The Eagles Aren't Good Enough To Make Mistakes

Despite disappointing results in the last two games, the Eagles are a good football team. Going into this week, they were first in the NFL in point differential and second in DVOA. Now, they remain third-best in the former and should stay near the top in the latter.

Two losses, by a combined score of eight points, do not end the season. But they do show us what kind of team the Eagles are: one that can't afford to make mistakes if they want to compete for the division title. 

The truth is that the Eagles don't have many difference-making players. Fletcher Cox and (old but still great) Jason Peters, perhaps. After those two, who can the team count on to consistently win individual match-ups? Carson Wentz has flashes of brilliance. Third down back extraordinaire Darren Sproles is the only explosive play maker on offense.

This roster isn't built to exploit mismatches in talent. It's built on competence. On defense, guys like Brandon Graham, Jordan Hicks, and Malcolm Jenkins form the core, but none of them are keeping offensive coordinators up at night trying to scheme around them. They are good because they do the right thing (most of the time). They won't get you killed and they can succeed within the scheme. Ditto on offense: Jordan Matthews and Zach Ertz are solid starters in the NFL as long as you're not counting on them to be the number one option.

The early season win streak was built on competence in all phases. The defense didn't do anything special with fancy blitzes; it just lined up and got pressure with four rushers. The offense took league-best field position and converted drives into points at the second-highest rate. They did so methodically, not gashing teams with big plays but marching down the field with a mix of efficient runs and short passes (part of the reason Carson Wentz scores so low in Air Yards). Limiting turnovers (to zero for the first three games) and capitalizing on opponent mistakes.

That strategy was effective until this started happening:

As far as I can tell, weeks five and six are the most penalties the Eagles have committed in consecutive weeks since 1989. And those penalties matter. According to friend of the blog Sean Taylor, each additional penalty a team has over its opponent is worth approximately -0.5 points. The Eagles have out-fouled their opponents by 16 in the last two weeks and, surprise, lost by a combined eight points. Add in a couple more unforced errors, like a rookie fifth round pick stumbling out of the gate and a veteran running back fumbling at the worst possible time, and you can see how the Eagles went from 3-0 to 3-2.

Again, this is not to bury them. They are still a good team that should be at least in the race all season. But it's not like teams we've seen in years past that could spot an opponent a three touchdown lead and roar back in the final minutes. There are too few #playmakers and not enough strengths. That means they either have to return to the suffocating competence of the early season—limiting turnovers and penalties, staying efficient on offense, and preventing big plays on defense—or come up with a new way to win... like putting more in Wentz's seemingly-capable hands.

The Truth About Bad Luck

Friend of the blog Justin F. wrote a post at Bleeding Green Nation in which he calculated the Eagles’ Pythagorean win percentage, a metric that estimates a team’s wins based simply on points scored and allowed. Justin notes that the Eagles jump up to just over .500 if you calculate win percentage using the Pythagorean method (compared to .385 in real life). Then he offers a common explanantion:

So why is there a such a discrepancy between the Eagles’ actual win percentage and Pythagenport win percentage?

A very good question, and the answer probably is not what people want to hear and/or believe, although it is true. In one word: luck. In two words: bad luck.

What follows in the comments at BGN is a discussion that unfortunately devolves into rambling incoherency, where the luck explanation is aligned with a belief in stats and doubters read like WIP callers. But the truth is more nuanced.

Explaning the difference between actual and Pythagorean win percentage as “bad luck” is overly simplistic. That is one possible explanation, but should not be used as an all-encompassing default. Rather, we must examine whether bad luck is actually the cause.

There are certain well-known measures for luck in the NFL. For example, fumble recovery, field goal percentage, schedule strength, and injuries. However, the Eagles haven’t had a problem with any of those.

The Eagles have recovered 48 percent of fumbles, just barely below average. They’re 10th-best in field goal percentage, with Alex Henery hitting 86 percent. Meanwhile, Eagles opponents are actually worst in the NFL, making only 65 percent of their attempts. The Eagles have faced the 18th-hardest DVOA schedule, and the three games Michael Vick lost to injury were average. No evidence of bad luck here.

However, there are plenty of other non-luck outlying factors that could account for the actual-Pythagorean discrepancy. Factors like:

• 30th in the NFL in opponent red zone TD percentage.
• 26th in fourth quarter points allowed.
• 32nd in interception percentage.
• 25th in first down-inducing penalties.
• 19th in DVOA variance.
• 27th in missed tackles (as of week nine).

If you’re going to call the 2011 Eagles underachieving — a label I’m not adverse to using — don’t blame it on the easy out. Luck is always involved, but it shouldn’t be an automatic determination. The above factors are much more likely to be the cause.

Photo from Getty.

By the Numbers: A Fall From Grace

At the end of Sunday’s game two players were firmly seated on their respective benches. One was Tom Brady, leader of an elite but far from perfect team, who took a rest after amassing a three touchdown lead. The other was DeSean Jackson, talented underachiever on a league-basement team, who was benched for poor play while the Eagles were trying desperately to come back.

