder to the 49ers for . . . Kentwan Balmer, picked just a few spots after Jackson in the 2008 draft. Balmer, for the Seahawks this year, was the worst-graded DE in football (see chart).
Jonah Keri just wrote a whole book on how the Tampa Bay Rays went to the world series by making these kind of deals, over and over again. By trading one commodity for another, similar commodity, and getting a 2% edge, over time it really adds up—and at some point, you’ll get a few surprise deals where the edge is way more than 2%. In this case, the Seahawks essentially traded Lo-Jack for Kentwan Balmer. The difference, as you see above, is way more than 2% . . . and where did the benefit go? To the Lions.
Lo-Jack was not a terrifying speed rusher at USC. He had 30.5 sacks in four full seasons: 6, 10, 4, and 10.5, in order. But with size, strength, and attitude sharpened by life in hardscrabble Inglewood, Jackson was a force in both dimensions of the game; in his senior year he paired his 10.5 sacks with 60 tackles—a prototypical Schwartz/Cunningham defensive end.
When injuries hit the defensive line, Jackson answered the bell in a big, big way. Getting a steady diet of snaps from week 10 on, Lo-Jack had a huge (+5.0) game against Buffalo, at left end. He was flipped to the right when KVB went down, and was average against the Bears in his first game on that side. He was flipped back to the left against Green Bay, and had a very strong game, then back to the right against the Bucs and dominated (+4.5). He stayed on the right side the rest of the way out, struggling against Jake Long, but finishing strong against Bryant McKinnie.
Bottom Line: Lo-Jack produced like an above-average starter in heavy rotation, and fron Week 10 on was one of the better 4-3 DEs in the game. I’ll hold off on anointing him the starter of the future for now, because I’d like to see more consistency—but there’s no doubt he’d be the perfect physical fit for the void KVB will eventually leave.
Next, the ever-controversial Turk McBride. Controversial, because I repeatedly dismissed him in my offseason assessments last year, much to the chagrin of the commentariat. This season, the 6’-2”, 278-pound ex-Chief played about 40% fewer snaps than Avril or KVB, but about 40% more than Lo-Jack—exclusively at the left end until Vanden Bosch went down, then exclusively at the right end for the rest of the season.
The stats suggest that McBride was nearly as successful as Jackson; McBride had five sacks to Lo-Jack's eight. The second Green Bay game sheds some light on why they graded out so differently, though. By PFF’s reckoning, both McBride and Jackson had two sacks, yet McBride’s grade for that day was –0.3, and Jackson’s was +2.1. Jackson is clearly having more of an impact down-to-down; Lo-Jack’s run support is much better than McBride’s, as well.
Bottom Line: McBride is an interesting case. Phyiscally, he’s a bit of a ‘tweener, and he did manage to get to the quarterback five times. However, I’ve never liked the cut of his jib for reasons I couldn’t fully explain. The PFF grades show why: when he’s not sacking the quarterback, he’s simply not a factor. PFF’s consistency/home-run bias may be coming into play here, but it supports my eyeball take: Lo-Jack has long-term starter potential here; Turk McBride does not.
There’s nothing that can be said about The Great Willie Young that Neil from Armchair Linebacker hasn’t already said (at breathtakingly profane length!), but I will say this: in only six snaps, he was graded at a +1.5 on the season. He also impressed during the preseason.
Bottom Line: Willie Young is a developmental prospect with a very lean frame, a long way to go, and an undeniable knack for playing football. I hope he has a place on the roster for next season.
We’ve already talked about Andre Fluellen as a defensive tackle. He played 81 snaps there, as opposed to 72 at end, during the Avril/KVB injury phase of weeks 11-15. As a DE, he graded out as a below-average pass rusher, and a well-below-average everything else. I believe his future on this team is as a tackle.
Bottom Line: At 6’-2”, 302, Fluellen is a defensive tackle. He should get this offseason to be developed and coached strictly within that role, rather than being moved all over the line. Instead of frantically gaining and losing weight to fit immediate need, the Lions should develop him this year as strictly a three-technique DT.
