I'm not sure how I happened on Derek Sarley -- may have been a smart football RT -- but I'm glad that I did. Some of the team bloggers we interviewed recently were a tad too "fansy" for our liking, so I'm glad to get in Derek who's about as hard core a film junkie as you're gonna read. He can be read at The Philadelphia Daily News & igglesblog.com as well as followed at www.twitter.com/igglesblog
We should get his take on the Chip Kelly offense and some NFL trends -- but, first, Derek, was Silver Linings Playbook a really good movie or is Jennifer Lawrence simply delightful?
Thanks, Zach. I went into that movie expecting it to be just another dumb romantic comedy that would pass a Friday night after we'd finally gotten the kids to bed. It ended up having an emotional depth for which I wasn't prepared. I'd say it was "really good," but then I'd risk enraging a friend of mine (www.twitter.com/t_mcallister) who wrote the *second* most popular recent book about being an emotionally turbulent Eagles fan. So, never mind. It was all Jennifer Lawrence.
The market for enraged Eagles fans must be robust!
There's a lot of superficial talk about the Chip Kelly offense. Can you give us a sense of how and when the Eagles offense is going to look different from the football we're used to watching on NFL Sundays?
Depends on the time of year and who's out of which playoff race.
It's worth separating Chip Kelly into two distinct roles. The first is Chip Kelly, head coach. That's the guy who has implemented state-of-the-art workout and nutrition programs, structures practice to maximize physical reps, has a roster bias towards size, etc. That's the "program builder" the Eagles said they were hiring back in the spring and his success or failure in that role will almost totally determine the outcome of his stint in the professional ranks.
With that said, everyone always wants to talk about Chip Kelly, offensive genius -- and for good reason, since that's the guy we get to watch on TV.
Based on what we've seen so far, Chip's Philadelphia offense is going to look a lot like his Oregon offense. If the defense is light in the box, he'll run the ball and use the QB read to "block" a defender. If it's clogged up inside, he'll attack the edges with outside runs and lots of different kinds of screens. And if the defense is selling out up front, he'll *try* to attack them with the deep passing game.
I say "try" because not even the great Chip Kelly will be able to run a deep passing game if his offensive line doesn't hold up better than it did last week against Jacksonville. And the loss of Jeremy Maclin means the wide receiver corps isn't as dangerous as it was before his injury.
If all of that sounds a lot like the kind of football we're used to watching on Sundays, that's because it is. He's centralizing a bunch of cool new stuff we've started to see around the league the last few years, but the amount that's totally new will probably be a small percentage.
The last thing on that point is that there's really an embarrassment of riches in terms of material on Chip Kelly's offense. Between FishDuck.com, Chris Brown on Grantland, and Sheil Kapadia (www.phillymag.com/eagles/) it's harder to find new angles to write about than anything else.
You, and others like Chris Brown have carved out an interesting niche of writers who talk about "smart football," but the emphasis is more on scheme nuance and rigorous film breakdown as opposed to numbers and advanced stats ala Football Outsiders, PFF etc.
Which, if any of the new advanced stats do you find helpful?
Sorry for the delay, guys, had a kids soccer practice to coach.
So, Riley Cooper. No, no one in Philadelphia is over what he did. That's a stain forever and only an honest-to-goodness troll will be wearing a #14 jersey to the Linc this year.
That said, we made our peace with the dog-killing thing and he's the best of a bunch of not-great options, so if they're not going to cut him, we'll have to root for him-as-player if not him-as-person.
I'm really not sure what to make of Ertz. You wouldn't think a 4-12 team would have that many positions where rookies couldn't come in and contribute, but they actually are pretty stocked at tight end. Just in terms of receiving yards, Brent Celek has been a top ten TE three of the last four years. He's struggled a bit with a lingering hand issue that has affected his catching, but he's a solid blocker and better after the catch than most people realize.
Behind him, they have Rice grad (and multi-position Swiss army knife) James Casey, Ertz and even Clay Harbor, who might be a long shot to make the final 53, but has been working out with the WRs as well the past few weeks. We've even seen packages with all four TEs on the field at once (http://www.phillymag.com/ea...).
That's all a long way of saying Ertz is clearly the TE of the future, but it's a bit of a logjam at present.
