Mining the New Football Outsiders Almanac 2012

The Football Outsiders Almanac appeared online yesterday, and you should obviously go get yourself a copy. With apologizes to our esteemed local publication, the FOA 2012 is the gold standard for the NFL offseason. The amount of statistical detail Aaron Schatz and everyone else at Football Outsiders puts into their work is nothing short of awe-inspiring. With that in mind, I’m going to highlight a few pieces that stuck out to me.

* * *

The Almanac jokes that, “of course we’re predicting a Philadelphia rebound,” but I don’t actually see it. They give a mean projection of 8.6 wins in 2012, which is barely more than the Eagles amassed last year. Moreover, it’s the lowest projected win total going back to at least 2009. The 2011 optimistic outlook pegged them at 11.7 wins. Oops.

* * *

FO marked Demetress Bell down for only five blown blocks in his last 20 starts. If he can stay healthy and Howard Mudd can work some magic, maybe there’s reason for some optimism at left tackle after all.

* * *

The Eagles dropped from third to eighth in offensive DVOA, but the Almanac suggests that “half” of that decline came from Vince Young’s poor play. Let’s hope Mike Kafka proves to be a better backup.

* * *

Two running backs and one tight end was the Eagles’ third most common offensive formation, but the Eagles ran the ball from it only one third of the time — the lowest percentage in the league by a long shot. On the other hand, this is the first year since 2009 that the offense ranked higher than 23rd in overall run percentage. The mantra appears to have been, “run, just not behind Owen Schmitt.”

* * *

The Almanac attributes only 12 sacks to blown blocks, the lowest figure in the league. Moreover, three of those are in LeSean McCoy’s column.

* * *

Despite those 18 sacks, I wonder if Jason Babin might be playing himself into a platoon job at left defensive end. Runs to his side averaged 4.91 adjusted line yards, second-worst in the NFL. On the other side, Trent Cole was second-best in the league, allowing a paltry 2.4 yards.

* * *

For the second year in a row, the Eagles defense ranked near the tail end of the league defending running backs coming out of the backfield. This is what’s colloquially known as the Casey-Matthews-covering-Brandon-Jacobs problem.

* * *

Bobby April is universally hailed as a special teams maestro, but his unit has declined in DVOA each year since he arrived. This year’s biggest problems came in the form of kick returners (Dion Lewis plus a down year for DeSean) and Chas Henry, who FO estimates cost the Eagles 11.5 points over the course of the season in field position alone. Yikes.

* * *

The Almanac is very optimistic about Michael Vick’s chances to rebound all the way back to his 2010 form. In fact, despite assuring us that his interception rate jump from 1.6 percent to 3.3 percent in 2011 was a normal regression to the mean, FO predicts he’ll go back to a 1.9 percent rate this season. Among starters, that would put him among the top five in the league.

* * *

There’s tons more where this came from, so go buy the book and share what stands out to you.

Low Expectations for Demetress Bell

I’m a big believer in the market economics of the NFL. If 32 NFL teams pass on a player until the sixth round, he doesn’t have a particularly high chance of success. It’s the same with free agents who don’t receive serious interest on the market.

Derek Landri, who re-signed with the Eagles yesterday, would be in the latter category. He was a productive back-up with the team last year, but even after perhaps his best season as a pro, no other team offered him a long term deal — and the Eagles weren’t anxious to get him back either. He seems to still have an uphill battle to make the roster, especially if the team jumps in with a first round defensive tackle.

The same thought process leads me to question Demetress Bell’s value as well. Bell shopped himself around quite a lot, visiting a handful of different cities in order to seek a long term deal. But he was never offered one, even by the suddenly desperate Eagles.

Technically, Bell’s contract is 5 years, $35 million, but everyone knows he’s not coming back after 2012. The $8.5 million roster bonus in 2013 makes that a foregone conclusion. What does it say about Bell that in the modern NFL where left tackles are one of the top two or three most important positions on the field, he couldn’t find one team to give him a legitimate multi-year deal? Is he really any better than King Dunlap?

