Breaking Down PFP

^{- Posted by Derek -}

The guys from Football Outsiders have changed the name, but it's still the same great book.  And since there's a good chance this is a temporary change -- what with them maybe even having a publisher and all next year -- I'm sticking with the old name.

I'm going to do this post a little differently than in years past.  Typically, I've gone through the entries for the Eagles and their NFCE brethren, picked out the juiciest morsels for sharing, and then riffed from there.

I'm mixing that up a bit this year, for two reasons.  The first is that the book is now a digital download.  And while I sort of miss having the big ugly volume to flip through, you can't really use the "Amazon hasn't shipped my order excuse" any more. 

The second is that I think they got a number of things wrong this year.  And I'm going to save my precious excerpting allowances to explain why.

- - - - - -

Starting point:  They don't love the 2009 Eagles, predicting only 9.3 wins and assigning the team a 27 percent chance of being a "Super Bowl contender" (11+ wins).  One reason?  Injuries:

One of our biggest concerns about the Eagles in 2009, in fact, is ... not a single member of the rotation at either defensive line or linebacker missed a single game a year ago, something that’s extremely unlikely to recur. That health even extended to the secondary, where the four starters combined for a total of 63 starts, with only Asante Samuel missing even a single game due to injury. That’s also not likely to recur in 2009.

In response, I would argue that the Eagles are extremely well positioned to handle ordinary injuries on the defensive side of the ball, due to the terrific depth they've assembled (major injuries, particularly to a guy like Trent Cole, are going to be tough for any team, but there's no reason to assume the Eagles are any more likely to suffer such a fate than any other team). 

In PFP's own words:

No team in the league has more quality or depth at corner, one through four, than Philly.

So that handles corners.  At linebacker, if Bradley goes down, that's going to hurt, but they seem to have replacement-level backups at both outside positions.  At tackle, Trevor Laws played well in limited action last year, and although you'd worry about weardown if one of the big guys blows a tire, he'd likely be fine as a short-term replacement against anyone but the Giants.  Defensive end is extremely deep and while no one's replacing Cole, there are plenty of able bodies.

The one position you'd worry about -- from an unproven back-up perspective -- is safety.  If Jones beats out Demps, we still won't really know how Quintin will do out there.  If Demps wins, then Jones is perfectly capable of starting.

So yes, injuries are always of concern.  Please know, dear football gods, that I am not mocking you.  We always worry about injuries.  But if a guy goes down, the Eagles' defense seems like it has ample reinforcements.

- - - - - -

Fun fact #1:  The Eagles ran in second-and-long situations only 37 percent of the time last year, the 19th-highest rate in the league.  So much for the theory that every drive has to start:  "Incomplete pass, Run up the middle."

- - - - - -

The FO guys also aren't all that high on Todd Herremans:

Herremans will compete with Max Jean-Gilles for the second season in a row for the left guard spot ... Had Jean-Gilles not been filling in for the injured Andrews at right guard, he very well might have lost his spot for good ... If it were up to us, we’d go with ... Jean-Gilles at left guard ...

As we discussed a week ago, this observation does not hold up to close scrutiny.  Herremans was very good last year, MJG was uneven, and now that we have the numbers to back this up, I feel pretty confident in knocking back the MJG train.

- - - - - -

Fun fact #2:  In 2008, the Eagles ran "max protect" 10 percent of the time.  Going back to 2007 and 2006, the figures were 13 percent and 8 percent, respectively.

- - - - - -

Looking at the rest of the line, they have this to say:

Furthermore, our research indicates that the best offensive lines are the ones that experience continuity, so while Philadelphia has added superior talent with their free agent acquisitions, it may take a season for that talent to gel into a superior line.

The last bit is certainly true.  It might take some time for these guys to figure each other out, especially with all the injury/rehab issues at play.  With that said, I remember their essay on offensive line continuity, and I also remember not finding it all that persuasive at the time. 

To re-hash, for the 2007 PFP, they ran a correlation analysis that found offensive line continuity was correlated with good things, like winning and offensive DVOA.  The directionality question was almost completely unaddressed (you don't make changes to offensive lines that are playing well, and injuries suck).  And I'll quote this part:

Continuity does not measure whether the new lineman is a top pick or a star free agent or some guy off the practice squad. 

Exactly.

- - - - - -

Fun fact #3:  "On offense, the Eagles had the league’s biggest difference in rushing DVOA depending on how many running backs were in the formation. With one back, they averaged 5.0 yards per carry with 29.1% DVOA; with two backs, they averaged 3.2 yards per carry with -13.1% DVOA."  Yep.