Derek, if you had to, what would you set the over/under for games started by Vick this season? I'm thinking not just of injury concerns (he's gained 4 pounds, so now he's invincible!) but also the possibility that Kelly turns to Foles, especially if the Iggles start 2-5 or something like that.
So, first off, there aren't really "others like Chris Brown." Not that there aren't other people out there with similar levels of knowledge, but no one who knows as much as he does writes as much as he does or as well for laypersons as he does. He's one of a kind.
What makes that question kind of funny, though, is that before I "retired" I was known as more of a stats guy. I've been in more arguments than I can count about "numbers always lying because [you said something mean about my favorite player]."
I'd add AdvancedNFLStats.com to your list of sites above. And I wouldn't put PFF in with the others. I really appreciate the way they track snaps played / in coverage / in the slot, etc., but their "objective" rating system is deeply, deeply flawed. Which wouldn't be so bad if their position was, pace Winston Churchill, that the system was the worst we have, except for all the alternatives, rather than "totally amazing and perfectly correct because we watch every play of every game."
I think Vick will play until he gets hurt, which averaging the last three years together means about week eight.
Vick is clearly the _type_ of player who fits Kelly's system -- decision-making cliches aside -- but he's probably too old to _be_ the player to fit Kelly's system. So they'll ride him as long and as hard as they can so everyone else can get used to playing that kind of offense, then go to Foles when he breaks.
Or, you know, Invincible 2 will happen and we'll all just enjoy the ride.
Hopefully Elizabeth Banks will be involved. I'm still trying to figure out why whoever directed that movie instructed Elizabeth Banks to go with a Texas accent when she was playing the role of a New Yorker. It's as baffling as some Andy Reid 3rd and short play-calls.
Let me go back to PFF for a moment -- because they've obviously gained quite a bit of traction over the last year or so -- (attitude aside) what, in particular, do you find flawed about their system?
PFF collects information that is absolutely invaluable and there's no other public source for it. We're not getting player snap breakdowns (covering, rushing or playing run defense) anywhere else. It's hugely useful to know how many snaps cornerbacks played in the slot last year. QB under pressure numbers are good too. I rely on PFF all the time for these sorts of things.
Some of the concerns I have with PFF are not limited to them. Scott Kacsmar posted something on Football Outsiders this week that spotlighted some of the issues we see in all advanced stats (http://www.footballoutsider...). Any time we're using stats to compare two secondary guys, we need to understand that all of that information was gathered by people who may not have known or cared what the coverage was supposed to be and so just assigned blame to the nearest guy.
Again, this isn't a PFF-specific issue, but if their 'net sweepers are out and they're listening, I'd encourage them to pull up the first Eagles preseason game and look at their grading for pass plays involving Patrick Chung.
PFF has a systemic issue. Football isn't baseball. You can't isolate one batter against one pitcher and get a pretty good idea of the skills of both players.
PFF would say you can. That you can focus on the LG vs the RDT, see who gets the better of the match-up, and then assign an "objective" grade to the outcome.
Coaches aren't stupid. They don't, generally, ask players to do things they can't do. So you may find that both LGs involved in a game blew four run blocks and two pass protections, so they get the same grade. The problem is that one of those LGs is an all-pro and was being asked to make difficult reach blocks all game, while the other is a stump who just slowed down the guy in front of him.
The same thing is true at other positions. Good RTs are going to be left to block alone more often than bad ones. Great quarterbacks will be asked to make more difficult throws. TEs who are abysmal at blocking won't do it very often.
By atomizing performance we're ignoring the way 11 players on a side function as a system that gets continually re-balanced to achieve some sort of equilibrium.
Another example: Playing defense gets easier and easier the better your teammates are. Take one MIKE who's playing behind two dominant DTs. That guy's going to look like an all-pro as he's running around clean knocking heads. Put that same MIKE on another team with a Swiss cheese line and now his job will be much harder. Same for CBs playing Ss and, really, the whole back seven playing with a good or bad pass rush.
"WE ARE NOT SCOUTS — We aren’t looking for (or grading) style or technique, merely the result of the play ... We are looking for the result of that poor technique, not the poor technique itself. If poor technique results in a positive play, that is graded at the same level as good technique yielding a positive play."
I'd criticize that passage, but then I'd just be re-stating what they said.