SHOPPING LIST: The Lions will need to find an impact two-way defensive end, ready to replace KVB as a starter by the 2012 season. Lawrence Jackson has the potential to be that end. Cliff Avril is an RFA who must be re-signed to a long-term deal. The Lions may look for a developmental speed-first end behind Avril, especially if Willie Young does not take major strides in the offseason.
12 comments:
I'm a big fan of these positional reviews you do Ty so keep it up.
This years draft is loaded with DE talent, are they any guys in particular you'd be interested in with the 13th pick?
So, basically, even though DL was a strong and deep area last year, we could potentially be staring at two holes on our DL in 2012 if LoJack doesn't blossom into a starter (KVB wears down, Avril leaves as FA). I guess that warms me up to drafting a first-round DE.
It also reminds me of the year we drafted CJ when our WRs seemed fine and DL was a big need.
James--
Thanks! I'm trying. :)
Unfortunately, I missed the DL drills at the Combine. I was NOT impressed with Ryan Kerrigan at the Senior Bowl, which is too bad because he's a perfect physical fit. In the second round, if Brooks Reed falls past hometown Arizona, he'd be a perfect "sit for a year or two" prospect with proven pass rush ability. Also, I see Adrian Clayborn sliding past the Lions' second pick in several mocks . . .
Peace
Ty
SomeChoi--
Exactly; if they want to sustain the performance of the defensive line, they'll need to keep it fed with fresh talent. In a year or two, Avril/Lo-Jack/Willie Young may be as effective as KVB/Avril/Lo-Jack was this year, but I'm not sure you can bet the future of the defense on it.
Peace
Ty
The thing to remember about the DL (both Es & Ts) is that Schwartz likes to run a full 8 or 9-man rotation. It isn't your standard start 4, insert a rush specialist in passing situations, and use back-ups as back-ups (i.e. pray you don't have any injuries). So don't look at KVB, Avril, and LoJax as two starters (one aging and coming off serious injury) and a back-up, but rather as 3 guys who will see a lot of snaps (maybe even equal snaps). The "problem" is we need someone to step into that 4th spot 2011 and, by 2012, we'll probably lose one of the guys we've got now. Bottomline: when 1.13 comes around, if the BAP on the Lions' board is a DE that Schwartz & Co. are in love with, I'm fine with that pick.
This goes a little bit back to the "need or luxury pick?" thread from the other day. Who should the Lions take at 1.13 (and later)? It's clear that the Lions should not reach for any one player just to fill a specific positional need, but instead should take the best combination of overall talent and immediate "usefulness." I'm going to assume there's at least one CB and one DE that the Lions really, really like (they would want the player on a team that was starting from scratch). If the CB is gone at 1.13, but the DE is available, I think it's more "useful" to take the end than "reach" for the next best corner (a guy the Lions aren't necessarily in love with) just 'cause corner is the more immediate need.
I will disagree about the idea that Schwartz and Gunny like to run a rotation on the D Line. Suh had about 70 snaps more at DT than the next closest DT in the NFL and Corey was ninth in the NFL in snaps played with just under 900. KVB played a ton of snaps when he was healthy, and the DE rotation did not really get started until first Cliff got hurt, then KVB went down.
They were basically forced into it. I hope that they have seen the benefits of rotating the D Line regularly, and do it more in 2011, giving Lawrence and Sammie Lee the snaps that their level of play deserves.
As to upgrading the D Line, I see Cameron Jordan as an excellent fit for the Lions at DE. The way the draft seems to be shaking out he would be the best player available at # 13 IMO. He could be the heir apparent to KVB. I am assuming that the Willie Young experiment has not played out yet. He will be the speed rusher of the future........I hope.
I am a big Lions fan, but having never actually played football, I am quite naive to the ins and outs and the nuances of each position. With that being said, explain to me why this does or does not make sense for a hypothetical scenario....move Avril to one of the vacant OLB spots the Lions have and use Lawrence Jackson and/or Willie Young to man the vacancy left by Avril....I feel a lot of the negative will be due to Jackson's and Young's youth/inexperience/inconsistency/unproven-ness but I'm interested to hear what others have to say.
