Writers of Pro Football Prospectus 2008

Most Recent FO Features

PerineSam15.jpg

» SDA: Bowl Spectacular Part II

The bowl games heat up with ranked teams facing one another, including matchups between Louisville-Georgia, Nebraska-USC, and Clemson-Oklahoma.

23 Dec 2009

Week 15 DVOA Ratings

by Aaron Schatz

New Orleans' first loss of the season knocks them from their perch on top of the Football Outsiders DVOA ratings, but their replacement in the top spot is not the last undefeated team, the 14-0 Indianapolis Colts. Instead, our new number one in both total DVOA and weighted DVOA is the Philadelphia Eagles.

We're all used to this by now, right? How many years now have we had where Philadelphia loses some close games early and it looks like DVOA is rating them much higher than their play deserves? Then they go on a hot streak and conventional wisdom decides the Eagles are one of the most dangerous teams in the league as we enter the postseason. Is there any point in explaining and/or apologizing for it at this point?

Indianapolis has gone 14-0 by excelling in the specific situations that we've found correlate best with wins. Our "estimated wins" metric uses total offense, total defense, and special teams, plus five other splits. Indianapolis is in the top three in the league in four of those five splits: week-to-week consistency, offense in the first quarter, and both offense and defense in the second half of close games. I'm pretty sure that most of the FO staff feels the Colts are really the best team in the league this year, no matter which teams have a slightly higher DVOA rating, and I would include myself as someone who feels that way. However, the fact remains that in the past, total DVOA has been a better indicator of future performance than the "estimated wins" metric.

The Colts have now tied an NFL record by winning seven games by four points or less in one season. The other teams that went to the playoffs based on so many close wins eventually discovered that clutch wins are a cruel mistress. It's sort of ironic how many of them eventually lost by a close score.

  • The 2003 Panthers got hot in the playoffs and made it to the Super Bowl, which they lost by three points.
  • The 1993 Raiders lost in the playoffs by six points.
  • The 1990 49ers are probably the closest comparison to the Colts, because they were an excellent team even when they weren't winning close, and finished the regular season 14-2. They lost the NFC Championship game to New York by two points.
  • The 1986 Browns lost in the playoffs by three points.
  • The 1998 Cardinals lost in the playoffs by a larger score, 20 points.
  • The 1989 Packers missed the playoffs by tiebreaker despite going 10-6.

A lot of the teams with six wins of four points or less also eventually lost by a close score. (I'm not saying this will happen to the Colts, it's just interesting.) The 2008 Colts lost in the playoffs in overtime. The 1999 Titans lost the Super Bowl by seven points (and one yard). The 1984 Broncos lost in the playoffs by a touchdown. However, there is one team which put together six regular-season wins of four points or less and still went on to win the Super Bowl: the 14-2 1986 New York Giants. Like this year's Colts and the 1990 49ers, that team would have had a very good win-loss record even if they had lost a couple of those close games.

Of course, there's a reasonable argument that Peyton Manning is so transcendent as a quarterback that the Colts' close wins are less fluky than, say, the 1998 Cardinals' close wins. I don't mean that sarcastically. The guy's unbelievable. On the other hand, so was the guy playing quarterback for those 1990 49ers, and he couldn't keep Roger Craig from fumbling in the final few minutes of the NFC Championship. It's still a team sport.

Two big changes in the stats pages this week. First, we've fixed a bug on the playoff odds report page that was switching home and visitor during wild card games. That was causing prospective fifth/sixth seeds to come out with better Super Bowl odds than prospective third/fourth teams. Second, the offensive line and defensive line pages now include the new method for marking long runs, with Second Level Yards per carry and Open Field Yards per carry on both offense and defense.

Also, to make up for this week's ratings being a day late, I'll put together a nice "what I learned from Jaws and Greg Cosell" post later today.

* * * * *

These are the Football Outsiders team efficiency ratings through 15 weeks of 2009, 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. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. WEIGHTED DVOA is adjusted so that earlier games in the season become gradually less important. It better reflects how well the team is playing right now. 

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

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
WEIGHTED
DVOA
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
1 PHI 35.4% 2 33.1% 1 10-4 17.0% 9 -12.1% 4 6.4% 3
2 BAL 31.0% 5 29.1% 2 8-6 17.0% 10 -12.8% 2 1.2% 12
3 NO 29.4% 1 23.8% 7 13-1 30.2% 1 -0.4% 12 -1.2% 25
4 IND 27.9% 3 25.8% 4 14-0 25.6% 3 -1.6% 9 0.6% 17
5 NE 27.5% 4 26.2% 3 9-5 29.2% 2 3.6% 18 1.9% 10
6 GB 24.5% 6 23.8% 6 9-5 21.4% 6 -10.5% 5 -7.5% 32
7 DAL 22.8% 9 25.1% 5 9-5 25.1% 4 2.9% 17 0.7% 16
8 MIN 18.6% 7 17.9% 10 11-3 14.0% 12 2.0% 16 6.6% 2
9 NYG 16.1% 13 14.0% 13 8-6 17.1% 8 -0.7% 10 -1.6% 28
10 DEN 15.5% 10 11.2% 14 8-6 3.4% 19 -12.5% 3 -0.4% 21
11 ARI 14.5% 8 16.8% 11 9-5 11.2% 13 -0.1% 13 3.2% 5
12 PIT 13.7% 12 15.7% 12 7-7 19.1% 7 0.7% 15 -4.7% 30
13 SD 13.7% 14 21.3% 8 11-3 23.0% 5 9.7% 23 0.4% 18
14 HOU 12.3% 11 19.4% 9 7-7 15.7% 11 5.4% 21 2.1% 9
15 NYJ 8.8% 15 4.2% 18 7-7 -11.5% 23 -18.6% 1 1.8% 11
16 CIN 5.8% 17 3.4% 19 9-5 6.4% 18 -0.7% 11 -1.3% 26
TEAM TOTAL
DVOA
LAST
WEEK
WEIGHTED
DVOA
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
17 MIA 5.2% 16 6.6% 17 7-7 6.7% 17 3.8% 19 2.3% 7
18 TEN 0.1% 21 7.2% 16 7-7 9.5% 15 8.5% 22 -0.9% 23
19 JAC -0.3% 19 -3.3% 22 7-7 10.4% 14 10.0% 24 -0.7% 22
20 SF -1.0% 18 1.8% 20 6-8 -9.6% 22 -8.4% 6 0.2% 19
21 ATL -2.1% 22 -6.1% 23 7-7 8.7% 16 10.4% 25 -0.4% 20
22 CAR -3.1% 23 9.9% 15 6-8 -3.9% 21 -5.0% 7 -4.1% 29
23 WAS -3.8% 20 -0.2% 21 4-10 -2.3% 20 0.6% 14 -0.9% 24
24 BUF -12.5% 24 -13.5% 24 5-9 -18.9% 27 -3.3% 8 3.1% 6
25 CHI -25.3% 25 -27.8% 27 5-9 -24.1% 31 5.4% 20 4.2% 4
26 SEA -27.4% 26 -33.0% 29 5-9 -14.0% 24 14.1% 29 0.7% 15
27 CLE -27.8% 27 -23.2% 25 3-11 -14.8% 25 21.9% 32 8.9% 1
28 TB -28.0% 29 -26.3% 26 2-12 -17.2% 26 13.0% 28 2.2% 8
29 OAK -31.1% 30 -28.8% 28 5-9 -21.7% 29 10.5% 26 1.2% 13
30 KC -32.6% 28 -36.1% 30 3-11 -19.8% 28 11.4% 27 -1.4% 27
31 STL -42.0% 31 -39.0% 31 1-13 -23.2% 30 19.7% 30 0.9% 14
32 DET -51.7% 32 -52.3% 32 2-12 -26.9% 32 19.7% 31 -5.0% 31

  • 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.
  • 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 VAR. RANK
1 PHI 35.4% 10-4 39.8% 10.2 4 -2.2% 27 18.1% 24
2 BAL 31.0% 8-6 33.1% 10.3 3 -0.8% 23 14.3% 16
3 NO 29.4% 13-1 33.1% 10.5 2 -2.0% 26 13.3% 14
4 IND 27.9% 14-0 30.4% 10.7 1 3.2% 12 7.4% 3
5 NE 27.5% 9-5 23.2% 9.5 6 4.9% 8 20.9% 30
6 GB 24.5% 9-5 33.5% 9.1 9 -9.5% 32 14.7% 17
7 DAL 22.8% 9-5 21.5% 9.7 5 1.5% 16 6.8% 2
8 MIN 18.6% 11-3 24.4% 9.2 7 -7.7% 31 9.6% 7
9 NYG 16.1% 8-6 8.5% 8.1 14 5.9% 5 16.5% 23
10 DEN 15.5% 8-6 15.6% 8.6 12 3.0% 13 18.4% 26
11 ARI 14.5% 9-5 18.2% 9.2 8 -6.9% 30 21.8% 31
12 PIT 13.7% 7-7 21.2% 8.6 13 -5.4% 29 12.6% 13
13 SD 13.7% 11-3 14.9% 9.0 10 0.4% 20 6.2% 1
14 HOU 12.3% 7-7 14.5% 8.7 11 -2.0% 25 8.5% 5
15 NYJ 8.8% 7-7 12.3% 7.4 17 1.2% 17 15.2% 19
16 CIN 5.8% 9-5 8.1% 7.8 15 0.7% 19 18.1% 25
TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK PAST
SCHED
RANK VAR. RANK
17 MIA 5.2% 7-7 3.0% 7.6 16 5.7% 7 9.9% 8
18 TEN 0.1% 7-7 -8.2% 7.2 19 6.3% 4 36.0% 32
19 JAC -0.3% 7-7 -1.3% 7.2 18 -0.4% 21 20.0% 28
20 SF -1.0% 6-8 -0.2% 6.6 23 1.6% 15 11.4% 12
21 ATL -2.1% 7-7 -6.5% 7.2 20 7.4% 2 10.6% 9
22 CAR -3.1% 6-8 -8.4% 6.8 21 5.7% 6 16.1% 21
23 WAS -3.8% 4-10 -3.3% 6.7 22 -1.3% 24 15.0% 18
24 BUF -12.5% 5-9 -12.2% 6.0 24 2.2% 14 13.4% 15
25 CHI -25.3% 5-9 -25.5% 4.4 25 1.1% 18 11.3% 11
26 SEA -27.4% 5-9 -21.7% 3.5 29 -5.4% 28 20.7% 29
27 CLE -27.8% 3-11 -32.9% 4.2 26 3.4% 11 16.4% 22
28 TB -28.0% 2-12 -31.2% 4.0 28 7.9% 1 15.6% 20
29 OAK -31.1% 5-9 -35.8% 4.0 27 6.9% 3 19.0% 27
30 KC -32.6% 3-11 -30.3% 3.4 30 3.7% 9 7.9% 4
31 STL -42.0% 1-13 -40.2% 2.8 31 -0.6% 22 10.9% 10
32 DET -51.7% 2-12 -52.2% 2.1 32 3.6% 10 9.6% 6

Posted by: Aaron Schatz on 23 Dec 2009

200 comments, Last at 27 Dec 2009, 11:40am by tuluse

Comments

1
by vesini :: Wed, 12/23/2009 - 12:02pm

First?

The Eagles back on top - again - Does this mean I can't make fund of Andy Reid anymore?

vesini, who did not use the proper stats, and is dead.

2
by crack (not verified) :: Wed, 12/23/2009 - 12:02pm

I assume you mean knocks them from the top of the DVOA rankings.

If the Lions weren't knocked out last year no one should be.

3
by jmaron :: Wed, 12/23/2009 - 12:03pm

"The 1998 Cardinals lost in the playoffs by a larger score, 13 points."

they actually lost by 20 to the Vikings. Viking playoff wins are a rare event so I felt I needed to point out the error.

4
by Aaron Schatz :: Wed, 12/23/2009 - 12:06pm

Heh. Thanks, I'll fix that.

5
by Erithtotl (not verified) :: Wed, 12/23/2009 - 12:06pm

I bet they are sure glad that it's Phily at the top spot so they can talk about the perennial DVOA favorite and not mention Baltimore sitting at #2 with their sparkling 8-6 record. Was a 31-7 win over Chicago really THAT impressive?