Now, to my point several posts ago about best/worst system. There's still value in what they're doing. The grading mostly seems like it's directionally correct most of the time. They identify interesting players who finally get a bit of attention.
But when you look at what actually happens after the season ends -- which free agents get signed, how much they're paid, why are all these PFF superstars still looking for work -- there's clearly a misalignment between their grades and what NFL Pro Personnel guys are seeing.
In that sort of dispute, between a bunch of passionate amateurs and a league full of hardened professionals playing a zero-sum game, I'll go with the latter group every time.
Was it just me or was that article kind of a joke? It was as if Shane Ryan wrote an article about the statistical analysis of Hawk Harrelson. (Sorry Shane, but this was terrible: http://www.grantland.com/bl...)
No disrespect to Tony Khan, but if he's relying on quarterback rating to assess Blaine Gabbert, he doesn't know shit about advanced stats. And if he considers the difference between an 82.8 QB rating and 84.5 QB rating (in a small sample size, no less) to represent a "jump", he doesn't know shit about statistics in general.
Teams have been using advanced analytics for years. One of the most striking things that comes with being an Eagles blogger is when someone does research showing what teams "should" do in a particular offensive situation and then you go back and look at how the Eagles operated in the same situation under Andy Reid and find the curves are mysteriously aligned.
Football Outsiders has been doing contracted research for NFL teams for a long time.
In the case of the Jaguars, I'd note the qualifications of the person leading up their statistical effort, which have more to do with genes than educational background. I guarantee the people doing the most interesting stats work for teams have PhDs, are much smarter than the rest of us, and won't be going public any time soon.
Ah see, Alex beat me to it. The platform should auto-update new comments :)
Someone asked for an example. I'll pick Quintin Mikell, a guy near and dear to the hearts of Philly fans.
Last year, PFF graded him as the league's fifth-best safety. This summer, he couldn't find a job. And the Eagles' safety situation is so bad that they went and signed Kenny Phillips, who can't stay healthy in the bubble boy's cocoon.
Derek, you blog Eagles first and foremost. What other teams around the league intrigue you? I know for me, I'm a Texans fan, but I love the way Seattle has been doing things for a while and love Bruce Arians run offenses.
As my kids have gotten older, it's been harder and harder to keep up with the rest of the league.
Like lots of fans, I'm curious about the second-year quarterbacks. Are they really as good as they looked last year, or was it a right time/place thing for most of them, as the league wasn't ready for those "college offenses" and at least the guys out West stepped into some pretty ideal situations.
I'm a big RGIII fan. I wish he'd gone anywhere but Washington.
I really don't. I just like watching football. Match up any two good NFL teams and I'm in.
College football is different. Bad spread football is the worst football, and there are lots of college teams playing bad spread football. I'd rather watch two bad Big Ten teams slam it up the gut for three yards a pop all day.
I do like Kubiak's offense. When it's working, it reminds of when Norv Turner and Mike Shanahan were at their peaks as playcallers.
Derek, you've been a fantastic guest (btw, Derek writes at igglesblog.com and is @igglesblog on Twitter) one last question for you --
I haven't done any studies on this, but it would seem to me that it's better for an offense to have 2nd and 1 than it is 1st and 10. 2nd and 1 leads to 1st and 10 way more than 1st and 10 does. And there are a LOT of big plays that come off 1st and 10. Hell, as a defense I think I'd rather the other team's offense have a clean first and 10. When you're defending against 2nd and 1, you're thinking, oh just get it over with and get the 1st down already.
Now, if that's true, wouldn't it be a smart move for an offense to try to get to 2nd and 1 or even 2nd and 2 more often? And then I saw a coach that I believe uses that strategy. Almost all of New England's first down plays look to gain about 8 yards.
Is this theory too "out-there" or should teams being making concentrated efforts to get 2nd and very short?
I'm not sure about the second part. Just for fun, I used the PFR play finder (http://www.pro-football-ref...) to see how many times New England threw a pass that gained 8-9 or 10-11 yards on first and ten. The database spits out 19 of the former and 15 of the latter, so if they're consciously trying to throw it just short of the sticks, they're having a limited amount of success.
My guess is that all has more to do with play timing and how the offense fits together, but it's certainly worth checking out more.