Steve, You are not the first to ask that as Cliff was a OLB at Purdue.
1. Cliff is one of the best pass rushing DEs in the NFL. The value of that is very high. Moving him to a "less important" position, that he probably does not have the skill set for does not make sense.
2. I don't think Cliff would be successful at that position. His cover skills would need a lot of work, and this would change what he has been asked to do by the Lions.
3. On a personnel management side, this is not what you do to successful employees. Cliff has worked his ass off to become a DE like the Lions wanted. To ask him to change positions after three years in the NFL would be demotivating in my mind. Not only to Cliff, but to the D Line as a whole.
4. Lawrence looks good, but Willie has not proven anything yet. I see our future [+ 2 years] DE rotation as Cliff, Lawrence, Willie, Turk(?) and another new guy. IMO KVB only has two years maximum of productive play. I hope I am wrong. He would make a great line coach.
NorthLeft12, I agree with most of what you said (lots of good points), but I wanted to take a closer look. Let me start with Suh. He proved pretty much from day 1 that he was a beast that you just don’t take off the field unless you have to. Just because Schwartz likes to rotate doesn’t mean he doesn’t know how to use a player like Suh. Though Haynesworth is a different beast, he was used the same way by Schwartz in Tennessee, i.e as the centerpiece of the defense.
Now let me shift to the “back-ups.” At tackle, Fluellen played all 16 games, starting none and recorded stats in 7 of them (as per nfl.com). His best, based purely on stats, were Weeks 1 and 14-16. Sammie played 15 games (missing Week 4), no starts, and had stats in every one. His best were also Weeks 1 and 14-16. At end, Turk McBride played 15 games, starting 8 (Weeks 2, 10, & 11 for Avril and 13-17 for KVB). He missed Week 6 and was unproductive in 4 games, including the start at Dallas. His best games were starts in Weeks 10 and 13-16. Finally, LoJax only played 11 games, beginning in Week 2. He was unproductive in 2 of them and had his best games in Weeks 10-11 & 14-16. What immediately jumps out is A) both Turk & LoJax did step in very well when Avril & KVB were out and B) all the back-ups had their best games during the Lions win streak (without KVB and missing Avril Week 14).
While I don’t have snap counts, I think these numbers convincingly show that the Lions were actively using all 8 defensive linemen from Week 1 on. Naturally, the back-ups played more when starters went down. That’s just the nature of the game. But the Lions didn’t wait for guys to start getting hurt to get their back-ups in the game and those guys didn’t wait to start making plays. McBride and Hill combined for 3.5 sacks in Weeks 1 & 2. LoJax had one in Week 3, then did, in fact, do most of his damage when Avril or KVB was out (again, this would be expected when your 4th stringer becomes your 3rd stringer). Fluellen’s lone sack was in Week 14. Overall, the 4th stringers (LoJax & Fluellen) were as productive, sack-wise, as the 3rd stringers (McBride & Hill). Again, I think this all shows the Lions were running a deep rotation early on. When injuries DID occur, which was essentially confined to the ends as both Suh & The CW started all 16, McBride & LoJax were both well prepared to step up a spot and Fluellen basically became the 4th stringer at both spots. The rotation became a little less deep, but no less productive because those back-ups had been getting PT all along.
As for your comments on Avril, you’re spot on. You don’t move a premier pass rusher, period. While LBs do blitz, they spend most of their time chasing RBs or dropping into coverage. While Avril is adept in those areas, they aren’t his strengths. It makes no sense to have him spend more time doing what’s he’s less good at. Now, if the Lions were going 3-4, it would make sense to use him in the Clay Mathews/James Harrison role, but that’s not happening. So, the team has no compelling reason to move him. Also, Cliff is not going to want to move for all the reason you mention, plus the fact that elite DEs get paid more than elite OLBs. Moving back would be a demotion AND a pay cut. Finally, your #4, totally agree. I think the only reason Turk started over LoJax was that he had already been in the system. LoJax takes that 3-spot over this season (assuming Avril & KVB are still #1/#2). Willie will continue as a project. Turk is the expendable one, IMO, if the Lions decide to draft an end. I also agree about KVB. A realistic successor needs to be on the roster in 2011, whether that’s LoJax, Willie, or someone else. KVB as DLine coach is A-OK in my book.