7
by ajdfsjlkafsd (not verified) :: Wed, 12/23/2009 - 12:10pm

If you haven't figured out by now that W/L record isn't the best determinant of who is a better team, you should probably give up on this site.

12
by Vince Lombardi (not verified) :: Wed, 12/23/2009 - 12:21pm

Winning isn't everything. I should stop right there.

8
by Aaron Schatz :: Wed, 12/23/2009 - 12:13pm

The Ravens' six losses have all come against playoff teams, and four of them were by less than a touchdown. However, they do have an interesting habit of playing close against other good teams and beating the holy hell out of bad teams, and it is possible the ratings like that a bit too much. I'll write about the Ravens more next week, once we see what they do against the Steelers. If they have a big win, they might move into the top spot.

13
by Hurt Bones :: Wed, 12/23/2009 - 12:22pm

Highest average margin of Victory
1. Baltimore 20.1
2. Seattle 18.8
3. New England 17.1
4. New York Jets 15.6

Lowest average margin of defeat
1. Pittsburgh 4.0
2. Dallas 5.8
3. Baltimore 6.0
4. New England 6.6

21
by Temo :: Wed, 12/23/2009 - 12:37pm

DVOA: marginalized by Pythagorean expectation since 1984!

/not really true
//Bill James forever

73
by John (not verified) :: Wed, 12/23/2009 - 2:32pm

Lowest average margin of defeat
1. Indianapolis 0.0
2. Pittsburgh 4.0
3. Dallas 5.8
4. Baltimore 6.0
5. New England 6.6

There, fixed that for you.

77
by Arkaein :: Wed, 12/23/2009 - 2:41pm

You didn't fix anything, Indy's average margin of defeat is undefined:

0 point margin in all losses / 0 losses = NaN (not a number)

Mathematically, this is no closer to zero than it is to positive or negative infinity.

20
by tonic889 (not verified) :: Wed, 12/23/2009 - 12:35pm

Hell, I'm a Ravens fan and I think that ranking is too high. A few weeks ago, our defense had given up the most plays over 20 yards in the league. I suppose DVOA would claim that most of those plays happened on 3rd and 35??

The rating reminds of the Eagles from last year. A team with an average record but high DVOA due to losing close games and killing bad teams. Then again, the Eagles went on to make the NFC Championship game.

captcha: emanate tumor. That's scary.

24
by Temo :: Wed, 12/23/2009 - 12:43pm

The Ravens are clearly ranked too high because the defense has given up the most plays over 20 yards in the league. Number of big plays (defined as above an arbitrary amount of yards) is way better than this.

Just saying, you probably need a better reason for your DVOA being too high other than a single cherry picked stat.

28
by Hurt Bones :: Wed, 12/23/2009 - 12:52pm

Combined with the "a couple of weeks ago" qualifier. Is that 2 weeks? 4? 8? The Ravens have given up some big pass plays but many were were in the first half of the season. Since Nov. 1 the Ravens are only allowing 11.9 points per game, tops in the league. There is a chance they could catch the Jets (current scoring defense leaders) in the next two weeks.

38
by tonic889 (not verified) :: Wed, 12/23/2009 - 1:09pm

So arbitrary dates (like Nov. 1) are OK? But arbitrary stats are not? Ok. I need to stay off these threads.

43
by Hurt Bones :: Wed, 12/23/2009 - 1:16pm

Except that the arbitrary Nov. 1 reflects the last 8 weeks or the most recent performance of the team. And though arbitrary gives an idea of when the defense started to improve.

20 yards is just a round number. Why not 19 yards or 21 yards?

69
by Big Johnson (not verified) :: Wed, 12/23/2009 - 2:29pm

20 yards is just a round number. Why not 19 yards or 21 yards?

How bout 9 games or 7 games? Its like when the announcer says "team A has won their last 3 home games" I always think yeah well how did they do in their 4th. Is their any correlation to 8 that makes it more relevant than 7 or 9?

114
by tonic889 (not verified) :: Wed, 12/23/2009 - 4:02pm

Here is the Passing Offense DVOA of Ravens opponents.

KC -17
SD 68.1
CLE -26
NE 56
CIN 26
MIN 39.9
Pre Nov. 1 average: 24.55

DEN 20.6
CIN 26
CLE -26
IND 48.3
PIT* 40.9
GB 34.9
DET -33.4
CHI -20.9

I asteriked PIT because Dennis Dixon was the QB for that game.
Post Nov. 1 average (incl. PIT) : 11.29%
Post Nov. 1 average (w/o PIT) : 7.06%

I'm not claiming that plays over 20 yards is The One True Metric of pass defense. But what I'm seeing is that Ravens Pass D had a much softer schedule post Nov. 1. So you'll forgive me if I laugh out loud at the claim that Nov. 1 is any less arbitrary than 20 yards, or that the Ravens D has improved in the second half of the season.

118
by Eddo :: Wed, 12/23/2009 - 4:17pm

I'd say your data suggests that the apparent decline of the Ravens' defense in the first half of the season was more due to them playing, on average, very good passing offenses, and that playing a more average (yet still on the good side of 0% DVOA) schedule in the second half has shown more of their true strength.

135
by jonnyblazin :: Wed, 12/23/2009 - 6:01pm

Or you could notice that their pass defense improved once Ladarius Webb replaced Fabian Washington in the starting lineup. Or that defensive coordinator switched from man to zone pass coverage after it was obvious that they were struggling.

Of course, that would require that you actually watched the games before forming an opinion.

134
by Hurt Bones :: Wed, 12/23/2009 - 5:59pm

In the first 6 games of the year the Ravens played teams that have averaged 22.345 points per game and the Ravens allowed 21.667 points per game in those contests. After the bye, the teams they played have averaged 20.875 points per game and they have allowed 11.9 points a game in those contests.

It may just be me but cutting the number of points you allow nearly in half is improvement in my book.

185
by crack (not verified) :: Thu, 12/24/2009 - 11:24am

Until you learn that the most important thing in the NFL is to have the highest DVOA there is no reason to engage the commentators in discussion.

97
by Kevin from Philly :: Wed, 12/23/2009 - 3:29pm

Put a "li" in there, and you're OK.

57
by Will Allen (not verified) :: Wed, 12/23/2009 - 1:36pm

For what it is worth (which may not be much), my intuition tells me that stomping crappy teams this year isn't as strongly correlated with quality play as it has been in year's past.

59
by Still Alive (not verified) :: Wed, 12/23/2009 - 1:45pm

Here is a repost from Advanced NFL stats comments on this very issue:

-Anonymous idiot
"How are the Redskins (4-10) ahead of the Bengals (9-5)?"

Because not only are wins and losses not the main thing going into these rankings, they do not actually appear in them at all. The factors used to make these rankings are right there on the page for you to look at...

Before you say "but wins and losses are what determine how good a team is!". You might also ask how my fantasy team could be ranked ahead of its opponents in all my friends' minds when it finished 7-7 and lost the first round of playoffs. If you are sophisticated and look beyond W-L at the actual statistics that determine how good a team is in FF (scoring) you would see that I ranked 2nd in scoring and was very consistent (great signs of success), but had an abnormally high pts against me scored (bad luck). This is how I can have the best or second best team and finish in a three way tie for 4-6th, and how WAS can finish above CIN in these rankings.

WAS plays in the toughest division in the league.

72
by zlionsfan :: Wed, 12/23/2009 - 2:31pm

please stop talking about my fantasy team. It scares me when you do that.

I finished 6-7 and lost in the first round ... I was a close 3rd in scoring (lost a spot on the last week), also reasonably consistent (I had the highest "all play" record in the league), but "gave up" the most points in the league.

82
by Still Alive (not verified) :: Wed, 12/23/2009 - 2:54pm

I have managed to be 1 or 2 in scoring in my main league for 5 years running, and so far my results have been something like .500, .500, undefeated, undefeated, .500. Of course I always manage to lose in the playoffs :( .

You can tell who the non-statistically savvy people are in the league because they tend to think those two undefeated teams were epic and don't at all remember the .500 ones despite the fact they were all very similar in strength compared to the league as a whole.

168
by RickD :: Thu, 12/24/2009 - 12:08am

Here's a better answer: it's not a perfect system.

It's a better argument than trying to argue that Washington really is a better team than Cincinatti.

(And no, the Ravens really aren't the 2nd best team in the NFL.)

6
by yoyodyn (not verified) :: Wed, 12/23/2009 - 12:08pm

'On the whole, I'd rather be in Philadelphia.'
~DVOA Computer

83
by Bruce G. (not verified) :: Wed, 12/23/2009 - 2:56pm

'Yes pHiladelpHia does feel good, on the (w)hole.'

9
by Danish Denver-Fan :: Wed, 12/23/2009 - 12:14pm

So the Eagles generally have a steady DVOA, but loose to cupcakes early in the season? Have the Eagles been connsistently inconsistent over the years? This could help explain some of those mindboggling losses to inferior opponents.

17
by Phil Osopher :: Wed, 12/23/2009 - 12:29pm

My friend was hollering about how they were going to lay an egg before the Oakland game. I thought that was ridiculous as JaMarcus was playing.

Philly laid an egg.

He said they always do this and that now was the time they were going to do it this year. I know stupid anecdote, but he was fired up about it and then they proceeded to lay an egg. humorous

42
by Rick (not verified) :: Wed, 12/23/2009 - 1:14pm

Also, to be fair to Philly, the big reason they lost in Oakland was their O-Line. At that point in the season the OL was in shambles and Jason Peters (their big $ LT) had left the game with an injury. The Eagles put some scrub 7th round rookie in at LT against Richard Seymour, so you do the math. McNabb got owned all day long.

65
by Tundrapaddy (not verified) :: Wed, 12/23/2009 - 2:10pm

And you haven't even touched on the creative playcalling, or lack thereof.

A tight spiral is Andy Reid's heroin.

26
by Nathan :: Wed, 12/23/2009 - 12:51pm

So the Eagles generally have a steady DVOA, but loose to cupcakes early in the season?

And the boy wins a cigar. Anyone who watches the Iggles on a regular basis will confirm this is their MO. They play to the level of their opponent. The Oakland loss was not shocking at all.

31
by DaninPhilly (not verified) :: Wed, 12/23/2009 - 12:56pm

I think the Andy tries out a lot of things in the first third of the season, to see what works and what doesn't. Sometimes this gets him into trouble later, but I have noticed that he doesn't do as many dumb things later in the year, and the wackiness you see early has pretty much gone by week 12 or so. That's my theory, anyway.

102
by Kevin from Philly :: Wed, 12/23/2009 - 3:42pm

Part of it is probably getting the free agents and draft picks used to the system. The Eagles Offense and Defense are supposedly both among the more complicated in the league. I'll leave it to you to decide if it's too hard for Andy to grasp, as well.

163
by Scott C :: Wed, 12/23/2009 - 11:25pm

Michael Vick has 4 plays down or so, so it can't be that complicated.

10
by Chief too :: Wed, 12/23/2009 - 12:17pm

I'd like to see a ranking based on variance as well as total DVOA. (I know estimated wins incorporates both, but it also incorporates other elements, like red-zone DVOA, IIRC.)

If a high total DVOA-high variance team (like the Eagles) would be expected to win fewer games against the same schedule and fewer games head-to-head against a team with a slightly lower total DVOA and a much lower variance (like Indianapolis or New Orleans), I wouldn't consider them better. Consistency matters, and I think it goes a long way toward reconciling DVOA with people's intuitive rankings.

Also, is Baltimore's ranking even more surprising than the Eagles?

22
by Danish Denver-Fan :: Wed, 12/23/2009 - 12:41pm

Seconded. It should be possible to run a head-to-head simulation 10k times (Monte Carlo simulation?), and try to analyze with what margin the games were decided. Maybe not as a permanent stat, but maybe some tidbits to analyze come playoff time - possibly even improve the betting portion of the premium package.

I will say though: Cosistency matters, but it's not always better to be consistent. I root for the Broncos who are very inconsistent. I don't mind this, 'cause the Broncos are almost certainly going to face a superior opponent at some point in the playoffs (if they make it ~= 73%). Lets say they will face Indy. If both teams had a variance of 0%, Indy would basically win every time. But because the Broncos are so inconsistent, they have a, say, 30% chance of playing to their maximum abillity and win.
If my team is mediocre i like the inconsistency cause that, ultimately, improves the chance of a championship. It is also raising the risk of going 6-10, but who cares right?

