Writers of Pro Football Prospectus 2008

Most Recent FO Features

FitzgeraldLar04.jpg

» Scramble for the Ball: The DVOA Schism

Mike and Tom try to figure out what kind of secret sauce Arizona is feeding the media to sit at the top of the power rankings and in the middle of our DVOA rankings.

14 Oct 2008

Week 6 DVOA Ratings

by Aaron Schatz

Well, all looked reasonable with the DVOA ratings after most of this week's games were played. Then the Giants had to go and lose to the Browns. As a result, we're left with the 3-3 Philadelphia Eagles as the number one team in the NFL according to DVOA ratings.

Obviously, a 3-3 team at number one looks a bit strange to our readers. Don't worry, it looks a bit strange to me too.

What's particularly odd is that most of the typical "filtering out luck" arguments that often explain why teams have DVOA higher than conventional wisdom don't apply to the 2008 Eagles. There's one exception: Philadelphia has had extremely poor luck when it comes to opposition field goals. Opposing kickers are 13-for-13 against the Eagles, including seven field goals of 46 yards or more. By our estimates, this has cost the Eagles 10.4 points compared to average kicker performance.

On the other hand, Eagles opponents have been average in both kickoff distance and punt distance. The Eagles have not played a particularly hard schedule so far. They have recovered the majority of fumbles on both offense and defense. Opponents have only scored one defensive touchdown.

Looking at the standard stats except for wins and losses, it is pretty clear that the Eagles are an excellent team. Overall for the season, the Eagles have outscored opponents 167-123, tied for the third-largest point differential in the league behind the Giants and Titans. Their turnover differential is +5, second in the league. If they were ranked number three or four in DVOA, it wouldn't seem so weird. Number one tends to jump out at you.

Obviously, a lot of Philadelphia's "value" was built in their Week 1 blowout of St. Louis, a 38-3 victory which is worth about 80% DVOA with current opponent adjustments. And yet, even if we did not count that game, the Eagles would rank second in DVOA, narrowly ahead of Tennessee. The Eagles rank 12th in VARIANCE, so it isn't like they've been bouncing back and forth between great games and terrible ones.

One common complaint among our readers is that DVOA seems to overvalue teams coached by men out of the Bill Walsh coaching tree, playing a short pass-heavy "West Coast Offense." Right now, that includes Seattle, Philadelphia, Tampa Bay, Washington, and Green Bay (with an asterisk; Mike McCarthy is not really out of the Walsh tree, but he's kept a similar offense to the one the Packers used before his arrival). In the past, it included the Steve Mariucci Lions, the George Seifert Panthers, Mike Sherman Packers, and so on. There is an argument to be made that Mike Shanahan and Gary Kubiak belong to this school as well, although their offense is far more run-heavy. The same goes for Greg Knapp, who was offensive coordinator under Jim Mora in Atlanta from 2004 through 2006.

To try to figure out if there was something to this argument, I plotted VOA (i.e. no strength of schedule or fumble luck adjustments) compared to actual wins for every team from 2000-2007. The dots in orange represent the "WCO" teams. The dots in blue represent the rest of the league. This chart does not include Shanahan, Kubiak, and Knapp as part of the "WCO" group.


As you can see, the West Coast Offense teams do slightly better compared to VOA, but the difference is so tiny as to be pretty much meaningless. The R-squared of that WCO trendline is .792, and the R-squared of the other trendline is .798, so VOA explains the win-loss record of WCO teams about the same as it explains the win-loss record of other teams. If we include the three years where Mora coached Atlanta, the difference actually becomes a little bigger, but still not much (.779 vs. .800), and if you toss the Shanahan and Kubiak teams in as well, it makes no difference. Based on this cursory analysis, it certainly seems like DVOA does not necessarily favor teams from the Holmgren/Reid coaching line.

Bill Barnwell actually addressed this in the Eagles chapter of Pro Football Prospectus 2008.

"Is it even necessarily the case that DVOA looks at the Eagles with some sort of rose-colored glasses? While DVOA has certainly seen them to be a better team than their record indicated over the past three seasons, that wasn't always the case... Much as anyone could've said that the Eagles were underrated by DVOA from 2002-2004, the Eagles have been overrated over the last three years. It just so happens that those three years have been the ones in which we've released books."


Philadelphia Wins vs. Advanced Stats
During Andy Reid Era
Year Actual
Wins
Pythagorean
Wins
DVOA
Expected Wins
1999 5 5.5 7.1
2000 11 11.2 10.8
2001 11 12.3 11.1
2002 12 12.5 10.7
2003 12 10.4 10.3
2004 13 11.5 11.4
2005 6 5.9 7.4
2006 10 9.8 11.6
2007 8 9.1 9.6
Avg 9.8 9.8 10.0

The biggest explanation for the difference between Philadelphia's win-loss record and total DVOA is that the offense has played much worse in the most important situations on the field. The Eagles rank fourth in DVOA on first down, 12th on second down, and 16th on third down -- including dead last on third-down runs. The Philadelphia offense ranks 17th in the red zone and a pathetic 31st in "close and late" situations (second half, score within a touchdown). They also have a poor red zone defense, although the defense has been very good in "close and late" situations.

We know that the "third-down rebound" effect is much stronger between seasons than it is during one specific season. I haven't had time to research whether poor performance in the red zone or in "close and late" situations tends to carry over for an entire season. However, if the Eagles offense can improve in these situations to match what it is doing the rest of the time, we won't be arguing about the team's place atop the DVOA ratings at the end of December.