Matt,
I used PFF snap count data. Before his injury in week # 12, KVB played just over 91% of the snaps at one DE spot. That is not being rotated.
That is ridiculous. Cliff played roughly 73% of the snaps when he was healthy. That is a good amount.
The idea of a rotation is to keep your best players fresh when the game is on the line in the fourth quarter. I think the Lions did a good job with Cliff, managing his snaps, but did a terrible job with KVB. He should be rested more. Lawrence has certainly earned a higher number of snaps, and I suspect that the Lions defence was better when Lawrence was in compared to KVB.
I understand KVB's role as a defensive leader and think that he has earned his starting spot, but lets not overdo it. I think playing about 70% of the snaps is a reasonable amount.
The DTs were Suh in virtually all the time [just over 90%], and Corey playing about 73% of the snaps. Sammie Lee spelling him for virtually all those snaps [372]. Fluellen got 81 snaps, which was not even what Suh missed. I realize Suh is a great player, but I think they could cut back his snaps by around 10% and he would still be in the top three most used DTs in the NFL. I think in the long run that is better for the Lions and Suh.
Let me try to crunch a few things off the numbers you gave me. If CW's missed snaps (372) were 27% over 16 games, that equals about 1378 snaps in a season or 86 per game. You're saying KVB played 91% in 10 games (through Week 11 and minus the bye), or 783 snaps with only 77 snaps off. That certainly doesn't appear rotational, but you have to factor in some other circumstances.
First, Avril missed three of those games which, as we've already discussed, made the whole rotation thinner (now, in theory, you've got Turk playing 73% of snaps with LoJax filling in the other 27% for him AND the 9% for KVB - if Avril's healthy, do you still need KVB in for 91% or does it dip into the 80s?).
Second, I'm assuming KVB has enough veteran clout, especially with this coaching staff, that he mostly takes himself in and out of games (as opposed to the coaching staff). If you give a guy that kind of freedom, you have to live with the results. Going hand-in-hand with this is that KVB likely knows the system better than, literally, any other player on the roster. That means he's in every single package. A gassed guy who knows what he's doing is better than a fresh guy that's still learning his responsibilities (i.e. LoJax).
Third, I think you MUST compare these numbers (Suh & KVB) to other STARTING DTs & DEs in the NFL, not to the other players on the Lions' roster and not to NFL-wide positional averages. I'm guessing (again, don't have the data myself) that, say, Kevin Williams and Jared Allen in Minnesota both played in the neighborhood of 90% of the Vikings' snaps. Again, rotation is great, but you don't take a dominant player off the field any more than you absolutely have to.
Fourth, remember who KVB was lining up with: a rookie, a guy brand new to the team, and Avril (a young, oft-injured veteran). You gotta' have a consistent veteran presence out there to make sure those other guys are lining up properly and know their responsibilities. Hypothetically, if you had the same KVB, but fast-forwarded the other guys' careers 3 years down the road, you'd be more comfortable whenever KVB did come to the sideline.
Finally, KVB's numbers are, obviously, from just over the first half of the season. We have no idea how the rotation might have changed down the stretch.
My overall point wasn't that the Lions use 8 guys equally, but rather that they use(d) all 8 significantly and this made for a better overall DL over the course of the entire season. Avril, Turk, LoJax, Hill, CW, and Flue all recorded half their sacks (give or take about half a sack) after KVB went down. I gotta' think they were able to increase overall DL production on a per game basis without KVB because those same guys got opportunities to be productive (the other half of those sacks came with KVB starting) in relief during those first 10 games.
I suppose to really prove my point, I need to get a PFF membership and do an appropriately comparative analysis of the Lions players to similar players league-wide. That ain't happening any time soon, so, for now, I'll have to stick with what I've got: limited stats, limited observation, and reputation (Schwartz's, not mine :-).
Just thought my rotation theory deserved a bump considering who the Lions drafted tonight.
Post a Comment