58
by MJK :: Wed, 12/23/2009 - 1:36pm

Consistency is good if you're good, bad if you're bad. Variance is bad if you're good, and good if you're bad. It's the phenomenon as why a superior team wants to extend a game and have more plays and possessions for both sides, but an inferior team wants to shorten the game and minimize opposing possessions. That's the real reason why you like to keep Manning or Brady or Brees on the sideline with ball-control offennse. It's not because less time of possession magically makes them less likely to socre on any given possession--it's because Manning/Brady/Brees led teams tend to be good teams, and against a better team, you want both sides to have fewer possessions so that a lucky bounce your way has a chance to win you the game.

186
by crack (not verified) :: Thu, 12/24/2009 - 11:28am

Very well said.

11
by Keith (1) (not verified) :: Wed, 12/23/2009 - 12:17pm

Have there been other years where two teams so disparate as the Colts and Packers end up both top-10 in DVOA on both offense and defense? That seems weird that they each hold those spots and yet there are five wins difference between them.

14
by Rick (not verified) :: Wed, 12/23/2009 - 12:23pm

Ok now I know you put a disclaimer in the first paragraph about this, but it really is starting to get ridiculous. Every year an average Philly team climbs to the top of the charts of this ranking system and then barely makes the playoffs or misses the playoffs even after a season of really poor play.

I've watched some of the recent Philly games and they are a decidedly mediocre team. Anyone with football knowledge will tell you that. They barely skated past a few bad opponents like Chicago, Washington, San Francisco, and they lost ALL of their games to the good teams: they got dominated by Dallas, blown out by San Diego, blown out by New Orleans. Their defense is a sieve, and their offense is highly inconsistent.

I know the record doesn't always match the true performance of a team, but after watching alot of the games this year I have to seriously question this DVOA system if it ranks the Eagles above clearly superior teams like New Orleans, Minnesota, Indy, and San Diego.

16
by Aaron Schatz :: Wed, 12/23/2009 - 12:29pm

Um... they lost to Dallas by four and San Diego by eight. That doesn't really qualify as "domination" or "blowout." The New Orleans blowout was a blowout, but it was also three months ago and it featured their backup quarterback.

23
by Rick (not verified) :: Wed, 12/23/2009 - 12:43pm

The game against Dallas was pretty convincing domination. They could barely move the ball and score points at all. The game against San Diego was a complete blowout until the 4th quarter when SD was playing PREVENT DEFENSE and allowed the Eagles to score some garbage time points. SD has a habit of doing this, but it was clear througout that they were the superior team. And backup quarterback or not, the New Orleans offense GOUGED the Eagles for 48 points.

45
by mfalc (not verified) :: Wed, 12/23/2009 - 1:18pm

The Eagles moved the ball just fine against the Cowboys. If you actually 'watched' the game, you would have realized the officiating decided that game. As for the New Orleans game, Kevin Kolb and Ellis Hobbs made sure that 21 points out of the 'GOUGING' came from a total of 58 gained yards.

Do your research before jumping all over a team. Or, actually watch football.

48
by Temo :: Wed, 12/23/2009 - 1:22pm

the officiating decided that game

No. Close short yardage plays and key penalties at importune moments went against the Eagles decided that game.

193
by Andrew B :: Thu, 12/24/2009 - 3:53pm

Temo: "Close short yardage plays"

Rick: "They couldn't convert 3rd & short and 4th & short WORTH A DAMN because they got stuffed consisently by the Dallas D-Line"

Officiating had nothing to do with it?

"I think I'd move it. There was a camera angle that showed exactly where the ball was when McNabb's elbow hit the ground. The Eagles logo at midfield provided a reference point. The spot where the ball was when McNabb's elbow touched the ground did not match the placement of the ball by the referees." (Mike Pereira, VP NFL Officiating)

Or, maybe Mike just has no idea what he is talking about, and the experts Temo and Rick do?

The Original Andrew

63
by Rick (not verified) :: Wed, 12/23/2009 - 2:01pm

Um, the Eagles scored 16 points against the Cowboys. McNabb threw 2 INTs. They couldn't convert 3rd & short and 4th & short WORTH A DAMN because they got stuffed consisently by the Dallas D-Line, penalties not withstanding. I think that might qualify as "couldn't move the ball". no?

If YOU had watched the New Orleans game you would realize that Kolb / Hobbs were not the problem. Sure, Kolb throwing that INT at the end of the first half was a critical mistake, but the problem was that the Philly defense could do NOTHING to stop Brees. THey put no pressure on him, and when they did he counteracted it effectively. Even with McNabb at the helm, Brees would have carved up the Philly defense with ease and the Eagles would still have lost, but maybe by not as much.

Also you failed to address the fact that SD totally owned Philly in almost every way.

88
by Andrew B :: Wed, 12/23/2009 - 3:08pm

"Um, the Eagles scored 16 points against the Cowboys. McNabb threw 2 INTs. They couldn't convert 3rd & short and 4th & short WORTH A DAMN because they got stuffed consisently by the Dallas D-Line, penalties not withstanding. I think that might qualify as "couldn't move the ball". no?"

Or MAYBE, it just had to do with field position, where the Cowboys started 5 of 11 drives beyond their 35, while the Eagles started just 1 drive beyond their 25 (the Eagles field position netted out to 143 yards of additional field to cover to get a touchdown); and the refs not being able to spot the ball correctly on a 3rd and 1 and a 4th and 1, as pointed out by the announcers and as anyone could see on TV, even after two Philadelphia challenges to get the spot corrected.

"If YOU had watched the New Orleans game you would realize that Kolb / Hobbs were not the problem."

Really? So Hobbs kick off return fumble on th ePhilly 22 to open the 3rd quarter, Kolb's interception at the Philly 24 on the next possession, and his pick 6 to end of the game were not what caused Philly to lose?

"Sure, Kolb throwing that INT at the end of the first half was a critical mistake"

Was that interception at the end of the first half right before or right after David Akers kicked a field goal as time in the 2nd quarter expired? Just curious, since you watched this game so intently.

"the problem was that the Philly defense could do NOTHING to stop Brees."

I see. Does that nothing include forcing 3 punts, stopping two drives for field goals, an interception, a fumble out of the end zone for a safety, 1 sack, 1 sack fumble (recovered by NOL), and 6 tackles for a loss?

The whole thing is pretty confusing, because if Kolb doesn't throw the 2 interceptions and Hobbs doesn't fumble the kick off return, NOL loses 21 points. And if Kolb instead got just a field goal off the drive he threw the 97 yard pick 6 on, if the Eagles had recoved Brees' sack-fumble on the NOL 16 instead of Shockey and got a field goal off that, the score would be 27 Saints-28 Eagles. So 4 plays changed the game from a close Eagles win to a huge Saints blow out. Kind of confusing thinking that Kolb and Hobbs, personally responsible for 24 points of the net point differential, weren't the focus of problems in the game.

The Original Andrew

162
by mm (not verified) :: Wed, 12/23/2009 - 11:24pm

Does that nothing include forcing 3 punts, stopping two drives for field goals, an interception, a fumble out of the end zone for a safety, 1 sack, 1 sack fumble (recovered by NOL), and 6 tackles for a loss?

The fumble out of end zone for a safety was intentional to give more space for a punt with the score already a blowout.

The Eagles' defense did pretty well for almost the entire first half (holding the Saints to 10 points), but then collapsed, giving up consecutive TD drives of 54, 22, and 24 yards, followed by a 70 yard field goal drive.

On their next drive, the Saints started to run out the clock (the 3rd quarter wasn't quite over yet). So the Eagles stopped the Saints for a punt, but on the next possession gave up a TD drive while the Saints were playing 'run out the clock'! (The drive covered 52 yards, with 5 running plays and a 5 yard pass to the backup tight end). The last time the Saints got the ball, they did indeed stop them again, with the Saints deliberately taking that safety on 4th down.

187
by Andrew B :: Thu, 12/24/2009 - 12:31pm

"The Eagles' defense did pretty well for almost the entire first half (holding the Saints to 10 points), but then collapsed, giving up consecutive TD drives of 54, 22, and 24 yards, followed by a 70 yard field goal drive."

Okay, so you've looked up the drive charts!

Did you notice that the 22 yard drive for a TD came after a kick off return fumble by Hobbs to open the half that gave the Saints the ball in field goal range? Or that on the very next Eagles drive, Kolb decided that the Saints needed another posession and tossed them the ball again so that they again could start a drive in field goal range?

Did the Eagles defense use their super ESP mindpowers to make Kolb and Hobbs screw up this way?

Its to be expected that the Saints would score some points on other drives. I don't have a problem with the defense not being able to shut them out. But giving them the ball with 20 yards to score is inexcusable.

The Original Andrew

192
by mm (not verified) :: Thu, 12/24/2009 - 3:29pm

Okay, so you've looked up the drive charts!

I saw the game, but looked up the drive charts so I could specifically show how the 2nd half went. The fact that you listed a fumble out of the end zone as a defensive 'accomplishment' for the Eagles lead me to wonder whether you had watched the game.

Did you notice that the 22 yard drive for a TD came after a kick off return fumble by Hobbs to open the half that gave the Saints the ball in field goal range? Or that on the very next Eagles drive, Kolb decided that the Saints needed another posession and tossed them the ball again so that they again could start a drive in field goal range?

What, you mean the refs didn't decide to give the Saints the ball there out of the goodness of their hearts? Of course the offense and special teams of the Eagles weren't playing well in the second half, but the Eagles defense was probably worse than the offense. They couldn't even stop them for a field goal until the game was far out of hand, and they gave up a long touchdown drive while the Saints were trying to run out the clock!

If the 2 teams play again in the playoffs, I expect a much closer game, but that still doesn't change the fact that the Eagles defense had a bad day in week 2.

95
by Rick A. (not verified) :: Wed, 12/23/2009 - 3:26pm

What you say is true. No question. But it is also true that Philly has done this just about every year in the Reid/McNabb era--i.e. fail to be fully focused until about game 11 on. You will see alot better rush from them from here on, and a better push from their bigs. Reid's entire m.o. is to go for the early lead, his whole philosophy is based around that. When they do not accomplish that in the game, earlier in the season, they are often a bit lost the rest of the game, become a bit dis-spirited. It won't be that now, even if they fail to get a big lead. They will again be highly dangerous, and teams sprouting leaks on their O Line at just the wrong time, such as Minny, will be in trouble facing a far more fierce rush from them. (or anyone with a focused pass rush, like Dallas e.g.)...

96
by Rick A. (not verified) :: Wed, 12/23/2009 - 3:27pm

I was replying to Rick, not The Original Andrew...

170
by RickD :: Thu, 12/24/2009 - 12:13am

You are replying to Rick, as am I. As will be the guy who replies to me.

49
by randplaty (not verified) :: Wed, 12/23/2009 - 1:22pm

The Dallas game was close, but the SD game was a blowout even if your metrics indicate it wasn't. That game was never in doubt from the first quarter.

http://wp.advancednflstats.com/nflarchive.php?year=2009&team=PHI&gameid=...

55
by Chief too :: Wed, 12/23/2009 - 1:33pm

Actually, I'd say that the Eagles' 23 percent chance of winning with 7:18 left is pretty reasonable doubt, all things considered. The Chargers' long drive after that came pretty close to sealing it (but then the defense let the Eagles climb back to a 16 percent chance with 13 seconds left). That game was pretty close to being a massive choke by San Diego.

People can blame "prevent defense" all they want, but the Chargers need to slam the door shut.

130
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 5:26pm

People can blame "prevent defense" all they want, but the Chargers need to slam the door shut.

If the Chargers were playing the Eagles soft when they allowed a TD that put the Eagles 5 points down, they were brain-dead stupid. This bizarro-idea that "oh, the Eagles had no chance because if they had stopped the Chargers, the Chargers defense would've prevented them from scoring and they still would've won" is completely crazy. If Rivers throws a pick-6 instead of that pass to Nanee, the Chargers defense couldn't've done a darn thing and the Eagles could've been up by as much as a field goal with little time left.