* * * * *

Special teams numbers are more variable from year-to-year than either offense or defense, but Oakland's turnaround is extreme even for special teams. The Raiders currently rank number one in our special teams ratings after ranking 29th, 26th, and 29th over the last three seasons. They are doing this with pretty much the same personnel as last year at the "main" positions: Sebastian Janikowski as kicker, Shane Lechler as punter, and Johnnie Lee Higgins on returns. Even stranger, the Raiders only rank in the top ten in one of the five elements of special teams: net punting, where they are second. They lead the league in special teams because they aren't weak anywhere, with positive value in all five categories.

The big difference on punts is not Lechler, who is always good. It is the punt coverage. Last year, Lechler ranked third in the league with 9.9 points of value on punts, but the Raiders gave up 8.3 points worth of field position on returns, 31st in the NFL. This year, Lechler once again ranks third in the league with 5.4 points of value on punts -- but this year, the Raiders have allowed -2.0 points worth of field position on returns, ninth in the NFL. I actually have a call in to a Bay Area reporter to talk about whether any particular player is the star of the coverage teams, and I'll let everyone know if I find out anything useful.

One other thing I noticed this week: Detroit has the second-hardest remaining schedule in the league. Yikes. Oh well, at least they have the Red Wings to enjoy.

All stats pages are now updated, including team and individual stats as well as playoff odds.

* * * * *

These are the Football Outsiders team efficiency ratings through six weeks of 2008, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)

OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. They also include opponent adjustments, currently at 60 percent strength. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season.

As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.

DAVE is a formula which combines our preseason projection with current DVOA to get a more accurate forecast of how a team will play the rest of the season. Right now, the preseason projection makes up 19 percent of DAVE, and this year's DVOA makes up 81 percent. For teams with only five games played, that split is 27/73. DAVE also discounts the first two games of the season by five percent.

To save people some time, please use the following format for all complaints: <team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>


TEAM TOTAL
DVOA
LAST
WEEK
TOTAL
DAVE
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
1 PHI 35.7% 3 34.4% 1 3-3 16.8% 9 -16.2% 6 2.6% 14
2 NYG 34.6% 1 28.7% 2 4-1 39.2% 1 7.5% 20 2.9% 10
3 TEN 26.4% 4 18.7% 4 5-0 -3.4% 20 -30.5% 2 -0.8% 23
4 TB 24.2% 12 22.4% 3 4-2 1.5% 16 -22.8% 3 -0.2% 20
5 WAS 22.2% 5 16.7% 5 4-2 23.4% 4 -3.6% 8 -4.8% 29
6 ARI 22.0% 8 15.4% 7 4-2 19.5% 7 -1.5% 13 1.0% 19
7 BAL 19.9% 2 15.1% 8 2-3 -8.2% 25 -30.6% 1 -2.5% 26
8 PIT 15.9% 9 13.2% 12 4-1 -7.2% 22 -19.9% 4 3.1% 9
9 NO 15.7% 13 13.3% 11 3-3 18.9% 8 6.1% 18 2.9% 11
10 CHI 14.8% 11 11.2% 13 3-3 2.7% 14 -9.9% 7 2.2% 17
11 IND 13.7% 24 14.5% 10 3-2 13.3% 11 -1.1% 14 -0.7% 22
12 MIA 12.2% 10 5.9% 15 2-3 21.0% 6 -3.0% 10 -11.7% 31
13 SD 10.3% 14 14.5% 9 3-3 14.9% 10 6.8% 19 2.1% 18
14 GB 8.6% 18 16.0% 6 3-3 2.0% 15 -3.0% 9 3.6% 8
15 DAL 8.3% 7 8.7% 14 4-2 24.7% 3 8.1% 21 -8.3% 30
16 ATL 8.2% 16 -4.8% 22 4-2 21.0% 5 17.2% 27 4.4% 5
TEAM TOTAL
DVOA
LAST
WEEK
TOTAL
DAVE
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
17 CAR 5.8% 6 5.7% 16 4-2 5.4% 13 -2.4% 11 -2.0% 25
18 DEN 5.4% 15 4.5% 17 4-2 26.6% 2 20.7% 29 -0.5% 21
19 NYJ 4.2% 20 1.9% 19 3-2 -0.1% 18 -1.6% 12 2.7% 12
20 BUF 0.5% 17 -4.1% 21 4-1 0.5% 17 5.0% 16 5.0% 3
21 MIN -2.5% 19 2.8% 18 3-3 -7.7% 24 -17.9% 5 -12.8% 32
22 JAC -4.8% 25 -0.9% 20 3-3 8.7% 12 16.1% 25 2.6% 15
23 SF -8.1% 21 -13.3% 25 2-4 -10.0% 26 2.3% 15 4.2% 6
24 CLE -11.6% 26 -11.6% 24 2-3 -7.4% 23 8.8% 22 4.7% 4
25 NE -19.9% 23 -10.9% 23 3-2 -5.4% 21 19.7% 28 5.2% 2
26 OAK -24.3% 22 -24.3% 28 1-4 -17.3% 28 13.2% 23 6.2% 1
27 HOU -28.0% 29 -21.0% 27 1-4 -0.6% 19 31.2% 31 3.8% 7
28 SEA -32.5% 27 -19.4% 26 1-4 -13.1% 27 16.8% 26 -2.6% 27
29 CIN -34.7% 28 -25.6% 29 0-6 -27.5% 31 6.0% 17 -1.2% 24
30 DET -50.1% 31 -46.4% 30 0-5 -22.4% 29 30.5% 30 2.7% 13
31 STL -55.6% 32 -48.3% 32 1-4 -26.1% 30 31.8% 32 2.2% 16
32 KC -57.5% 30 -46.4% 31 1-4 -38.2% 32 16.1% 24 -3.2% 28

  • ESTIMATED WINS uses a statistic known as "Forest Index" that emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close. It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles. Teams that have had their bye week are projected as if they had played one game per week.
  • PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • VARIANCE measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from most consistent (#1, lowest variance) to least consistent (#32, highest variance).


TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR. RANK
1 PHI 35.7% 3-3 41.6% 4.4 1 -0.4% 18 7.1% 5 10.9% 12
2 NYG 34.6% 4-1 47.8% 4.2 3 -18.7% 32 16.3% 1 37.3% 32
3 TEN 26.4% 5-0 30.4% 4.3 2 -8.4% 29 -8.1% 27 5.4% 4
4 TB 24.2% 4-2 23.2% 3.6 12 9.8% 5 -11.9% 29 20.0% 21
5 WAS 22.2% 4-2 14.1% 4.1 4 10.1% 3 -2.3% 18 6.4% 7
6 ARI 22.0% 4-2 15.6% 3.7 9 6.6% 8 -13.1% 30 19.4% 20
7 BAL 19.9% 2-3 17.6% 3.7 10 1.6% 15 2.6% 12 33.6% 31
8 PIT 15.9% 4-1 19.4% 3.9 5 1.9% 14 3.5% 10 5.6% 5
9 NO 15.7% 3-3 18.2% 3.8 6 2.8% 13 -2.2% 17 7.6% 9
10 CHI 14.8% 3-3 15.2% 3.8 7 6.2% 9 -8.4% 28 5.0% 3
11 IND 13.7% 3-2 16.4% 3.8 8 -0.1% 16 -6.2% 24 29.4% 26
12 MIA 12.2% 2-3 19.0% 3.2 17 -1.9% 20 -16.7% 32 31.6% 29
13 SD 10.3% 3-3 15.0% 3.7 11 -2.7% 21 -5.6% 23 8.6% 10
14 GB 8.6% 3-3 11.1% 3.5 13 -7.4% 28 0.6% 15 9.7% 11
15 DAL 8.3% 4-2 11.1% 3.2 18 7.0% 7 9.1% 3 16.9% 18
16 ATL 8.2% 4-2 11.8% 3.3 14 -9.0% 30 3.0% 11 24.4% 23
TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR. RANK
17 CAR 5.8% 4-2 5.2% 3.3 15 -0.4% 17 6.0% 7 24.6% 24
18 DEN 5.4% 4-2 8.3% 3.0 21 -6.1% 26 -7.2% 26 19.2% 19
19 NYJ 4.2% 3-2 2.1% 3.2 16 -1.7% 19 -15.3% 31 15.4% 17
20 BUF 0.5% 4-1 5.7% 3.1 20 -15.9% 31 -6.8% 25 14.4% 16
21 MIN -2.5% 3-3 -5.4% 2.7 24 3.4% 12 4.4% 9 7.5% 8
22 JAC -4.8% 3-3 -5.1% 2.9 22 5.7% 10 -4.4% 21 3.3% 1
23 SF -8.1% 2-4 -7.1% 2.6 25 -4.9% 25 -4.0% 20 11.6% 13
24 CLE -11.6% 2-3 -12.8% 2.9 23 7.3% 6 7.2% 4 32.0% 30
25 NE -19.9% 3-2 -16.4% 3.2 19 -6.5% 27 -3.8% 19 23.5% 22
26 OAK -24.3% 1-4 -25.8% 1.8 26 -4.3% 24 -1.5% 16 28.8% 25
27 HOU -28.0% 1-4 -31.2% 1.4 27 10.6% 2 -4.5% 22 5.9% 6
28 SEA -32.5% 1-4 -26.9% 1.3 28 -3.3% 22 6.7% 6 30.9% 28
29 CIN -34.7% 0-6 -38.6% 1.2 29 13.6% 1 2.1% 14 4.9% 2
30 DET -50.1% 0-5 -49.5% 1.0 30 3.5% 11 9.6% 2 12.5% 14
31 STL -55.6% 1-4 -58.5% 0.2 32 10.1% 4 2.3% 13 13.9% 15
32 KC -57.5% 1-4 -43.0% 0.5 31 -4.1% 23 5.0% 8 29.8% 27


Worst DVOA Ever Watch


WORST TOTAL DVOA
AFTER FIVE GAMES
WORST DEFENSIVE DVOA
AFTER FIVE GAMES
2000 CIN -64.2% 2006 HOU 34.7%
1999 CLE -62.4% 2008 STL 31.8%
2005 SF -61.1% 2008 HOU 31.2%
2005 HOU -59.2% 2005 HOU 30.5%
2001 WAS -57.8% 2008 DET 30.5%
2008 KC -57.5% 1998 WAS 30.3%
1996 TB -55.9% 2000 SF 29.9%
2008 STL -55.6% 2000 ARI 29.7%
2002 CIN -55.4% 2007 NYJ 29.1%
2000 SD -52.2% 1998 PHI 28.4%
2006 TEN -51.9% 1998 CIN 27.8%
2006 OAK -51.7% 2000 SD 27.8%

Posted by: Aaron Schatz on 14 Oct 2008

63 comments, Last at 18 Oct 2008, 6:48am by Anonymous

Comments

1
by Matt :: Tue, 10/14/2008 - 5:49pm

During Week 6, teams ranked in the top ten from Week 5's ESPN power rankings went 0 and 7 (with three byes). The average margin of defeat was 11.7 points.

During Week 6, teams ranked in the top ten according to Week 5 DVOA went 2 and 6 (with two byes). Those eight teams were outscored on average by 7.75 points.

ESPN Power Rankings are clearly ranked too high because they suck compared to DVOA. DVOA is way better than this. Being right 25% of the time is way more AWESOME than 0%, biatch!

2
by Bobman :: Tue, 10/14/2008 - 5:52pm

Man did the Colts shoot up in one game. Not entirely sure it's justified, but the next few weeks will tell. I'm looking at the estimated wins and see only five ahead above their 3.8 wins. That makes me feel pretty good. Whoo-hoo, I'll take the small breaks where I can find them.