On the other hand, he was undoubtedly the best player the Eagles could get when they learned that Jason Peters was lost for the year. That has to count for something. And over the last two seasons, he’s been as good or better than Todd Herremans according to Pro Football Focus’s pass blocking efficiency statistic:

Maybe Bell will play up to his potential this season, or even exceed his past performance now that he has Howard Mudd as a guru. Maybe he’ll manage to stay healthy the whole year. Or maybe he’ll give the Eagles exactly what they paid for, a questionable veteran on a relatively meager deal.

Impossible to say for sure, but I’m not really looking forward to finding out the answer.

Photo from Getty.

Watch and Judge All 24 Sacks of Michael Vick

Jimmy Kempski does an amazing job in this post, culling the video for all sacks Michael Vick took in 2011. More than anything, the Eagles quarterback doesn’t come out looking too good. Many of the sacks are poor decisions by Vick, and a number of the others look suspiciously like poor blitz recognition and protection adjustments at the line of scrimmage.

The one thing that is worth keeping in mind is these are only 24 plays, and by definition are likely to show the worst mistakes. There isn’t a companion video that shows how many times Vick escaped pressure in his face and turned it into a positive gain.

I’m also not sure I follow Jimmy’s logic at the end, where he notes poor right tackle play from Winston Justice in 2010 as a reason to keep Todd Herremans on that side. But all in all, really informative and interesting.

Inside Trent Cole's Contract Extension

Brian McIntyre has the details of Trent Cole’s contract. Here is the basic breakdown:

2012: $8 million signing bonus, plus original $3 million salary now guaranteed.
2013: Original $3.5 million salary now guaranteed.
2014: New $5 million salary (plus $500,000 $ack$-based bonus).
2015: New $10 million salary.
2016: New $11 million salary.
2017: New $14 million salary.

Cole turns 30 this year, which should give you a sense of which years are more or less fake money. To my eyes, the last three years all look unlikely. The Eagles gave Cole a lot of security by guaranteeing the final two years of his old contract and handing him a signing bonus on top of it. The 2014 additional year also looks attainable and very reasonably priced. After that, I don’t see the Eagles paying $10 million or more per year for a 33-year-old and up.

Still, don’t let that detract from what the deal really means. It’s not intended to purchase many more years of performance, but rather to serve as a thank you gift, a reward for Cole’s quiet excellence over the last few years. And in many ways, that’s more important.

Pass Blocking Efficiency, from Peters to Kelce

Considering how much fun we had the last time we talked about offensive line play, we might as well go back to it.

Pro Football Focus calculates “Pass Blocking Efficiency” as a percentage of plays in which offensive linemen prevent quarterback pressure. Sacks are weighted more heavily than hits and hurries, but they all count.

Here are the numbers for the Eagles starting linemen last season:

I listed the players in order of their Pass Blocking Efficiency, but truthfully that doesn’t quite give you the right context. 95 percent might sound good, but you have to prevent the quarterback from pressure the vast majority of the time if you even want to see playing time as an offensive lineman.

For a more accurate picture, look at the ranks on the left side. Those numbers are where the player stands among all those at his position (tackle, guard, or center) who played at least 50 percent of his team’s snaps. Therefore, even though Evan Mathis has the highest raw score, Jason Peters was ranked higher among his peers. He scored 5th-best among tackles, where Mathis was 11th among guards. Presumably that means that Peters’ job is more difficult.

Todd Herremans ranked 35th with 94 percent PBE, but that’s not as bad as it might sound. 58 tackles qualified with more than half of their team’s snaps, so he is below average among all of them. On the other hand, he still scored better than the majority of “second” tackles, so I suppose there’s a positive way to look at his performance. Danny Watkins, at 43rd out of 55, looks worse than Herremans on this measurement.

It was Jason Kelce that really stood out to me, though. At 30th overall, he was the single worst center measured. I was critical of Brian Baldinger’s comments that Kelce was having a Pro Bowl year, but that was based solely on my own impressions. These numbers certainly back me up though. In fact, Kelce’s 33 total pressures allowed was 10 more than the next highest center.