I could maybe buy the first TD was against a prevent defense, but no defense in their right mind would run a prevent with 12 minutes remaining and only up 12 points. At that point, the opposing offense has plenty of time to run two dink/dunk drives all the way down the field.

142
by Marver :: Wed, 12/23/2009 - 7:11pm

The Chargers were, in fact, brain-dead stupid in the fourth quarter of that game. There have been two scream-at-the-television moments for Chargers fans this season: Running the ball against the Ravens on the final drive, and the fourth quarter of the Eagles game. It may not have been 'prevent', but it most certainly was very soft zone defense.

I agree, the game wasn't a 'beat-down', but the Chargers won the game pretty soundly. They outgained the Eagles per pass 7.9 to 7.5, per run 4.4 to 2.2, won the turnover battle 1-0, and committed less penalties (3/31 to 9/70). If you were forced to choose who the better team is based entirely on that game alone, you'd have an anonymous answer.

149
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 8:43pm

It may not have been 'prevent', but it most certainly was very soft zone defense.

It really didn't seem any different from the remainder of the game to me - it's not like the Eagles were having any problem driving the ball down the field prior to that. They were really only having problems in the red zone after the first-quarter disaster of "let's try gimmicky crap and just try chucking it downfield."

I agree, the game wasn't a 'beat-down', but the Chargers won the game pretty soundly.

I really, really strongly want to point out that I never thought the Chargers didn't win the game solidly. But for the game to be within 1 score, and it being a home game for the Chargers, it's hard to claim that the Eagles looked incapable of beating the Chargers.

They outgained the Eagles per pass 7.9 to 7.5, per run 4.4 to 2.2,

The yards/offensive play was in the Eagles favor, so it's hard to really claim that the Chargers were strictly better on offense/defense. The Eagles passed more, but that's because they could, and it yields better returns anyway.

won the turnover battle 1-0

Yeah, if you want to claim that the Eagles' yardage was boosted due to a soft zone defense, that last interception has to be discounted similarly.

157
by Marver :: Wed, 12/23/2009 - 9:45pm

I threw the interception in there as a subtle, 'look how meaningless some interceptions are' hint at the other thread in which we were arguing.

I think we're on the same page here. Any Chargers fan who says 'we ANNIHILATED the Eagles' is about as dilusional as any Eagles fan who says 'we almost won that one'. It required at least another drive (in either argument) with the best possible outcome for that to be close to the truth.

However, I do think we should stop calling an 8 point lead "one score". One possession, maybe. But 8 points requires the best possible outcome, followed by converting something with worse odds than calling a coin-flip; there's a better chance it's a two score game than a one score game.

90
by Andrew B :: Wed, 12/23/2009 - 3:18pm

So the game wasn't in doubt with the Eagles down 5 with 5 minutes to play? Because no NFL team has ever come back from a 5 point deficit with 5 minutes left in the game, right?

Was the game in doubt at 14-6 at the half, with the Eagles having barely been held short of the end zone? I mean, because no team in NFL history held out of the end zone in the first half and down only 8 ever scores touchdowns in the second half, right?

The Original Andrew

122
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 4:45pm

Please, for the love of God, define "doubt."

Are you saying "less than 50% of the time, a team wins from this state" (i.e. less than 0.5 WP)? That's a stupid definition - look at BAL-MIN. That game was never 'in doubt' by that definition from the first quarter to 6 minutes left in the game - and Minnesota won that game by a missed field goal by the Ravens.

If the Ravens don't stop the Vikings on the second-to-last drive, that game looks almost exactly the same as the Eagles-Chargers game.

Also, please don't take the Advanced NFL stats WP calculator too seriously. It's missing at least one obvious game state which changes the results fairly noticeably - namely, the fact that the Eagles got the ball first in the second half.

The WP predictions are based on historical results of games in the same state - which means in half of those games, the score was 14-6 at the half, with the other team getting the ball first. That's dramatically different than the situation in the actual game, where it was 14-6 and the Eagles got the ball with a chance to tie the game, which means the WP behavior in the entire first half is lower than it should be.

But still, saying that having a 30% chance of winning a game is "not in doubt" is... bizarre. In any given week, ~5 teams who have a 30% chance to win a game would win!

196
by randplaty (not verified) :: Fri, 12/25/2009 - 3:48pm

So by "doubt" it seems like you mean less than 5%. Well if that's your definition I concede it. The point of the wp calculator is to show how San Diego dominated that game. If a team only has a 30% chance of winning, I would assert that's domination. That's a significantly better team at that moment. That's the point.

I'm not sure what you mean when you say that Philly getting the ball in the second half means that I shouldn't take the wp calculator seriously. The wp calculator charted the entire game not just the wp of the first half.

The fact remains that the SD-Philly game showed that SD was significantly better that week and DVOA and Aaron both seem to be discounting that fact. I would argue that you shouldn't take DVOA seriously until it fixes this.

198
by Eddo :: Fri, 12/25/2009 - 9:19pm

I'd say the who-has-the-ball-to-start-the-second-half issue is pretty huge, especially when it's a one-to-two-score game at halftime.

It seems like this wouldn't be that difficult for Burke to model, though.

18
by Jeff (not verified) :: Wed, 12/23/2009 - 12:33pm

You've gotta come with more to the table besides "you've watched a lot of their games this year", Rick, that's really weak. I've watched a lot of Philly games too, and I see a team that moved the ball at will against an NYG defense that looked pretty darn good Monday Night. Their defense definitely is built to rattle the bad QBs and good QBs like Rivers, Brees, and even Romo this year all have handled their blitz looks, but I think the DVOA ratings are capturing the Eagles accurately. You mean to tell me the Eagles couldn't go into the Metrodome and win a game the way the Vikings look right now?

60
by Will Allen (not verified) :: Wed, 12/23/2009 - 1:48pm

If the Eagles and Vikings meet again in the playoffs in the Metrodome, it will be a fascinating example of how teams can change dramatically from one year to the next, even if their records, and perhaps DVOA rankings, are similar from one year to the next.

If Favre had engineered his emigration to Minnesota last year, and stayed healthy, THAT would have been a tremendously formidable team, even without Percy Harvin, given how good the Vikings were last year on the line of scrimmage. In many ways I enjoyed watching them more last year. The Eagles were not the homerun team on offense they are now. Two teams, similar records and even advanced rankings from one year to the next, but very different in how they arrived at those places.

110
by jmaron :: Wed, 12/23/2009 - 3:57pm

The more things change the more they stay the same.

Last year heading into week 16 the Vikings were 10th in weighted DVOA and Philly was 1st. This year the Vikings are 10th in weighted DVOA and the Eagles are 1st.

The difference - last year in week 16 the Vikings were coming off their biggest win of the year a 35-14 drubbing of the Cardinals vs the horrible loss to Carolina.

33
by TXNiner :: Wed, 12/23/2009 - 12:57pm

You probably should have left the 49ers out of your discussion. Anyone with any football knowledge will know that the Eagles beat a tough 49ers team by 2 touchdowns (only the 2nd time they have been beaten by more than 7 points) while a not-so-clearly superior team like the Vikings required an incredible catch to get past them, and that without their best offensive player for all but one play of the game.

As far as the Eagles games against good teams, they've beaten the Giants twice as well as the Falcons. The Vikings haven't exactly had a tough schedule (Detroit twice, Chicago, the NFC West, and the underwhelming AFC North), New Orleans is not the same dominant team it was earlier in the season with their D all but disappearing, and San Diego's wins have not been all that impressive, barely skating by Cleveland, Dallas, and Cincinnati the past three weeks.

39
by Rick (not verified) :: Wed, 12/23/2009 - 1:09pm

The Giants are pretty bad this year. The Falcons were playing their backup QB, second-string OL, and backup RB in the game against the Eagles.

46
by jebmak :: Wed, 12/23/2009 - 1:19pm

If the Giants are pretty bad this year, then that doesn't leave many teams in the average, pretty good, good, and great categories.

70
by TXNiner :: Wed, 12/23/2009 - 2:30pm

Can't quite figure out what your definition of pretty bad is. There would be a lot of teams who would like to be as bad as the Giants are. And in that Falcons game, the Eagles were also without their top RB, down a couple of their opening day starters on their OL, and were without their top WR. Yeah, the Falcons were dinged up, but so were the Eagles.

92
by Andrew B :: Wed, 12/23/2009 - 3:23pm

Wasn't that an Eagles team playing the Falcons that was missing their top wideout, their projected #2 wideout, top running back, the starting right side of their offensive line, and two of three starting linebackers?

Maybe I'm confusing them with another Eagles team that was 100% healthy and played a beat-up Falcons team that almost beat the Saints.

The Original Andrew

98
by Rick (not verified) :: Wed, 12/23/2009 - 3:29pm

The quarterback is arguably the most important position in football. The Falcons had to deal with the loss of their most important player, AND the loss of their offensive linemen and their top running back.

The Eagles only really had to deal with the absence of their top WR and their MLB, which is not nearly as bad. I don't count the loss of Westbrook the RB as too detrimental because RB production is more often the result of the way the OL plays rather than the individual skillset of the RB, and his replacements (McCoy and Weaver) have been more than adequate.

106
by Andrew B :: Wed, 12/23/2009 - 3:49pm

"AND the loss of their offensive linemen"

You mean their left tackle and right guard? Their right tackle, center, and left guard are all starters on their depth chart.

"The Eagles only really had to deal with the absence of their top WR and their MLB, which is not nearly as bad"

So playing an entire scrub/benchwarmer right side of the offensive line, like the Eagles did (and have done all season), doesn't count as "real injuries"?

And if you think the Eagles don't miss Westbrook, you clearly never watched Westbrook play. Maybe you should "watch the games."

The Original Andrew

108
by Eddo :: Wed, 12/23/2009 - 3:51pm

So playing without some offensive lineman and their top running back only hurt the Falcons, not the Eagles? You can't have it both ways.

143
by Rick (not verified) :: Wed, 12/23/2009 - 7:11pm

If you can't understand that the Falcon's missing quarterback had a far bigger impact on the game than the Eagles' missing pieces, you clearly don't understand football.

And if you don't realize that Westbrook is maybe 1/4 of the player he was at his peak (he's 30 years old right now, and clearly in decline) ... then you definitely haven't seen any Eagles games in the past two years.

148
by bubqr :: Wed, 12/23/2009 - 8:41pm

1. 1/4 the player he was at his peak ? You haven't seen Westbrook play recently (pre-injury/last year).
2. That Falcons team, the same one that nearly beat the Saints, wasn't competitive ? The same Chris Redman ? Plus it's not like the as-much-injured Eagles barely got a win vs this Falcons team: it was a blowout, and a shutout if not for a last tip-drill TD.

150
by Eddo :: Wed, 12/23/2009 - 8:47pm

You said:

"The Falcons had to deal with the loss of their most important player, AND the loss of their offensive linemen and their top running back."

However, Eagles were also missing their top running back and several offensive linemen in that game. You can't use missing those positions as an excuse for the Falcons, but not the Eagles.

Additionally, I never said that losing Ryan for that game wasn't worse than any other player; I just said you can't use losing some positions hurt the Falcons, while ignoring that the Eagles also lost starters at those same exact positions.

And, while Westbrook is no longer the player he was even last year, Turner is not some stud running back. He's a good back in a good running system; losing him is no worse than losing Westbrook.

188
by Andrew B :: Thu, 12/24/2009 - 12:40pm

The Falcons lost Matt Ryan, and had to play Chris Redman. Redman won the Tampa game coming off the bench and nearly beat the Saints.

The Eagles lost DeSean Jackson and had to play Reggie Brown.

If you want to maintain that Redman << Ryan, then you also need to grant that Brown << Jackson. And if Redman is so bad, how did he win a game off the bench, and then almost beat the Saints?

The Original Andrew

91
by Andrew B :: Wed, 12/23/2009 - 3:21pm

More to the point, you only really have to beat the good teams in the playoffs.

These Dallas and San Diego teams are so expert at that, of course.

The Original Andrew

171
by RickD :: Thu, 12/24/2009 - 12:20am

"These Dallas and San Diego teams are so expert at that, of course".

Ask the Colts about how the Chargers are a pushover in the playoffs.