22
by turbohappy (not verified) :: Tue, 10/14/2008 - 10:41pm

Look at their variance - went from #1 to #26 ;o)

3
by Bobman :: Tue, 10/14/2008 - 5:56pm

How are the Colts ST only ranked 22nd?

Aaron, how many times do you need Russ Purnell to wash and wax that new Bentley he got you before you show them a little love? Hard to fault them for minimal return yardage this week since they only saw 2 KOs. And they consistently buried the Ravens inside their 15 on their own KOs.

Okay, I know, consistency... let's see them put together five more games like that.... arrrgh.

40
by turbohappy (not verified) :: Wed, 10/15/2008 - 11:04am

The coverage game has been good for the Colts. Oddly enough it's the FG/XP kicking (only 5/7 with no >50 attempts) and the return game that is killing their rating. The coverage is well into the positive on both punts and kickoffs.

59
by turbohappy (not verified) :: Wed, 10/15/2008 - 8:50pm

Thinking about it, I think it is primarily due to penalties on returns (in addition to not being impressive kicking FGs).

4
by scatman (not verified) :: Tue, 10/14/2008 - 6:01pm

This is surely sweet reading for Arizona fans (sic)? Maybe, just maybe this year......

5
by ammek :: Tue, 10/14/2008 - 6:08pm

"if the Eagles offense can improve in these situations to match what it is doing the rest of the time, we won't be arguing about the team's place atop the DVOA ratings at the end of December"

Although that 5th-ranked strength of future schedule might have some effect.

Did your research suggest that the previous years' over-optimistic DVOA projections for Philadelphia happened for the same reasons: third down, red zone, bad offense close & late? If so, then the Tanierians might have a point: that Reid's gameday coaching is the problem. And it might also explain the repeated expectation of improvement: DVOA says 'this year they must improve on those power situations'; fans say 'but they still haven't signed a proper fullback or a tight end'.

6
by Grafac (not verified) :: Tue, 10/14/2008 - 6:13pm

Wow Indy. Talk about an important game. Blowing out superior competition (number one defense) counts for a lot. What is the largest one game change (minus and negative) in DVOA history?

7
by tuluse :: Tue, 10/14/2008 - 6:16pm

The Bears have the best offense in their division through 6 weeks, who would have guessed that? Also, they have the best passing and rushing offense, so they aren't using one to cover the other or are just less bad then the Packer's rushing offense or the Viking's passing offense.

37
by Eddo :: Wed, 10/15/2008 - 9:32am

Can't say I saw it coming, but after watching the Bears (and the Packers and Vikings and Lions) all year, DVOA echoes my qualitative observations. It even has the pass offense at 10.3%, which makes sense, and the rush offense at -5.1%.

Orton looks good. Not just "better than expected", but actually good.

Forte looks good, but I think the line is overrated (and they rank 26th in ALY). If you take Forte and his tremendous vision (and he's a rookie!) out of the offense, they wouldn't be able to run on Northwestern, let alone any NFL team.

8
by Joe :: Tue, 10/14/2008 - 6:22pm

The Eagles problem is really that they can't run with any power, something that should improve if Westbrook can stay on the field and Shawn Andrews can play again this season. Not necessarily running for more power but being more successful in power situations. It's a bigger deal than people want to acknowledge - Reid can run shuffle passes with Westbrook, or pitch it out to him, or line him up as a wideout, or have him run an option route - these are integral parts of the Eagles offense that they can't run as effectively with the backup talent at RB. The two games this year where Westbrook was 100% (Rams and Cowboys) the offense was much better. Small samples against bad defenses apply, but it's something to consider.

But watching them every week I can see both why DVOA likes their offense and also why they are 3-3. They march down the field 15 yards at a clip and then stall inside the 10 yard line - it's weird but they don't have an answer in short yardage situations. Against San Francisco the Eagles we in the red zone a ton but couldn't convert - Akers kicked 4 field goals and McNabb threw an interception. Andy Reid has to get more creative with Westbrook out - I'm not sure why he's not using McNabb's athleticism other than if McNabb is hurt more than they are letting on. They should be rolling him out or bootlegging him on a play fake and telling him to run it in.

11
by MJK :: Tue, 10/14/2008 - 6:43pm

It seems to me like the Eagles are a team with a mobile QB, a shifty RB who's good at catching the ball and moving in space, and pretty fast WR's. These sorts of teams, when they have plenty of field to play with, can move the ball, because the defense has to worry about containing all the moving parts in a very large space. Your observation that the Eagles' real problem is that they can't run with power fits with this assessment, although I'll have to take your word for it (I've only seen one Eagles game this year so far).

How are the Eagles' TE's? Frequently, red-zone and critical down situations depend heavily on how much of a big receiving threat the TE is--a big athletic guy that can run good routes, get a little separtion, and catch a high-thrown ball in traffic on 3rd-and-6 from the 10 yard line (or on 3-rd and goal from the 5)--think Gates or Gonzalez--goes a long way towards succeeding in the red zone and on critical third downs. But for the life of me, I can't think of any notable TE name playing for the Eagles... Could that be part of the problem?

As to why DVOA is so high on them...I wonder if that relates to their lack of power running. Have they had a hard time holding leads in games? One thing that DVOA totally neglects is how well your offense allows you to manage the clock. A run for 1 yard on 1st and 10 (or 2nd and 10, for that matter) is nearly a total failure for the offense by DVOA's measure, but if you're up by nine with four minutes to go and the other team is out of timeouts, then it's actually a heck of a lot more valuable than a 4 yard pass where the reciever steps out of bounds, which DVOA would view as a "success". And in late and leading but close situations, a series of 4 yard runs is much more valuable than a series of 10 yard passes, even though DVOA likes the latter more.