Assuming Mathis returns to Philly, it looks like the left side of the line will be more than fine. But everyone from Kelce on down the right side definitely has room for improvement in 2012.

Photo from Getty.

On the Inaccuracy of SackSEER

Jimmy Kempski:

I think just about any knowledgeable football fan that is familiar with Football Outsiders is generally very complimentary of the work they do (as they should be), but SackSEER has its fans and detractors. The model was formed from data collected prior to the 2010 season and was first introduced in 2010. Therefore, the the model pre-2010 is understandably extremely successful. Its fans point to the pre-2010 success. Its detractors feel that predicting a player’s future in the NFL based on a couple jumps and a few passes through some cones at the Combine is a little ridiculous. It also is a pure data model that ignores things like watching game tape.

Jimmy rightly points out the utter failure this model has been in the last two years, although I’d make two additional points. First, in defense of SackSEER, don’t read the projections as predictions, but rather the most likely general range based on past precedent. For example, to knock the model for projecting “only” 36.4 sacks in his first five years for Von Miller is silly. It argued that he would be good and he is. Jerry Hughes, on the other hand…

That said, any model that leans too heavily on something as luck-dominated as sack totals is going to have trouble. I wonder if a better SackSEER could be constructed based on total pressure (sacks, hits, hurries).

Michael Vick: Under Pressure

I don’t think anyone (with the possible exception of Andy Reid) was shocked when Juan Castillo didn’t work out on defense. But nearly everyone was surprised at just how far Michael Vick regressed from his star season in 2010.

The big problem Vick had was simply turnovers. After posting zero interceptions through his first eight games the previous season, Vick was picked off 14 times in 13 games in 2011. That’s not nearly good enough if the Eagles want to rebound and make a run into the playoffs next year.

So let’s go a little deeper to try to understand some of Vick’s interception woes, using some stats from Pro Football Focus. Today I want to split Vick’s plays into two categories: regular and under pressure.

Let’s look at his non-pressured stats first. When not threatened by sacks or hits, Vick completed 69 percent of his passes, good enough for 8th-best among the 24 quarterbacks who took at least 50 percent of their team’s offensive snaps. That completion rate jumps to 76 percent and 6th-best when you don’t count drops against him.

As to interceptions, Vick was slightly below average, but not by much. He was 15th out of 24, with an interception rate of 2.5 percent. For reference, Eli Manning was 14th with 2.4 percent, and Drew Brees was 11th with 2.1 percent.

Overall, Vick showed room for improvement, but no big problems when he wasn’t pressured. When he had defenders in his face, however, Vick’s performance was more of a mixed bag.

On one hand, he had the 5th-highest touchdown rate and the 2nd-lowest sack rate among those 24 quarterbacks. With only 11.6 percent of all pressured dropbacks turned into sacks, Vick’s famed elusiveness served him well avoiding a big loss.

However, Vick completed only 42 percent his passes under pressure, which ranks 18th. Worse, 4.9 percent of his passes were intercepted, beating only Matt Hasselbeck, Tarvaris Jackson, Rex Grossman, and Ryan Fitzpatrick. (Matt Ryan scored about average, at 3.5 percent, while Andy Dalton, Aaron Rodgers, and Sam Bradford were blemish-free.)

There are two takeaways from this information. One is relatively straightforward: Vick needs to handle the pressure better, even take a few more sacks rather than expose himself to interceptions. The rookie Dalton, who threw zero interceptions under pressure, also threw the ball away in those situations more than anyone else in the league. Vick could learn something from that example.

But the second takeaway is more nuanced. For the last two seasons, Vick has been under pressure more than nearly any other quarterback. Last year he ranked first overall with pressure in 39.8 percent of his dropbacks. In 2010 he was second, with 41.8 percent.

Given the improvement along the offensive line year-over-year, it’s likely that this has more to do with Vick than his blockers. Football Outsiders sack timing stats show that more than half of his sacks take longer than three seconds, which partially can be attributed to avoiding defenders, but also results from his tendency to hold on to the ball too long.

Vick is a playmaker when he scrambles around — his touchdown rate is actually higher with pressure than without. But avoiding sacks and trying to score big also led him to turn the ball over far too much.