179
by Big Johnson (not verified) :: Thu, 12/24/2009 - 4:51am

good one. let me guess, romo is no good in december as well?

btw the chargers have won a playoff game in each of the last 3 playoffs. Im guessing they are the only team that has atleast 1 win in each of the last 3 playoffs. Maybe philadelphia. I guess the chargers are better in the playoffs than you give them credit to be.

67
by bravehoptoad :: Wed, 12/23/2009 - 2:25pm

Rick is a troll. Do not feed.

71
by TXNiner :: Wed, 12/23/2009 - 2:31pm

Dang it, I got Rick trolled!

119
by dryheat :: Wed, 12/23/2009 - 4:17pm

It certainly seems as though "IF you watched the game..." and "Anybody with football knowledge knows..." are the new black.

120
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 4:24pm

Every year an average Philly team climbs to the top of the charts of this ranking system and then barely makes the playoffs or misses the playoffs even after a season of really poor play.

2008: Barely made the playoffs - advanced to NFC championship game.
2007: 10th in DVOA, 8-8, barely missed playoffs. Never at top of DVOA rankings (nor even ever close - basically at 10th by Week 6 and stayed there the entire time)
2006: 3rd final, 10-6 overall, won division, lost in 2nd round.
2005: 6-10, only near the top in the first few weeks (before McNabb, TO, Westbrook injuries and the disaster of Mike McMahon)
2004: 13-3, clinched first-round bye in the playoffs crazy-early, Super Bowl loss.

Only one of those years fits your criteria. In 2006 the Eagles had clinched a playoff spot in Week 16. You can't "barely make the playoffs" if you clinch a playoff spot with games remaining.
In 2007 the Eagles were never very highly ranked by DVOA.

blown out by San Diego

It was a one-score loss. One score! That's it! One stop in the 4th quarter and a drive is all the Eagles needed. Which they couldn't get, which is why they lost, but it's far from a blowout.

To lump the San Diego game in with the New Orleans game is just complete lunacy. The Eagles would've needed miracles to win the New Orleans game. They would've needed one pass from Rivers to be dropped on the final Chargers drive to have the ball with a decent chance of winning at the end of the game.

141
by Rick (not verified) :: Wed, 12/23/2009 - 7:07pm

I should have rephrased that:

"Every year an average Philly team climbs to the top of the charts of this ranking system and then disappoints in the playoffs or misses the playoffs after a season of mostly poor play."

151
by Eddo :: Wed, 12/23/2009 - 8:52pm

The last two years that the Eagles climbed to the top of the DVOA rankings were 2004 and 2008.

In 2004, they represented the NFC in the Super Bowl.

In 2008, they won two road playoff games before losing in the NFC Championship.

Since the Eagles did not climb to the top of the rankings in 2005, 2006, or 2007, it is 100% false that they climb to the top "every year".

And if you meant that they disappoint in or miss the playoffs every time they do climb to the top of the rankings, that's also wrong, considering they won two playoff games following each season in which they did reach #1.

153
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 8:58pm

They actually were on top one week in 2006 - at 4-4, if memory serves. However, I don't see how losing a 3-point game on the road in the second round is disappointing.

159
by Rick (not verified) :: Wed, 12/23/2009 - 10:26pm

They were top 5 DVOA in 2006, too, weren't they?

And then of course the whole string of seasons from 2001 to 2004...

166
by Scott C :: Wed, 12/23/2009 - 11:58pm

In which, IIRC, they were UNDER RATED by DVOA.

172
by RickD :: Thu, 12/24/2009 - 12:22am

"from 2001 to 2004..."

when they were going to the NFC championhip game every year?

176
by Alex51 :: Thu, 12/24/2009 - 1:16am

And then of course the whole string of seasons from 2001 to 2004...

You mean, when they made four straight NFC Championship Games? That string of seasons?

You're right, that's so disappointing for teams ranked, by DVOA, from 2001 to 2004: 2nd, 5th, 6th, and 7th. I mean, seriously, any team ranked 7th or higher should, at least, make it to the Super Bowl. And every team that's ranked in the top 2 had better win the Super Bowl. Anything else is disappointing.

180
by Big Johnson (not verified) :: Thu, 12/24/2009 - 4:56am

"I mean, seriously, any team ranked 7th or higher should, at least, make it to the Super Bowl. And every team that's ranked in the top 2 had better win the Super Bowl. Anything else is disappointing."

best post since raiderjoe

152
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 8:55pm

The only years that Philly has ever been on top in any week have been 2004, 2006 and 2008. In 2004 they went to the Super Bowl, and lost by 3 points (to a 7-point favorite). In 2006 they lost a playoff game on the road by a field goal in the second round. In 2008 they advanced to the NFC Championship game, and lost on the road by 7 points. They have never been on top and missed the playoffs.

Since when is a 5-3 record in the playoffs, including always having at least 1 win, with all losses within 1 score, "disappointing"?

I don't think you're helping your point here. Care to try again?

160
by Rick (not verified) :: Wed, 12/23/2009 - 10:33pm

I'm not talking about being "on top" with DVOA necessarily as being #1. The Eagles consistently climb to the top 5 of DVOA and then just as consistently disappoint in the playoffs.

In 2001 they lost to the clearly superior Rams. Ok.

In 2002 they lost to a Tampa Bay team they had beaten numerous teams previously, at home, and in poor weather that favored the Eagles.

In 2003 they lost to an inferior Panthers team that was markedly worse in DVOA rankings.

In 2004 they lost a close game to a slightly superior New England team that they definitely could have beaten.

In 2005 they failed to make the playoffs at all despite having a Super Bowl team the previous year.

In 2006 they made played a New Orleans team that was lower than them in the DVOA rankings in the divisional round and lost.

In 2007 they failed to make the playoffs despite having a higher DVOA than some playoff teams, and even the eventual Super Bowl champs.

In 2008 they finished #1 in DVOA at year's end and lost to an Arizona team ranked far, far lower in overall DVOA; the same Arizona team that they had plastered in November in a huge blowout victory.

If that's not a record of disappointment for a team that is supposedly a consistent DVOA contender, I don't know what is.

165
by Jeff Fogle :: Wed, 12/23/2009 - 11:54pm

By the numbers off Rick's post:

Playing to DVOA rankings in losses
2002: Philly (5) lost to TB (1)
2004: Philly (7) lost to NE (2)

Underachieving DVOA rankings
2003: Philly (6) lost to Carolina (17)
2006: Philly (3) lost to NO (12)
2008: Philly (1) lost to Arizona (21)

Didn't make playoffs:
2005: Philly (18)
2007: Philly (10)

Was curious what the DVOA rankings were and looked them up. Figured I might as well type them in after doing that...

189
by Andrew B :: Thu, 12/24/2009 - 12:56pm

The rest of the Reid era:

1999: Philly (24) no playoffs
2000: Philly (9) lost to NYG (10) for the third time in a season.
2001: Philly (2) lost to STL (1)

"Underachieving DVOA rankings"

Yeah, because that NEVER happens, right?

1999: OAK (3) and KC (4) missed playoffs
2000: TEN (1) blown out by BAL (3)
2000: PIT (4) missed playoffs
2001: STL (1) lost to NEW (12)
2002: KC (4) missed playoffs
2003: TB (11) missed playoffs, losing record, had won Super Bowl previous year
2004: BUF (3) missed playoffs
2007: NEW (1), GB (6), DAL (4) lost to NYG (16)

The Original Andrew

178
by Pat (filler) (not verified) :: Thu, 12/24/2009 - 3:56am

Your comment regarding 2005 has nothing to do with anything. Injuries cost them their starting QB, two WRs, and RB, kicker, punter, and probably half of the coaches were suffering from meningitis or something. DVOA had them as a bad team, and especially bad in weighted DVOA, so it's not particularly surprising they were bad.

They didn't make the playoffs in 2007 because they won the division in 2006. Swap their non-common games with Washington and Philly's in, and Washington's out.

If that's not a record of disappointment for a team that is supposedly a consistent DVOA contender, I don't know what is.

It's not. Losses like that happen, frequently. All of the losses you mentioned were close - two of them were on the road, so the 'much worse' in DVOA you're mentioning is much less than you think.

Why were they on the road? Because the Eagles, like Jacksonville, are in a crazy-good division, and so their regular season suffers and hurts their playoff position.

Not every team can win the Super Bowl every year.

191
by Andrew B :: Thu, 12/24/2009 - 1:05pm

The 2007 Eagles were pretty much one doinked Akers field goal, or one less play on that brutal 97 yard Bears drive from replacing the Giants or Redskins in the playoffs.

NFC East - 10 teams in playoffs in 4 years, all teams in at least twice. Giants 4 times, Eagles twice, Cowboys twice, Redskins twice. Has any other division done that? No.

The Original Andrew

15
by RMGreen :: Wed, 12/23/2009 - 12:28pm

I'm an Eagles fan and I don't get their great defensive DVOA. I know they're generating turnovers, but I can't feel good about them after the game against New York. Beating up on Alex Smith is indeed what a good defense should have done, but it doesn't make me feel much better about what happens when a good offense hits the field against them.

19
by Phil Osopher :: Wed, 12/23/2009 - 12:35pm

Maybe Philly is just a good team and Andy Reid is a good coach and McNabbus is a good QB. I know its hard to believe, but seems every year they win lots of games and we all think they suck and then they win some more games.

Just a thought.

Baltimore having the 2nd ranked defense surprises me as Ed Reed has been hurt on and off all year and Ray Ray is becoming a (somewhat of a) liability in coverage. They just haven't seemed to be the old Baltimore bash your face defense. Maybe they are a decent team overall. Their offense looked like crap against a horrendous Cleveland team on Monday night. Only reason they got that victory is Cleveland is a pool of feces when it comes to QB play and any type of offense (unless you are Detroit or KC)

199
by Alex51 :: Sun, 12/27/2009 - 6:10am

Maybe Philly is just a good team and Andy Reid is a good coach and McNabbus is a good QB.

No way. If that were true, then how do you explain the fact that they've lost seven playoff games so far in this decade? Seven!! They're tied with the Colts for most playoff losses in this decade. That just goes to show how the Eagles and Colts are consistently disappointing in the playoffs, despite great stats in the regular season. They're chokers, plain and simple. Well, except for the Colts, they won a Super Bowl a few years back, so they're apparently a clutch team that just had a little bad luck the other 7 times. But the Eagles? They clearly suck. And before you start making excuses for those seven playoff losses, I don't even want to hear this one again:

"But the Eagles and the Colts are the only two teams to have as many as seven playoff appearances so far this decade! The only reason no other teams have lost that many playoff games is that nobody else was so consistent in reaching the postseason!"

Excuses, excuses! You don't hear the Patriots making excuses like that, do you? They at least have the common decency to win the Super Bowl whenever they make the playoffs...well, except the last three times. But that's neither here nor there.

Point is, it doesn't matter that the Eagles are one of only three teams to win 10+ playoff games this decade. The only thing that matters is that none of those wins happened in the Super Bowl.

Therefore: Andy Reid sucks, McNabb is a choker, and the Eagles are doomed to once again disappoint in the playoffs despite being considered a top team by DVOA and the spreadsheet-obsessed nerds that run Football Outsiders.

QED

200
by tuluse :: Sun, 12/27/2009 - 11:40am

Haha, that was amazing.

25
by Podge (not verified) :: Wed, 12/23/2009 - 12:43pm

Am I getting confused by the Open Field Yards tables? For example, on Offensive Lines, Tennessee is ranked #1, with 1.93 open field yards, KC is ranked second with 0.70, Carolina is ranked third with 1.16 and the 49ers are ranked 23rd with 1.41. What am I missing, because that makes literally no sense to me whatsoever. Is it just a screw up with the table, with the ranks for SF and KC being switched?

30
by Aaron Schatz :: Wed, 12/23/2009 - 12:54pm

I had to put them together manually because the automatic HTML builder isn't written yet for the new table style, and I made some mistakes. Fixing now.

35
by tally :: Wed, 12/23/2009 - 1:02pm

KC is ranked 19th with 0.70. Are you looking at SF listed directly above them with 1.41 and ranked 2nd?

Oops. Ignore.