Could this be some of what is messing things up? Someone who's watched more Eagles games than I have should weigh in...

13
by tuluse :: Tue, 10/14/2008 - 6:48pm

LJ Smith is the Eagles TE. He is not very good, and frequently hurt. Also, DVOA adjusts for garbage time, but I think that more research is needed on clock management.

Take the 2005 Bears for example. They knew their defense was so good, that as soon as they got points, they should just start running out the clock. This was good for 11 victories.

33
by qed (not verified) :: Wed, 10/15/2008 - 8:15am

The tight ends (L.J. Smith and Brent Celek) haven't produced much this year, and they usually don't have a tall "fade" wide receiver for goal-line situations (they're just starting to use Hank Baskett in this role). In previous years their power running game in short-yardage situations has been pretty good (I don't have my PFP here but I think they were close to the top of the league in short-yardage runs last year) so I think this may be a one-year no-Westbrook aberration. The usual Andy Reid red-zone offense involves quite a bit of misdirection and McNabb bootlegs so the lack of Westbrook and tight end production is definitely hurting that.

43
by Tom (not verified) :: Wed, 10/15/2008 - 12:40pm

The Eagles Tight Ends are L.J. Smith and Brent Celek. Celek is more of a big bodied guy who seems to be able to block well in sweeps, which comes in handy as the Eagles don't have a fullback and run a lot of single half back sets with Westbrook. However, he doesn't seem to have much ability to seperate himself from linebackers in coverage and on the rare occasion he does get open his hands are inconsistent at best. L.J. Smith is a very athletic guy who has all the tools to succeed and flourish in the NFL. However, he doesn't seem to have much blocking ability. He's very good in space and runs crisp routes making it difficult for linebackers to keep up with him and as he's too big for a safety to cover he can create matchup problems. That all being said, his hands are possibly worse than Celek and he shows no fundamentals whatsoever. He fumbles much too often (he carries the ball almost away from his body with his right arm), and whats worse, despite his athleticism, he is just awful once the ball is in the air, showing no ability whatsoever to make a play. L.J. Smith's lack of ability to go after a ball is what caused McNabb's red zone pick last week. The Eagles also have a lot of fast guys like Greg Lewis and Desean Jackson. Jackson is lighning fast and a very good route runner, he's almost always open. Jackson is little but makes up for it in fearlessness over the middle, however he will drop catchable balls now and again. Greg Lewis is speedy, frequently reminding me of a fast kid playing with the big boys. I've heard he has the best hands on the team but i've never seen anyone drop as many easy passes. He's afraid of getting hit and is, quite frankly, a complete pussy. Avant and Baskett are big bodied guys, Baskett a bit taller and Avant more thick. Avant is solid over the middle, with average speed but good field awarness, Baskett is fast and big with a great vertical. Neither of them seem to be able to catch passes for their lives, frequently dropping passes they should catch. The Eagles biggest problem this year are Number 1.) a lack of physical players and Number 2.) a complete lack of hands, they lead the league in dropped passes by like 7. Westbrook is sure-handed, quick, smart, and while not incredibly fast he shows great acceleration, making him capable of getting through holes before you can see him. He is also, surprisingly, a great pass blocker, always helping out a tackle with a chip block on his way to the flat or flat out taking out linebackers by the knees when necessary. The Eagles operate so well untill they hit the red-zone, with a few three and outs and frustrating third down failures that can be attributed pretty well to our lack of hands. We are fast and can pick up yards in bunches. But right now our operating without Kevin Curtis, who is terrifically underrated, quick and with great hands, a solid blocker on outside runs, his only weakness is his inability to get out of a jam. His return, coupled with Westbrook, Reggie Brown, and Shawn Andrews getting healthy and on the field, should make us a much better football team. Unless L.J. Smith suddenly lives up to his potential we should probably struggle in the red zone all season.

9
by Joe Skolnik (not verified) :: Tue, 10/14/2008 - 6:27pm

As an eagles fan, i obviously think we will 60 13-3 and win the super bowl. The Eagles have the most impressive victory of the year ( a 15- 6 domination of the Steelers that was not as close as the score would suggest) except perhaps the Colts over Ravens yesterday. The eagles barely lost 3 close games by a total of 14 points. The Eagles when playing well have a top 5 ofense and Defense.

19
by Wait, what? (not verified) :: Tue, 10/14/2008 - 8:16pm

As a fellow Eagles fan, I have to wonder if we're watching the same team. How is it "obvious" that they'll go 13-3? I'd put a large sum of money down on that not happening, if I could find anyone to take the bet. We are apparently operating under different definitions of "impressive victory" as well.

58
by armchair journe... :: Wed, 10/15/2008 - 7:36pm

@ Wait, what?

IggleJoe is saying that as a Philly fan, "obviously" he predicts the best possible outcome for his team.

... Which doesn't sound like a real Philly fan at all, come to think of it!

_______________________________
armchair journeyman quarterback

26
by The Ninjalectual :: Tue, 10/14/2008 - 11:10pm

Hooray! We have found an EagleJoe! His name is even Joe!! Woo-hoo.

Chris Horton for defensive rookie of the year.

39
by Eddo :: Wed, 10/15/2008 - 11:03am

At the risk of being shunned from the community, I'll even go so far as to claim that using "60" for "go" is better than any single spelling or grammar peccadillo that Raiderjoe has demonstrated.

57
by Alex51 :: Wed, 10/15/2008 - 6:16pm

Blasphemy!

36
by TomC :: Wed, 10/15/2008 - 9:22am

As [a Bears] fan, i obviously think we will 60 13-3 and win the super bowl. The [Bears] have the most impressive victory of the year ( a [29-13] domination of the [Colts] that was not as close as the score would suggest) except perhaps the Colts over Ravens yesterday. The [Bears] barely lost 3 close games by a total of [8] points. The [Bears] when playing well have a top 5 ofense and Defense.