- - - - - -

The FO guys are very high on Bunkley and Patterson:

The ascension of the defensive tackle combination of Mike Patterson and Brodrick Bunkley from promising to elite came over the second half of the season ... They’re not dramatic, penetrating one-gap tackles who accrue tons of defeats, but simply a pair of players who teams struggle to account for on every play.

As we've discussed before, I think those two guys are very, very solid, but I'm not ready to give them elite status yet.  (And you'll kindly note how closely that description above mirrors my prediction for Patterson from before the 2007 season, when he was still getting a lot of crap.)

Speaking of previous discussions, let's take another look at that sack post from May.  This is from the comments:

Let's revisit the numbers question when PFP 2009 comes out (in whatever form they're going to end up doing). When that happens, we'll get solid numbers on things like QB hurries, which will give us a more complete picture of their pressure.

Well we've got the numbers ... and they ain't pretty.  Patterson can add three hurries to his 0.5 sacks.  Bunkley had one hurry, in addition to his two sacks.  Talk about roles all you'd like, but that's pretty much case closed on how much pressure those guys got.  (Unless Joyner can bail them out ...)

With that said, those aren't remotely the scariest numbers in that table.  This one is:

Player .......... Sack .. Hurry
Darren Howard ... 10.0 .... 4

That is, quite simply, an unsustainable ratio.  I'll have to do some more data collection to put that in better context, but what it means is that Howard was converting an insanely high percentage of his pressure opportunities into actual sacks.  That's a bit like a baseball player having an unreasonably high batting average on balls in play.  You're going to get regression to the mean the next year.

That's one more reason to worry about the interior pass rush.  At this point, if I'm on the Eagles defensive coaching staff, I'm drawing up lots of ways to get second-level guys free coming up the middle.

- - - - - -

Fun fact #4:  "Asante Samuel broke up 31.2 percent of the passes in his direction, a percentage surpassed only by Corey Webster amongst starting corners."

- - - - - -

Here's a head-scratcher:

Middle linebacker Stewart Bradley was about what we expected last year; he’s effective against the run and a sure tackler, but he still has a ways to go in coverage. Strongside backer Chris Gocong is probably the unit’s best player ...

The most charitable explanation is that it's doubtful FO's advanced stats have yet figured out how to account for plays where the linebacker misreads the action and gets washed out of the play.  But it figures the smart guy would do well on the smart guy stats, however.

With that said, no, he isn't.

- - - - - -

Fun fact #5:  Asante Samuel really has no interest in tackling running backs, thank you.

- - - - - -

Quick hits:

LeSean McCoy -- "Well, the bad news is that he can’t pass block and he has serious fumble issues ..."

Brian Westbrook -- "Simply put, there’s not enough cartilage left in Westbrook’s knee ... About the only option the Eagles would have left would be Synvisc injections, the same treatment Randy Johnson uses to keep his knees lubricated. That’s Westbrook’s last resort, and honestly, it’s not that far away from coming into play."

Kevin Curtis -- "[A] player like Curtis, who operated out of the slot and didn’t play on every down, was less likely to get injured in his part-time role than as a primary receiver for the Iggles."

DeSean Jackson -- "One thing Jackson needs to work on is disguising his cuts and his routes; just under 16 percent of the passes thrown to him were defensed, far worse than the 4.5 percent averaged by the rest of the team."

[Ed. note: Or is he just one of the few guys McNabb has trusted over his career to actually make a play on the ball?]

Maclin -- "Expect Maclin to be a part-time receiver and return specialist this year before emerging as the Eagles’ best receiver by 2011, assuming that he can stay healthy."  More on Maclin in a bit.

McNabb -- "He’s accurate enough to play quarterback in a West Coast offense, but his erratic mechanics yield stretches of extreme wildness (he led the league with 14 incompletes where our game charters marked the defender as 'Hole in Zone')."

[Ed. note: That's just funny.]

- - - - - -

Fun fact #6:  Buy the book.  "Turn" to page 507.  Check out the chart at the bottom.  Seems awfully definitive.

- - - - - -

The last thing I feel I should at least mention -- Sam pointed it out -- is the "Playmaker Score" analysis in the essays at the end of the book.  They create a new wide receiver metric for college production (TD / G * YPC) and find that higher numbers are better predictors of NFL success.

Maclin's college production gives him a 10.0, which puts him in the third tier, not a good place to be. 

I would note in response that DeSean Jackson's "Playmaker Score" appears, if I'm doing the calculations correctly, to be 9.4. 

He seems to be working out ok.