27
by KJ (not verified) :: Wed, 12/23/2009 - 12:51pm

The birds always lose to crummy AFC opponents early in the season. It is so consistent, I almost think it's nearly planned. Here's Andy Reid's pep talk before the Oakland game:

"OK guys, as you know, we've put together a ridiculously inept game plan that will surely hurt our odds of winning this game. That being said, if you execute the game plan perfectly, we still should be able to pull this one out. Of course, if we end up stinking in this game and lose, at least it's an AFC team and then every team that tries to game plan for us in the future will stupidly use this game as a blueprint for success. Now go out there and give 'em quasi-hell!"

34
by DaninPhilly (not verified) :: Wed, 12/23/2009 - 12:58pm

We have a winner!!!!!!

37
by N8 (not verified) :: Wed, 12/23/2009 - 1:06pm

PRICELESS!

Also, as a Packers fan the Eagles scare me, the Vikings, not so much. History suggests the Eagles would smoke Favre in a 4 INT performance.

51
by ammek :: Wed, 12/23/2009 - 1:23pm

Even allowing for a dash of hyperbole, history suggests no such thing. In a dozen games against Philadelphia, Favre has thrown 12 picks. Only once have the Iggles intercepted him three times — and that was back in 2000 in a 6-3 Packer win.

Also, Favre has never played Philly as anything other than a Packer. Given that the Packers haven't won in Philadelphia since 1962, what makes you think it's just a Favre problem?

Recent history — ie, this season and last — suggests the Packers have plenty of reason to fear Minnesota, also.

74
by Arkaein :: Wed, 12/23/2009 - 2:36pm

I'd add that I don't think there is a single surefire matchup advantage within the entire NFC playoffs.

I really liked GB's chances in a rematch with the Vikings up until a few weeks ago, given the progression of their defense, the return of Jermichael Finley, and the recent vulnerability Minny has shown. Once Al Harris went down it has become clear that the Nickel and Dime back positions are likely GB's Achilles heel against any team with a decent 3rd WR, or teams with good TEs that can draw coverage from Woodson.

The Saints have always scared me with their ability to run their offense through any player on the field, but their defense and offensive line have looked vulnerable.

The Eagles, Cardinals, and Giants seem inconsistent. The Cowboys may be playing some of the best ball in the NFC right now, but they've lost 3 combined games to the Giants and Packers.

The NFC playoffs are shaping up to be quite exciting.

89
by jmaron :: Wed, 12/23/2009 - 3:14pm

right now the Vikings would fear Det

29
by Ryan D. (not verified) :: Wed, 12/23/2009 - 12:52pm

Baltimore almost won in Minnesota, but lost on a missed FG as time expired. Being 9-5, with a win over Minnesota (before their recent collapse) would look a lot more impressive than 8-6, especially in the current playoff race for the AFC wild card. Their early win @San Diego also looks nice, even if it is less of a factor in weighted DVOA now. They were also the first team to beat Denver, and did it 30-7 right after Denver won @SD and vs New England. They also lost to Indy by 2 in what was definitely a very close game. I think one could easily make a case for Baltimore as one of the very best teams in the NFL.

138
by SteveNC (not verified) :: Wed, 12/23/2009 - 6:41pm

They also barely beat the Steelers with Dixon at QB.

147
by jonnyblazin :: Wed, 12/23/2009 - 7:51pm

And the Steelers lost to the Browns, Chiefs, and Raiders with Big Ben at QB.

190
by SteveNC (not verified) :: Thu, 12/24/2009 - 12:57pm

So Baltimore and Pittsburgh are clearly worse than the Browns, Chiefs, and Raiders!

32
by Fred (not verified) :: Wed, 12/23/2009 - 12:57pm

In the 1990 NFC championship, Joe Montana was injured late in the game and it was Steve Young who handed off to Roger Craig. So yeah, Joe couldn't have kept Craig from fumbling.

36
by tally :: Wed, 12/23/2009 - 1:04pm

We all know that handoffs from lefty QBs are fumbled more often than those from right-handed QBs.

40
by Fred (not verified) :: Wed, 12/23/2009 - 1:10pm

As I recall, Young actually handed off with his right hand on that play since the run went to the left. I remember someone speculating after the game that had the QB been right-handed, Craig might not have fumbled. I'm not saying that was a factor. I'm just saying someone else said it.

66
by Will Allen (not verified) :: Wed, 12/23/2009 - 2:17pm

Ah, yes, one of the most significant fumbles in NFL history, in terms of how we form opinions on who the best coaches are. I've long believed that if that handoff isn't fouled up, George Seifert likely gets at least one more Super Bowl victory, leaves the the 49ers on his own terms, never goes to Carolina to experience disaster, likely retires with the highest winning percentage among all NFL coaches, and is viewed by very many people as the greatest football coach ever.

Instead, he will never get inducted into the Hall of Fame, and is barely thought of anymore. All due to one botched handoff. We fool ourselves when we begin to think so highly of our analytic abilities, in regards to something so random as a football game or career.

104
by Rick A. (not verified) :: Wed, 12/23/2009 - 3:45pm

Man, you got that right. One of the most astute posts I've ever seen on this site. Pretty much no one thinks this way, human nature being what it is, except the players and coaches whose life fate is often determined by the combination of the small things that can and will go wrong, together with the aforementioned human nature that only remembers results...

109
by Kevin from Philly :: Wed, 12/23/2009 - 3:55pm

Don't kid yourself. Seifferts' fate was sealed once he did that "bossa nova" commercial.

145
by patriotsgirl :: Wed, 12/23/2009 - 7:40pm

And Belichick's reputation likely would not have received the same boost, as a) the defense wouldn't be remembered for the great game they played against SF, as they would have lost; and b) more importantly, he's never given the chance to design the Bills-stopping game plan that ended up in the Hall of Fame.

Of course, I think to say Seifert "very likely" gets another SB victory might be overstating a little, as the Bills themselves were no pushover that year. It would have been an interesting game.

Speaking of plays that changed football history, Norwood's FG also falls in that category. If he makes it, the perception of the 1990s Bills is completely different (even if they lose the next 3 SBs). I still believe those Bills teams were pretty awesome, but had the bad luck to lose that game - and then play three juggernauts in a row (Redskins and Cowboys twice).

Of course, my memory of the Bills may be exaggerated by the fact they regularly beat up on the Pats, and I saw them virtually every freaking week (living in the WNY market).

78
by MC2 :: Wed, 12/23/2009 - 2:49pm

Actually, Aaron didn't mention Montana by name, and the claim he made ("the QB of that team was unbelievable") could apply to Young as well.

41
by Paul R :: Wed, 12/23/2009 - 1:14pm

I don't know how to search for this, so I'll just ask it here:

How do DVOA rankings compare to Super Bowl wins?

I've only been following this site for three SB's, and none of the top DVOA teams won any of those three.

50
by jebmak :: Wed, 12/23/2009 - 1:23pm

Even the best team only has about a 30-40% chance to win the Super Bowl. Say that they have a 70% chance to win every game (pretty high vs playoff teams). They have to win three games, so .7*.7*.7 = 0.343 or 34%.

53
by Aaron Schatz :: Wed, 12/23/2009 - 1:30pm

Please see table in last week's DVOA analysis that shows how the team with the better record actually has a losing record in playoff games since 2005. The last four postseasons have been significantly different than those that came before.

56
by Danish Denver-Fan :: Wed, 12/23/2009 - 1:34pm

I wrote this in the DVOA thread back in week 9:
------------
"Even if DVOA have you top-ranked after week 17, winning the SB is still very hard!

Say you have a first round bye. You start out in the divisional round, and face a team you are 65-35 to beat. Thats a big advantage considering that it's the playoffs we're talking about. Then you go to the Championship game were you, at best, will be a 60-40 favorite. Then you go to the SB where you are basically 50-50 (assuming the AFC and NFC are somewhat equal in strength). The probabillity of succeeding is: .65 * .60 * .50 = 19%... Feel free to adjust the numbers any way you want, but I doubt your gut-feeling will get the final number much over 25%."
-------------

79
by zlionsfan :: Wed, 12/23/2009 - 2:50pm

It's similar to what people say about baseball playoffs some years ... how could X possibly happen? (And if you play sports sims, you hear it this way: "How could team X possibly beat my team?") Being a 2-1 favorite in each game is a huge advantage, but that still means that once in 27 times you'll get swept in a best-of-five series.

Using your 65-35 split (which, as you point out, is a big advantage), if every top seed had roughly that kind of advantage, we'd see a top seed lose about every year or two ... or looking at it another way, if you had a team that was so strong that it would be a 3-1 favorite against any playoff opponent, it would still only win the Super Bowl about 42% of the time.

Not that the probabilities determine the outcome, of course. Just trying to emphasize how difficult it is to advance in any kind of tournament, whether it's single-elimination or best-of-seven rounds. (Extending playoff rounds to seven games from five or three really doesn't make a lot of difference in terms of chances to win overall.)

81
by Tom Gower :: Wed, 12/23/2009 - 2:54pm

The last #1 playoff seed to win the Super Bowl was the 2003 Patriots. Before that, it was the 1999 Rams. Whatever's going on, it's not just DVOA-related.

87
by jmaron :: Wed, 12/23/2009 - 3:08pm

Super Bowl Participants - Week 15 DVOA Ranking

1994 - SF (2) vs SD (10)
1995 - Dall (2) vs Pitt (3)
1996 - GB (1) vs NE (10)
1997 - GB (5) vs Den (2)
1998 - Atl (7) vs Den (1)
1999 - Stl (1) vs Tenn (10)
2000 - NYG (10) vs Balt (2)
2001 - StL (2) vs NE (13)
2002 - TB (1) vs Oak (3)
2003 - Car (19) vs NE (6)
2004 - Phil (2) vs NE (3)
2005 - Sea (4) vs Pitt (6)
2006 - Chic (2) vs Indy (7)
2007 - NYG (17) vs NE (1)
2008 - Arz (20) vs Pitt (4)

Number 1 teams - 5 appearances 4 wins 1 loss
Number 2 teams - 7 appearances 4 wins 3 losses
Number 3 teams - 3 appearances 1 win 2 losses
Number 4-6 teams - 5 appearances 3 wins 2 losses
Number 7+ - 10 teams - 3 wins 7 losses

112
by Kevin from Philly :: Wed, 12/23/2009 - 3:59pm

Damn! If only the Colts were in 3rd place, I'd be feeling prety good right now.

167
by Scott C :: Thu, 12/24/2009 - 12:07am

I wonder what it is in week 16 or 17 weighted DVOA.

44
by tally :: Wed, 12/23/2009 - 1:18pm

I think you'll be hard pressed to find any rankings predictive of Super Bowl wins. One game is simply too random to predict with any great degree of certainty prior to the beginning of the playoffs.

47
by Jamie C (not verified) :: Wed, 12/23/2009 - 1:21pm

A comment about the Eagles:

they are 7-1 with Jason Peters at left tackle and Todd Herremans at LG. they lost to oakland because King Dunlap was playing LT and they couldn't stop any pressure. since they have their starting 5 Oline playing they have remarkably dominating on offense.

it also seems that they are generating turnovers at a far better rate than ever before.

so the eagles have:
a dominating offensive line(can you remember the last mcnabb pressure/sack?)
outrageously good receivers
quality running backs that get better when Westbrook comes back
a top 5 DE in the league.
the best cornerback tandem in the league
2 extremely under-appreciated DT's.
a quarterback that looks like he's playing the best football of his career.

62
by Will Allen (not verified) :: Wed, 12/23/2009 - 1:57pm

Yet they are capable, just recently, of going out and tackling like a bunch of drunken frat boys playing intramural football.

I am not anti-Eagles at all; just saying that handicapping this year's playoffs is a mug's game.

64
by Rick (not verified) :: Wed, 12/23/2009 - 2:04pm

Best cornerback tandem in the league?

Asante Sameul is an overrated, overpaid gambler who gets burned more often than he picks off bad quarterbacks (and he mostly just picks off mediocre qbs - the good ones usually don't fall for his route jumping), and he basically couldn't tackle a small child.

Sheldon Brown is slightly above-average but has definitely lost a step in coverage because he's getting up in years.