OK, maybe not that "top 5 ofense" part.

10
by Tom Gower :: Tue, 10/14/2008 - 6:34pm

I'm surprised to see BAL still #1 in defense after this past Sunday, though it is narrowly so and TEN is tops in VOA.

This weekend should be interesting: TEN D (pass DVOA -47.3%) v. KC O (pass DVOA -61.3%). I wonder what the last team was to have negative net yards passing.

12
by Bucs Fan (not verified) :: Tue, 10/14/2008 - 6:45pm

Wow the Bucs shot up to 4th, probably because of the return of Garcia, one of the stingiest D in the league, and a good special teams play against Panthers.

14
by DaveInTucson :: Tue, 10/14/2008 - 6:53pm

The Packers' defense is rated 9? I'm a Packer homer, but their defense hasn't been very good this year. The Vikings, Cowboys, Buccaneers, and Falcons all rushed for at least 176 yards on them (and even the Seahawks managed 113).

They haven't been doing well against teams that can run and pass (Cowboys, Bucs, and Falcons so far). Up next: the newly-rejuvinated Colts! Even if Addai can't play, it could get ugly fast. :-(

I have a blog where I post objective, predicive power rankings of NFL teams.

52
by GGtheMad (not verified) :: Wed, 10/15/2008 - 3:01pm

You've got to look at the breakdown on the defense though and while I haven't been able to watch the games the stats seem to bear out a bit.

The Packers run defense is awful and if you look at the Team Defense page here on the site it will tell you that, a 17.4% DVOA. Only Detroit (20.6%), St. Louis (25.8%), and Houston (30.7%) rank worse against the run.

Statistically the pass defense looks pretty good. #3 behind Baltimore and Tennessee. Now some of that is because teams can run all day, every down on the Packers. Some of that is because they have been playing "Run First" teams. Minny doesn't pass the ball. Any pass D would look good against them. Detroit passes a lot but Detroit just sucks it doesn't matter. Dallas has 201 passing attempts to 161 rushes, relatively balanced, but you'll note that with the exception of a couple of big pass plays to a guy no one really had scouting info on and who was being covered by the 3rd string safety (Charlie "I've got some" Peprah "in my eyes") on the plays they didn't let Romo have his way. 56% completion rate for Romo means the D did something to him. Tampa Bay is still a run offense. I know 225 pass attempts to 165 rushes but can you really count that 67 attempt game vs the Bears? Chop 30 off that game for what would still be a season high 37 and you are looking at a team that likes to run it's Graham/Dunn duo at you first, then pass. Atlanta is again another run oriented team. Seattle is 50/50 split on pass to rush attempts and they just suck right now too, but yeah they would lean on the run against a crap run D like the Packers have.

So yeah, I don't think their pass defense is as good as the numbers say, but it's not horrible (even if you really only have the Dallas game to measure it against and if they would have had any starting safety on the field the big pass plays probably wouldn't have happened and a couple of the big pass plays were more or less a run anyway and the failure of the D-line and the backers to deal with it was part of the issue).

You can expect that ranking to drop. I don't see a revitalized Colts offense having problems with it. But Woodson is still playing well and I think Tramon Williams is playing better than Al Harris. Yeah I said it. Nick Collins (who again was out for most of the Dallas game) has finally started to play the way we wanted him to for the past 3 years. Aaron Rouse is better than Atari Bigby in some aspects of the game so I don't mind him being on the field while Bigby is hurt, though I'd like to see if Bigby that played the last few games and the playoffs last year is the Atari model we have or if someone swapped in a Playstation for him for those games and if he would revert to the old Atari model that just couldn't read plays and was out of position so often early in the year.

But the D-Line is horrendous. No pass rush, no ability to stop the run. It's very easy for teams to double Kampman on pretty much every snap because with Jenkins injured, KGB back behind the iron curtain, and no defensive tackle that isn't playing at above 70% health offenses don't have to worry too much. But the secondary has generally done a pretty good job. Mistakes have been made, but some of those you expect from 3rd stringers forced to play 2-3 quarters a game due to injury.

The lack of a run offense for the Packers doesn't help either. They can't eat the clock with this offense so the defense ends up in bad situations as well.

We'll see. I expect the pass defense to start looking worse than it is listed (though I hope they keep up the crazy int rate 3 players with at least 3 ints at this point is nice) but I do have some hopes that the run defense can maybe climb up from 29 to 22 or 23.

15
by BucKai :: Tue, 10/14/2008 - 7:03pm

Just looking at the Buc's schedule, they've had the 5th hardest through the first 6 weeks and managed four wins, and have the 4th weakest schedule from here out. Of course, their opponents also have the 21st ranked variance so I guess it'll depend on which opponent shows up each week.

I was really looking forward to the Dallas game in two weeks as a serious test, but with Romo & now Pac-Man gone, the Jerry Jones crew just doesn't seem as intimidating.

16
by Gold_Star_for_R... :: Tue, 10/14/2008 - 7:05pm

This is surely sweet reading for Arizona fans (sic)? Maybe, just maybe this year......
Their future schedule ranks 30th.
This year, indeed.

27
by The Ninjalectual :: Tue, 10/14/2008 - 11:13pm

Beating Dallas is fun, no?

Chris Horton for defensive rookie of the year.

17
by scottb (not verified) :: Tue, 10/14/2008 - 7:53pm

I'm curious how DVOA handles penalties and the resulting down. Is a team judged as having 2 downs? For example, if an offense has a first and 10 and gets a 5 yd. false start does that give them a negative VOA, then they complete a 10 yd pass on first and 15 giving them a positive VOA and they registered 2 1st down plays? If this is the case, does the defense receive a corresponding negative VOA for the offensive penalty? or does the defense only have 1 1st down in this scenario, the actual play?