There may be a more apt and fair comparison, but to me that distinction illustrates exactly how far the Eagles have fallen. They used to be an elite squad. Not anymore.

Let’s check the numbers:

7 = Times the Patriots entered Eagles territory. They scored 5 touchdowns and went 1 for 2 on field goal attempts on those 7 possessions. That touchdown percentage would be worst in the league for opponent trips to the red zone, let alone crossing the 50 yard line.

2 = Official tackles by Jamar Chaney. There is no doubt that on Sunday Chaney missed more tackles than he made. See the entries under “No need for linebackers” and “fundamentally sound” in the Eagles coaching handbook.

400 = Largely worthless passing yards by Vince Young. That is, however, a single game career high.

40% = Completion percentage on passes targeted at DeSean Jackson. Young’s total would have been even higher if not for Jackson letting two touchdowns and another 75 to 100 yards slip through his fingers. DeSean’s stock has never been lower. It is increasingly likely not only that he’s playing for another team next season but that the Eagles won’t be able to get back much for his services.

10 = Eagles penalties, for 60 yards.

6 = First half carries by LeSean McCoy. The Patriots secondary is bad and the Eagles were able to take advantage of that matchup frequently, so I won’t belabor the “Why won’t you run the ball?” point. But the Patriots were also deficient against the run, no matter what Reid said in his post game press conference. Especially in a game when the goal needed to be keeping Tom Brady off the field, McCoy should have been more of a factor. You can bet that’s what Jim Washburn was heckling Marty Mornhinweg about on the sideline.

.125 = Eagles win percentage at Lincoln Financial Field since Week 16 of last season. Seeing so many fans stream out of the stadium with a quarter left to play can’t have made Jeff Lurie happy.

Photo from Getty.

Which NFL Referee Was the Strictest in 2010?

Back during the 2010 season, I started charting penalties by referee crew to see if there was any significant and consistent difference in how strict particular referees are, or whether they favor home or away teams.

The data I used was just the standard total penalties from each game, and thus it does have some problems. First, 17 weeks is obviously a small sample size. It’s better than one or two games, but we will still need to take a few years to see if the same referees keep popping up. Furthermore, the total penalties don’t account for those that were declined, nor do they break them down into categories by type: holding, roughing the passer, facemask, etc.

However, the results are still interesting to examine. Let’s start with average penalties per game in 2010:

Just like it was most of the way through the season, Ed Hochuli’s crew gave out the most penalties, followed by Jeff Triplette and Tony Corrente’s teams. There’s a three to four penalty gap between those referees and the outlier on the other end, Pete Morelli.

Whether that difference is statistically significant is hard to say. Certainly there’s a good deal of variance game-by-game. However, Hochuli’s crew had only three games where they gave out just nine penalties. Refereeing the same teams, Morelli’s group had seven games in which they gave out nine penalties or fewer. It seems likely that somebody’s approaching the rules a little bit differently.

The above graph breaks down the penalties as to whether they were against the home or away team. This data was more volatile when I added in the final few games, but overall the trend is similar to the previous version. Some referees, like Walt Anderson, penalized one side on average much more heavily than the other.

The Takeaway Message

The reality is that we can’t tell for certain, using these numbers, whether any one of these referees is particularly biased or is unfairly enforcing or ignoring the rules. But the graphs do at least pose the question. When we see a football game turn on a single call by a referee, it’s tough not to conclude that a few extra violations called or disregarded can have a big difference.

Originally published at NBC Philadelphia. Photo from Getty.

Are NFL Referees Biased?

The data shown in the graph above depicts the average penalties called per game against away teams versus home teams. Those referees who call more fouls against the away team have positive bars, and those who call more against home teams have negative bars.

Once that becomes clear, it’s easy to see which referees have bias for or against the home team. For example, our good friend and “strictest NFL ref” Ed Hochuli is also the umpire with the second-most against home teams. He calls, on average, 1.67 penalties more against the home team than the visiting. The crew led by Walt Anderson is even worse.

Meanwhile, referees like Gene Steratore, Terry McAuley, and Pete Morelli have the opposite bias: against the visiting team by over 1.5 calls a game. if the Eagles had two home games, the first refereed by Anderson’s crew and the second by Steratore’s, there would be a swing of almost four penalties, enough to halt a drive, call back a long run, or miss a blatant late hit…

Which NFL Referee is the Strictest?

Most of the names on the list are relatively anonymous. Perhaps the most famous referee in the NFL, Ed Hochuli, also happens to be by far the most strict — by an average of one more penalty called a game than the next highest, Jeff Triplette.

Hochuli’s crew also calls an average of five more penalties a game than the two most lenient units, headed up by Pete Morelli and Scott Green. That amounts to a difference of about 50 penalty yards per week. Over the course of a season, that’s 80 extra penalty calls and 800 yards of game-changing decisions.