84
by DaninPhilly (not verified) :: Wed, 12/23/2009 - 2:57pm

Rick, I know you're a troll, but you cannot say that anyone with as many INTs as Samuel over the years is anything but a great DB.

94
by Rick (not verified) :: Wed, 12/23/2009 - 3:24pm

Seriously ask yourself if 9-10 interceptions a season are worth it when he gives up maybe that same number in touchdowns by blowing his coverage to jump a route?

Also consider that most of Asante's interceptions come as the result of mediocre opposing quarterbacks. I can guarantee you that Asante will rarely intercept a Peyton Manning, Drew Brees, or a Kurt Warner.

And then you have to factor in his abysmal tackling ability which results in tons of YAC for opposing WRs who beat his coverage, and you have the makings of a highly overrated CB.

It's easy to see all of the INTs that come as a result of risk-taking, but you have to look at the stats more closely and examine in-game performance over the course of a season.

100
by Jamie C (not verified) :: Wed, 12/23/2009 - 3:38pm

i don't think the eagles would be 4th in passing defense if Asante Samuel was as bad as you say he is. he hasn't gotten beaten nearly as often as you claim. he might of given up 1, maybe 2 long TD's this season. i'm not going to disagree that he's not a good form tackler. but he's also been playing with a stinger all season.

103
by DaninPhilly (not verified) :: Wed, 12/23/2009 - 3:43pm

I forget, how many TDs to WRs have the Eagles given up this year? You should know, given the certainty with which you present your arguments.

Considering you said "Seriously ask yourself if 9-10 interceptions a season are worth it when he gives up maybe that same number in touchdowns by blowing his coverage to jump a route?" then surely Asante by himself has given up 9-10 INTs this year, correct? Otherwise, your entire argument is exposed as a half baked piece of crap, correct? So, how many TDs to WRs have the Eagles in general and Asante in particular given up?

105
by Eddo :: Wed, 12/23/2009 - 3:48pm

While I think "great" is indeed too much praise for Samuel, he does indeed intercept good quarterbacks.

Of his 41 career interceptions (34 regular season, 7 playoff), he has intercepted:
Carson Palmer (2x), Tony Romo, Ben Roethlisberger, Matt Ryan, Eli Manning (2x, including once in the playoffs), Chad Pennington (playoffs), Peyton Manning (playoffs), and Philip Rivers (playoffs). Those eight good quarterbacks account for just under 25% of his total interceptions.

He's a gambler, but he's not just preying on weak QBs.

174
by RickD :: Thu, 12/24/2009 - 12:33am

You guys are far too polite to somebody who's just talking out of his a$$

99
by Andrew B :: Wed, 12/23/2009 - 3:36pm

"Sheldon Brown is slightly above-average"

Really? Can you name some other starting cornerbacks who've given up fewer touchdowns per season over a 4 or 5 season stretch compared to what Brown has done from 2004 to 2009? Just curious.

The Original Andrew

52
by Eddo :: Wed, 12/23/2009 - 1:29pm

Color me confused as to why there is still uproar over the Eagles' ranking. Last year, everyone nearly had a shitfit when a very similar Eagles team (dominant at times, "short-yardage woes", a loss to a terrible team) was ranked #1, and that team went on to the NFC Championship. Doesn't that validate DVOA somewhat regarding the Eagles?

(I'm not saying that an NFC Championship run should be used as any legitimate validation metric, but it's the kind of thing you would think impresses those who don't understand the Eagles current high ranking.)

121
by dryheat :: Wed, 12/23/2009 - 4:25pm

Actually, I think DVOAs biggest flaw is captured in that the Eagles are always among the NFL's elite year-in year-out, despite on-the-field evidence to the contrary....and nobody seems to be able to correct for it.

125
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 5:09pm

Actually, I think DVOAs biggest flaw is captured in that the Eagles are always among the NFL's elite

Really? When's this happened? Oh wait - in 2008. That's it.

In 2007 they were ranked 10th as early as Week 6, which is where they ended - with an 8-8 record and the 5th hardest schedule in the league. They actually played the team directly above them, and lost by 4 points with their backup QB (McNabb was injured) throwing 4 picks. That seems pretty strong evidence they were exactly where they should be.

In 2006 they were ranked 3rd, finished 10-6, winning their division, won the first round game and lost the second, versus New Orleans by a field goal.

In 2005 they were only ranked high for a few weeks, before they plummeted after injuries knocked out the entire offense, where they stayed.

In 2004 they were ranked high and massacred their conference before losing the Super Bowl by a field goal.

Why do people keep mentioning the Eagles? The Jaguars have the same freaking issue over the exact same timeframe, and I have never heard them mentioned.

2008: ranked 20th, but at 5-11, ranked around 8-8/9-7 teams. (4.9 games under their est. win total)
2007: ranked 3rd, at 11-5 between 13-3 teams (but at their est. win total)
2006: ranked 5th, at 8-8 between a 12-4 and 13-3 team (2 wins under their est. win total)

On average, ~2.1 est. wins higher than their total, exactly the same as the Eagles. (P.S.: what do these two teams have in common? Oh yeah - their division's crazy good. Hint.)

54
by MJK :: Wed, 12/23/2009 - 1:32pm

I was rather surprised at Baltimore sitting at #2 as well. I've watched a few Baltimore games this year, and while they definitely are a top-half-of-the-league team, I wouldn't say that their play is better than Indy's, New Orleans', Minnesota's, or even the Packers', at least to the naked eye.

Maybe Aaron just hasn't caught that the "plays in the Chesapeake Bay area" modifier is turned up too high? Of course, it doesn't seem to help the Redskins... :-)

Seriously though, regarding Philly, I know we've rehashed this to death, but I still think there are some very real reasons why their DVOA often seems to run higher than their win-loss record. On offense, I think it relates to DVOA not taking into account the time management aspect of late-game plays into account, so a team that is good at moving the ball, but that has some questionable clock-management strategies and an allergy to running the ball even if they do so effectively will tend to underperform (W/L wise) than their DVOA would predict.

On defense, Philly's defense strikes me as a "gambling" defense--they rack up a lot of sacks and INT's, but give up a lot of yards in the process. An INT costs an offense an average of 6 "success points". So in order to be penalized for their gambling ways by DVOA, Philly would have to give up roughly 6 first downs or perhaps 80 yards of field position for every INT they get. I think bad and mediocre teams are just not capable of moving the ball that much against a decent defense, so every time Philly gets an INT against such a team, they get a huge DVOA defensive bonus. I wonder...how much has the success point value for INTs been tuned? Is the -6 points the value that maximizes DVOA's predictiveness, or could tweaking this parameter additionally perhaps improve things? I assume Aaron has looked at this...

61
by Crushinator (not verified) :: Wed, 12/23/2009 - 1:53pm

On top of which, I think their offense is generally overranked by DVOA because of the value placed on moving the ball across the entire field. I think the Eagles red zone situation makes them a team that should be a top 3 offense in the league but having few physical playmakers holds them back.

It seems like their red zone offense has dramatically improved since involving Weaver more, but watch the Eagles in the red zone and how McNabb just stares down Celek every play. There's no way opposing defenses miss that.

68
by Kasmir :: Wed, 12/23/2009 - 2:28pm

I wish you guys would investigating increasing the weighting of variance into your weighted DVOA. Much of the bayesian angst with DVOA is explainable by looking at variance: IND, DAL, and MIN are the only top 8 DVOA teams with top 8 (minimal) variance. PHI and NE are the only top 8 teams with bottom 8 (maximal) variance. Most folks would believe that the first three teams are better than the latter two. I would hazard the guess that trended variance correlates even better with popular perceptions.

How much offseason analysis do you do in impriving the predictive value of weighted DVOA? It would seem to me you have a tremendous amount of data to run alternative predictive scenarios against, i.e. how well to various alternative mid-season DVOA weightings correlate with end-season DVOA?

79
by Eddo :: Wed, 12/23/2009 - 2:50pm

"Most folks would believe that [IND, DAL, and MIN] are better than [PHI and NE]."

Most folks might believe that, but is it such a sure thing?

IND beat NE by one point at home. If Kevin Faulk doesn't bobble a catch (or if a difference coaching decision (playcall or punt) is made), that game may turn out differently.

DAL did beat PHI twice, but PHI has a better record than DAL. I'd say it's awfully tough to determine which team is better.

Week 8 MIN may have been better than week 8 PHI, but is that true of week 16 MIN and week 16 PHI? MIN does have a better record, but has lost two of three, looking bad in the process. Is it obvious that MIN is the better team? I don't think so.

---

You do have a point regarding variance; it would be nice to see the theoretical normal distribution of a team's DVOA. However, I think people are getting too caught up in the ranking of teams instead of actually interpreting the actual ratings and variances. What DVOA is currently saying is that PHI has a higher mean value across all their plays than any other 2009 team. However, it's also saying that, while IND, MIN, and DAL don't have as high a mean value, they're more consistent in their outcomes.

Think of it as students taking a test: PHI is the student that scores a 95, 100, 75, 75, then 90, for an average score of 87; IND is the student that scores 85, 80, 85, 85, then 95, for an average score of 86. PHI's average is higher, but IND scored higher on three of the five individual tests.

101
by Andrew B :: Wed, 12/23/2009 - 3:39pm

"DAL did beat PHI twice"

When did Dallas beat Philly the second time? Two weeks from now?

The Original Andrew

107
by Eddo :: Wed, 12/23/2009 - 3:50pm

Ah, my bad. That fact further strengthens my point, though.

126
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 5:12pm

... Why? If Philly kills Dallas, it directly undermines your point.

127
by Eddo :: Wed, 12/23/2009 - 5:21pm

Well, the post I was responding to said that DAL was definitely better than PHI; I asked if that was really a sure thing. DAL beating PHI twice would have been a big point in the original poster's favor; DAL beating them once weakens his point, thus improving mine.

129
by Chief too :: Wed, 12/23/2009 - 5:24pm

Because, however he phrased it, his true point was that it's not clear that Dallas is better than Philadelphia.

132
by Eddo :: Wed, 12/23/2009 - 5:33pm

Thanks Chief.

137
by Kasmir :: Wed, 12/23/2009 - 6:28pm

"Most folks might believe that, but is it such a sure thing?"

My remark was about the mismatch of DVOA rankings vs popular perceptions. Of course no ranking system is "correct", and even the ultimate W-L rankings while definitive would probably not predict a continued season very well.

My sole point was that perceived rankings and DVOA rankings might correlate better, for what it's worth, if variance were weighted more.

Variance -- properly measured -- is an index of "quality control" of a team. A good team with low variance is likely a truly good team, and likewise a poor team with low variance is likely a real stinker. Teams with high variance are prone to random ranking, so they're less likely to belong in the ranking quartile they end up in.

New England is a particularly interesting case, as their offense is lowest in DVOA variance while their defense is highest! Since their aggregate variance is 30th, special team variance must be pretty bad as well.

Amusingly, Jacksonville's 32nd ranked variance offense is going up against New England's 32nd ranked defense this weekend. Anything could happen :)

140
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 6:47pm

You're looking for estimated wins, which is in the table above.

"Variance -- properly measured -- is an index of "quality control" of a team."

Variance here is the spread in a team's actual previous performance. It's not particularly predictive, because most of it probably came from actual changes in the team which aren't repetitive. That's why DVOA tends to be a better predictor than estimated wins.

146
by Kasmir :: Wed, 12/23/2009 - 7:50pm

"[variance is] not particularly predictive, because most of it probably came from actual changes in the team which aren't repetitive."

How do you know that? I'd be interesting in seeing any evidence or at least an argument from data.

154
by Pat (filler) (not verified) :: Wed, 12/23/2009 - 9:00pm

How do you know that?

Estimated wins is less predictive than DVOA. Variance is pretty much the only component that isn't in DVOA, and DVOA is more predictive. The other factors are just weighted differently.

161
by Kasmir :: Wed, 12/23/2009 - 11:09pm

You're just repeating your assertion. And I'm not asking about estimated wins. How do you know that variance is based on "non recurring changes" anymore than any other performance factor is? If anything, I would expect low variance in DVOA to suggest there were *less* special causes in that teams DVOA-measured performance history. Sounds like you think variance is a particularly useless statistic to report on. Do you have any actual evidence that variance is uncorrelated with predicted DVOA?