Perhaps Reid and Marty Mornhinweg should focus on teaching Vick to avoid more of those situations by making pre-snap reads and quick decisions about where to go with the football. If DeSean Jackson returns, there will still be plenty of opportunities to create big plays, even without dancing around in backfield. If he doesn’t become more consistent and less turnover-prone, Vick will continue to be a liablity going forward.

Photo from Getty.

A Second Level Look at the Eagles Defense

10th in points allowed, 8th in yards allowed.

Those numbers are seductive. Maybe this Eagles defense wasn’t as bad as we thought. Maybe, counter to conventional wisdom, Juan Castillo doesn’t have to be fired. Maybe he’s already figured things out.

If you feel those sentiments catching up to you, it’s time for an intervention. Total points allowed, total yards allowed are flawed statistics at best. Imparting any meaning to them requires understanding them in the context of a host of other statistical factors.

When we take the whole picture into account, the picture of the Eagles defense doesn’t look so rosy after all.

Opponents

Here’s a list of the Eagles opponents, ranked by offensive efficiency (Football Outsiders DVOA):

• NWE #3
• NYG #7
• DAL #11
• ATL #12
• BUF #16
• SF #18
• WAS #19
• MIA #20
• NYJ #21
• SEA #22
• ARI #28
• CHI #30
• STL #32

The Eagles played one elite offense this season: the Patriots. They only faced one other offense in the top ten, and Juan Castillo could pull a gameplan off the shelf for that one. Same with the Cowboys, who the Eagles faced once with Tony Romo and once without.

Next you have the Falcons, the juggernaut that couldn’t manage a single offensive point against the Giants this weekend, along with the Bills and 49ers, offenses that represent the height of mediocrity. Every other opponent the Eagles faced was as bad — or worse (!) — than the Rex Grossman-led Redskins.

Consistency

If you want to win week in and week out in the NFL, you can’t have wildly variable performances week by week, let alone quarter by quarter.

The Eagles defense had good moments, for sure, especially against the worst NFC East in years. But they could never sustain that success. We lived that roller-coaster ride throughout the season. So it’s unsurprising to note that the defense had the second-worst variance rate of any team in the NFL.

But even more telling is the volatility throughout a single game. The Eagles were 11th in the NFL in points allowed through the first three quarters. But in the fourth, that dropped to 25th.

That wasn’t an anomaly. Castillo’s defense was consistently out-schemed and out-played late in games. The Cardinals, for example, engineered a comeback largely by exploiting coverages that they knew were coming. That will lose you games even when you look alright on the stat sheet.

Situational

Carrying along from inconsistency, it’s worth noting that the Eagles defense wasn’t bad at everything. Much like the Jim Johnson defenses of old, which relied on a bend-but-don’t-break approach, Castillo’s unit had its own pluses and minuses.

Unfortunately, the 2011 Eagles had more of the break-but-don’t-bend qualities.

For example, the defense ranked 8th in the NFL in first downs per drive. That means they stalled out a large number of opponent drives — a positive indicator. That led to the Eagles allowing the 9th-fewest points per drive.

However, these numbers come with large caveats. For starters, the Eagles allowed the second-fewest number of field goal attempts in the league, and opponents converted a league-low 66 percent of those attempts. Some of that is indicative of good defense, some just of luck.

More importantly, despite fewer first downs, the Eagles were below average (19th) in touchdowns allowed per drive. Why? Two reasons. For the second straight season, the Eagles had an atrocious red zone defense (60 percent TDs allowed). The defense also allowed far too many big plays. They were 4th-worst in the NFL in 40-yard plays given up per drive.

Again, some things were great. Jim Washburn’s defense line posted the most sacks in the league. But at the same time, the Eagles were 24th in first downs allowed by penalties (unsurprising), and 22nd in turnovers (surprising). Not enough positives to outweigh the negatives.

* * *

In short, that’s how you can end up with seemingly solid overall points and yards allowed, without actually having a defense good enough to win more than eight games. And it’s also why Juan Castillo needs to go… or at the very least be demoted post-haste.

Photo from Getty.