Also have you looked at what specific penalties carry over from game to game? Is a team called for holding (for instance) 4 or 5 times in one game more likely to be called for holding in the future?

18
by KevinNYC (not verified) :: Tue, 10/14/2008 - 8:08pm

The Giants defense deservedly took a 13 spot drop after yesterday's embarrassing performance. I choose to see last night's game as a fluke... if their defense was truly that bad they wouldn't have won a game yet.

Anyone know what the highest finish for an offensive running game is? The Giants' running VOA is double the next closest team's before adjustment and slightly below double after the adjustment.

20
by maxpower19 (not verified) :: Tue, 10/14/2008 - 9:16pm

I would like to point out that in deciding whether DVOA overrates WCO teams, correlation to wins is not the best measure to use because wins are strongly affected by special teams and defense. Remember that only 3/7 of Philly's DVOA is from their offense. Better questions would be: Do WCO teams score fewer offensive points, after adjusting for starting field position, than non-WCO teams with similiar offensive VOAs? Also, does DVOA under-penalize the Eagles' offense for not having scored in goal-to-go situations?

34
by mawbrew :: Wed, 10/15/2008 - 8:57am

I think this is a really good point. The suspiscion about WCO bias in DVOA is related, I think, to a sense that DVOA undervalues successful 'big plays' relative to lots of successful 'little plays'. And it's actually a relatively few teams that this would seem to effect - the Eagles, Bucs, and Seahawks probably. The Packers always seemed to have more than their share of big plays and the Lions were so bad that, well, the bias didn't matter.

Similarly, there was a sense there was a DVOA prejudice against big play/'Tampa 2' defenses. The sense was that DVOA undervalued big defensive plays (sacks, turnovers) that ended drives. That seems to be less of an issue so far this year .

35
by Andrew B :: Wed, 10/15/2008 - 9:11am

In 2006 the Eagles were the big-play team par excellence, and DVOA was still overvaluing them. That was the year the Eagles put on multiple one, two, and three play scoring drives in some games.

The Original Andrew

21
by panthersnbraves :: Tue, 10/14/2008 - 10:30pm

The Panthers' drop of 11 spaces is entirely too much because ..., um ..., you see ..., and then ... nevermind.

NFC South roolz! (almost)

23
by Chad Gerson (not verified) :: Tue, 10/14/2008 - 10:53pm

Link to playoff odds?

25
by Telamon :: Tue, 10/14/2008 - 11:10pm

I too would like such a link.

I know that one can use the drop down menu, but unfortunately, some of the banner ads (especially the Identity Theft one) break the drop down menu by loading over it and making the lower options inaccessible.

48
by Tom Gower :: Wed, 10/15/2008 - 2:32pm

There's a link to playoff odds (and the other stats) at the bottom of the page. Just scroll down, or refresh the page to get rid of the annoying ad (which problem FO knows about).

24
by zlionsfan :: Tue, 10/14/2008 - 10:59pm

2-1, tied for first in the Central, road wins over Ottawa and Carolina. Yep, it's a pretty good start to the season for the defending champs ...

Oh yes, football.

right.

let's not talk about Detroit.

How about the Dolphins, with the easiest remaining schedule at this point?

You know, the Bengals have had the hardest schedule so far, so they're going to win a game too, right?

Not that I'm paying attention to the upcoming draft or anything.

28
by Mig (not verified) :: Tue, 10/14/2008 - 11:37pm

PHI actual wins vs. estimated wins 2005, 2006, 2007, 2008.

2005 estimated wins +1.4
2006 estimated wins +1.6
2007 estimates wins +1.6
2008 estimated wins +1.4

Talk about consistency. When a team STRUGGLES to beat the #23 team, and rises to #1, something is not right.

Other than that, DVOA rocks !!!!

30
by raidersfan (not verified) :: Wed, 10/15/2008 - 2:56am

As much as I agree with you on how Eagles shouldn't be #1, with the Titans and G-men as the number 1 (though, to be fair, the Eagles with a healthy Westbrook may be the top team after the bye week, and isn't the point of DVOA to be predictive?), as an obsessed football fan who actually watched the game, the Eagles only had trouble in that game because of the stupid decision of Andy Reid to kick a long field goal, and one good drive that the 49ers had. While watching the way the game progressed, it was obvious that the Eagles were the significantly better team, and final result was fairly indicative of how the two teams played.

61
by Alex51 :: Thu, 10/16/2008 - 5:35pm

You do realize that estimated wins assumes an average schedule, right? And that the Eagles have faced tougher than average opponents over the 2005-2008 timespan? In fact, schedule strength accounts for roughly half of the variation in (actual wins - estimated wins) during the Andy Reid era. So, basically, the Eagles win less often against good teams than they do against bad teams. Interesting. Who knows, this strange tendency might be shared by all WCO teams. We might be on to something here!

29
by RickD :: Wed, 10/15/2008 - 2:07am

Do you think the disparity between DVOA and actual wins for Philly is related to their coaching and their inability to win games that it seems like they should win? It seems like Andy Reid is great at consistently fielding a competitive team, but his teams have an annoying tendency to find a way to lose.

38
by Andrew B :: Wed, 10/15/2008 - 10:15am

Well, they've certainly been losing more close games than they've been winning recently. Here's their record in games decided by 6 points or less (i.e. another touchdown wins it or loses it) including the playoffs.

2002 1-3
2003 4-2
2004 4-1
2005 5-4
2006 3-4
2007 1-5
2008 0-3

I think a lot of the recent struggles to win can also be tied to David Akers missing field goals and Eagles opponents missing almost zero field goals.

In 2006, Eagles opponents were 29 of 30 on Field Goals. 31 of 37 in 2007. 13 of 13 so far in 2008.

The Eagles are 4-10 in games decided by 4 points or less and 3-7 in games decided by 3 points or less since 2006.

Occasional missed field goals by the opponent and made long field goals by Akers would have changed some of those games into wins.

The Original Andrew

41
by Eddo :: Wed, 10/15/2008 - 11:29am

I wonder if there's any way to explain opponent's FG percentage being higher for the Eagles (if I recall correctly, it was high last year, too EDIT: I completely overlooked the figures you gave, it looks like I do recall correctly.).

Maybe the Eagles defense is most likely to stop an opposing offense early in a drive, resulting in a punt, or let them get into the red zone, where their defense clamps down, resulting in an easy field goal?

Maybe, due to their blitz-heavy defense, when an opposing offense has the ball from the 30-yard-line outward (FG-miss-range), they get sacked out of field goal range and are forced to punt?

I guess my question is this: are teams making field goals they should expect to miss against the Eagles? Or are teams only attempting shorter, easier-to-make field goals?

31
by ammek :: Wed, 10/15/2008 - 6:03am

Why is the Cowboys' DVOA for kickoffs so much worse than everyone else's?

51
by Anonymous (not verified) :: Wed, 10/15/2008 - 2:46pm

Because the Cowboys kickoff coverage is THAT much worse than everyne else's.

32
by jimm (not verified) :: Wed, 10/15/2008 - 7:46am

I know the Vikings have been terrible running the ball of late but have they really dropped to 30th in DVOA rushing. If so why does the offensive line rank to high?

54
by Tundrapat (not verified) :: Wed, 10/15/2008 - 4:13pm

I'll go out on a limb here, and suggest that it's due to:

1) the complete and utter predictability of the vanilla offense (do they EVER pass on 1st-and-10?), and;

2) Peterson's newer, more tentative running style. The guy is getting the ball 20+ times a game, but he's not running with the same authority as he showed last season. My guess is that the injury scared him into running softer. Does anyone else see this, or am I just being too pessimistic about my Purple Jesus?

42
by Mac (not verified) :: Wed, 10/15/2008 - 11:30am

Aaron, if your system says that a .500 team is the best team in football, your system is broken.

45
by Wanker79 :: Wed, 10/15/2008 - 12:44pm

DVOA is clearly ranked too high. Everyone knows that judging a team based solely on their W-L record is way better than this. Screw you and your attempt to be more objective than the average douche nozzle who comes to the site to complain about something they clearly don't understand. All Kerry Collins does is win, baby!

55
by Alex51 :: Wed, 10/15/2008 - 5:04pm

Well, I guess somebody should tell the 1980 Raiders and 2001 Patriots to give back their Super Bowl rings. They both started 3-3, so apparently neither of them was the best team in football.

44
by Richie :: Wed, 10/15/2008 - 12:43pm

I do not wish to alarm anybody, but Playoff Odds say that at 96.8%, the Arizona Cardinals have the best chance of making the playoffs of any team in the NFL.

- A Dolphins fan wrote this.

56
by RickD :: Wed, 10/15/2008 - 5:57pm

Is somebody else going to win the NFC West?

The Cardinals also get to play the AFC East. They don't exactly have a tough schedule.

46
by Paul M (not verified) :: Wed, 10/15/2008 - 12:48pm

Mike McCarthy is from a WCO background
He would be under the Paul Hackett branch having worked with him in Pitt and KC

47
by Logan Cash (not verified) :: Wed, 10/15/2008 - 1:22pm

I don't mean to simplify things, but doesn't the variation just have to do with the fact that NFC East has been pretty darn good for the last couple of years. By playing more good teams each year - the Eagles can still have a high DVOA but lose more games.

Had the Eagles been in the NFC West, they would likely average an extra 1.5 wins per season over the past three years. So, that in itself would make up for the discrepancy.

And isn't that the whole point of the DVOA... to show how good a team is independant of the strength of schedule. So we should expect a team in a tough division to have a higher DVOA rank than their record would otherwise indicate.

49
by tuluse :: Wed, 10/15/2008 - 2:36pm

Variance is within the year, not multi-year.

50
by Eddo :: Wed, 10/15/2008 - 2:43pm

I don't think he meant "variance" when he said "variation"; he probably meant "discrepancy".

He'd have a point if other NFC East teams traditionally had the same issues as the Eagles, but they don't. Last year, the Giants' record was better than their DVOA. Ditto for the Cowboys.

53
by Tracy (not verified) :: Wed, 10/15/2008 - 4:06pm

I think Aarons analysis about voa and wco teams is backwards:

The only 16 win point on the chart is orange, and we all know that correponds to the 2007 Patriots, which are not a wco team. So while Arron said that the wco teams are colored orange, the orange points must be non-wco teams. I think the data suggests that voa has no significant bias regarding wco teams, but that if a bias exists, it would tend underrate wco teams, not overrate them.

60
by PhillyFan (not verified) :: Thu, 10/16/2008 - 12:26pm

The only thing I notice about the chart is that WCO teams are both the best (14+ wins, high DVOA) and worst (1-4 wins, very low DVOA). Not sure if it means anything, though.

62
by MdM (not verified) :: Fri, 10/17/2008 - 1:50am

Philly #1--incredible.

In any case, has there ever been a more unachieving team? They are either underachieving W-L, or when they are not, they are losing in inferior teams in the championship game.

We're coised!

63
by Anonymous (not verified) :: Sat, 10/18/2008 - 6:48am

How can Indy have the 10th best rushing offense? Against Houston's #32 rush defense the might Indy ground attack gained 79 yrds. For the year Indy averages 69.40 rushing yards ranking #32. Yards per carry 3.3 ranking #30, but you have them at #10?

No.