76
by jmaron :: Wed, 12/23/2009 - 2:40pm

7th in variance for the year. They must be 32nd in the last 4 weeks.

Last 4 weeks:

Chic 43.5,
Arz -46.9,
Cinn 29.2,
Car -28.5

The Vikings have struggled against defending the pass for much of the season:

SF 37.6, StL 54.7, Bal 43.4, Arz 76, Car 65.7

But they also had 7 games with a off pass DVOA over 60%. If they can't figure out how to get the pass game going again they will be looking at a quick exit out of the playoffs.

When I read Childress start talking about getting back to our roots running the ball and imposing our will I was concerned. Favre may be a self centered jerk but I suspect he has a better handle on what will make this offence successful, the Vikings need to pass the ball very effectively if they have any chance. Plowing Peterson into the line every 1st down for no gain is not going to get it done.

86
by Will Allen (not verified) :: Wed, 12/23/2009 - 3:01pm

Yet if they don't force teams to scheme to stop Peterson, their chances of passing effectively drop quite a bit. It's a real conumdrum.

All this Childress vs. Favre hoopla is a byproduct of bad offensive line play. I said last spring that the coaching staff's judgement would receive a test with how much faith they placed in Sullivan playing adequately. Right now, it doesn't look so hot. They can't run between the tackles nearly as effectively as they have the last couple of years, and that makes it easier to defend the pass.

This team needs to block better, period, and if it doesn't, what plays are called won't make a damned bit of difference.

93
by jmaron :: Wed, 12/23/2009 - 3:24pm

yeah that really is at the crux of it. But I thought in the two games against Arz and Car they left the tackles on islands far more than they should have. And they made next to no adjustments in either game.

At this point I'm feeling pretty negative about this team because they got their asses kicked at the line of scrimmage against Car and Arz. But like you said predicting this years playoffs is really a mugs game.

113
by Will Allen (not verified) :: Wed, 12/23/2009 - 3:59pm

When they stepped up and traded punches effectively against decently physical Bengal lines, I thought the Arizona game was just an anomaly. Then they go out and get pushed around as much as they did Sunday night. I don't have any effin' idea as to what might happen, but I sure as hell ain't plannin' to scalp tickets come February.

169
by jmaron :: Thu, 12/24/2009 - 12:10am

Of course injuries could be a big factor. If I'm not mistaken every member of the starting offensive line has been on the injury report over the last 3 weeks. Hutchinson has had a bad back since early on. Not exactly a good thing to have in that line of work.

Up until Arizona the passing game was excellent. From weeks 4-12 they only had one passing DVOA under 57.8 (Pitts was -12.8 and that had a lot to do with called back TD's and two very fluky plays). But since Arizona when the injury bug hit the passing game has suffered (14.6, 5.8, -3.7).

The running game has never been there this year - They've only had 4 positive DVOA rushing games against (4.3 to 30.6 range) and only one in the past 7 weeks (19.0 vs Cinn).

There are only 4 guys listed on the report this week, only one offensive lineman (Hutchinson - shoulder). Let's hope the troubles were injury related and that they are getting healthy.

111
by Rick A. (not verified) :: Wed, 12/23/2009 - 3:58pm

Exactly. But I still think "Chilly" is quite overmatched as we move toward nitty-gritty time. He could adjust play calling to ease the pressure that has mounted on Minny's Bigs--but he has always appeared so rigid in his football thinking, and thinking in general, that you have to doubt he will be able to think a little bit outside the box. They need MORE designed roll-out plays right now, which Favre excels in and always has, where are they ? "Chilly" doesn't want to go there--my own opinion is his ego is involved. If he does that it will look to the world that Favre is taking over even more, because that's what Brett wants--more of those plays. What does this cat Childress want, anyway ? Favre's performance clearly has put them in a position they would very, very, very likely not be in--and that position just got the guy a nice, fat 7 figure contract extension...

115
by Rick A. (not verified) :: Wed, 12/23/2009 - 4:02pm

I was replying to Will Allen. But to jmaron : there's no reason, as of yet, for outright pessimism. I think certain players are going to take control of this situation. We'll see if I'm right. And, therefore, we'll see just exactly how much they WANT it. Most of the time, in football, when it comes to the best teams, the one's that want it more find a way. I know Brett Favre REALLY wants it. And they have the team, the tools, the ability...

136
by jmaron :: Wed, 12/23/2009 - 6:03pm

One possible explanation for the major drop off in performance could be injuries - I believe they've had as many as 13 players on the injury report in the last few weeks including 3 offensive lineman.

117
by Will Allen (not verified) :: Wed, 12/23/2009 - 4:12pm

I've always been ambivalent about Childress, because I've never seen evidence that he evaluates talent well. It doesn't take a genius to figure out that Steve Hutchinson and Jared Allen are good players. Having said that, I also think trying to mind-read people from afar is pointless. Hell, doing so with people you know very well, and are in close proximity to, is usually a waste of time.

123
by Rick A. (not verified) :: Wed, 12/23/2009 - 5:07pm

Well, #4 and "Chilly" are grown men. They'll get it, at least, together in a workable way. But one of them is a known winner--the other ???? Childress must lower his ego and rigidity. However, as I said, if he doesn't I would bet my bone marrow a few key players, not just #4, will take that ship over...

75
by Big Johnson (not verified) :: Wed, 12/23/2009 - 2:38pm

a sidenote for the playoffs odds page. can we get the percent chance at the "Eli crybaby bowl" NYG vs SD.

116
by tally :: Wed, 12/23/2009 - 4:07pm

Just treat NYG and SD getting to the Super Bowl as independent events and multiply their conference win %. Sorry to disappoint, but it's only a 0.4% chance of seeing this matchup.

85
by zlionsfan :: Wed, 12/23/2009 - 3:01pm

6% chance? No way, man. We already have a weaker schedule than Tampa Bay and have two sub-.500 teams remaining (SF and Chi) while the Bucs have New Orleans and Atlanta.

Realistically, though, it'll be the Rams. St. Louis missed their best opportunities. The chance of them beating either SF or Arizona is null.

ha ha.

124
by Still Alive (not verified) :: Wed, 12/23/2009 - 5:09pm

A simple divisional & conference summation for those interested.

DVOA
AFCE 29.0
AFCN 22.7
AFCS 40.0
AFCW -34.5
AFC 57.2

NFCE 70.5
NFCN -33.9
NFCS -03.8
NFCW -55.9
NFC -23.1

Advanaced FB Stats GWP
AFCE .11
AFCN -.06
AFCS .43
AFCW -.07
AFC .41

NFCE .78
NFCN -.39
NFCS -.06
NFCW -.72
NFC -.39

Some pretty clear patterns there and the rankings are the same at both sites. it feels like the NFCE has been strong for so many years now, and the NFCW so weak. not surprising to see the AFC so far up on the NFC overall.

128
by jmaron :: Wed, 12/23/2009 - 5:21pm

"not surprising to see the AFC so far up on the NFC overall."

I don't think there is really much difference between the conferences now. Det and StL are simply weighing down the rankings. If you look at the top 6 in each league the NFC is actually better (146.8 vs 129.3).

133
by Still Alive (not verified) :: Wed, 12/23/2009 - 5:54pm

Even without those two the WAS/CHI/ATL/CAR/TB/SEA group is much weaker than the corresponding AFC CLE/BUF/JAC/TEN/MIA/CIN.

The tops may be similar or even favor the NFC, but the middle and bottom of the AFC is much better.

131
by t.d. :: Wed, 12/23/2009 - 5:26pm

same top two as last year

139
by SteveNC (not verified) :: Wed, 12/23/2009 - 6:44pm

"Teams that have had their bye week are projected as if they had played one game per week."

So estimated wins are out of 15 games instead of 14? Not sure I see what the point is.

144
by Aaron Schatz :: Wed, 12/23/2009 - 7:25pm

No, no, that's only before Week 10. I'll try to get rid of that line after Week 10 next year.

155
by SDfan (not verified) :: Wed, 12/23/2009 - 9:23pm

Since the Chargers are #1 in variance now, and if it was factored into DVOA more than it is currently (as one poster has suggested), would that help explain why SD is still only ranked 13th? Even with the more in-depth analysis into the Chargers last week in the rankings, it's still crazy how low they are. At least weighted DVOA puts them 8th, which is a little better, but they have beaten a number of teams ranked ahead of them.

156
by chemical burn :: Wed, 12/23/2009 - 9:45pm

They have also lost to 3 teams ranked ahead of them. Two of their wins versus teams ranked higher in DVOA were squeakers (versus NYG and DAL) and they've also skated by some really terrible teams. My question for SD fans is what are the signature wins they've had this season that make you confident they are being under-rated? A 1 point win over NYG or a 3 point win over an ok CIN team? The split decision with Denver? Sneaking by OAK and CLE? 8th in weighted DVOA seems very, very reasonable to me...

158
by Marver :: Wed, 12/23/2009 - 10:15pm

If you're going to point out the narrow victories they have had (I still don't understand how you call the Cowboys game a squeeker...the 0:02 TD was meaningless), at least point out the narrow losses. They barely lost to Baltimore on a late-game touchdown followed by a drive stalling deep in Baltimore territory (due more to play-calling than anything). The loss to the Broncos was the result of TWO return touchdowns by the opponent, and was later reciprocated with a thorough beatdown. The loss to the Steelers required a late-game, field-goal drive by Pittsburgh to halt what may have become the greatest regular season comeback in history.

They're likely going to sweep "the best division in football", and have beat the only other division leader on their schedule. If they win this week, they'll have swept the reigning AFC East and AFC South division champions. They beat three teams ranked higher than them in DVOA on the road.

And plenty of teams would love to have 'snuck' by Oakland. Ask the DVOA leader or the reigning Super Bowl Champion. And I'm not sure you using 'sneaking' in the Cleveland game, when the Chargers led 27-7 at the conclusion of the 3rd quarter, is the best way to describe that game.

164
by jmaron :: Wed, 12/23/2009 - 11:41pm

You're being too much of a fan - SD is a solid team - they may well be better than their weighted DVOA but there is nothing they've done that sets them apart from some other good teams this year.

SD has played 7 games against teams with double digits DVOAs and they've won 4 lost 3. They have scored 208 and allowed 190. Which is pretty good considering. But it pretty much tells you they are roughly on the same level as those teams.

Balt, Pitt, Den, NYG, Phil, Dall

173
by Scott C :: Thu, 12/24/2009 - 12:32am

The team has changed quite a bit since the Balt and Pitt losses. The defense has had a slow transformation in scheme since Jamal Williams went down and the D-line had to shift to a 5-6 man rotation. The only real starter there is Louis Castillo. The coaches have determined that everyone else is only good enough when fresh, or a rookie that is too raw.

Overall, the team has played 60 different players so far. I'm not sure where that ranks in the league, but injuries and turnover have been a big story. The injuries have had a broad effect on the front 7 on defense and a focused effect on the O-line's right side.
The recent close game of some concern is Cincinati (Dallas was not that close -- they were up by 10 points with 10 seconds left in the game; Cle was way ahead until falling asleep in the 4th quarter). But considering who was out temporarily due to injury in the last several weeks and the various backup juggling its not so bad.

IF -- and this is an if -- the team can get a bye and get healthier on the D-line and LB'ers, and Hardwick can come back at C (providing much needed run blocking up the middle) there is good reason to think the team will play very strong in the playoffs. So many backups have seen lots of playtime, and the coaching staff has had to try out lots of adjustments and 'experiments' throughout the year.

Its strange that they have such low variance considering all the injury turnover. I'm not sure what to think of that.

175
by Marver :: Thu, 12/24/2009 - 12:59am

You're forgetting that among those teams, only the Chargers and Baltimore have 'run the slate' in the other games. The other teams all have losses to CLEARLY inferior opponents.

To only consider games played amongst one another is forgetting the other half of the picture: no slip-ups. This team didn't lose to Oakland (Pittsburgh, Denver, Cincinnati, Philly). This team didn't lose to Cleveland (Pittsburgh). This team didn't lose to Kansas City (Pittsburgh, Dallas pushed in regulation).

The only team in the entire group that you can make a logical case as being better than San Diego is Baltimore, who both beat San Diego (barely) and has lost only to strong opponents.

197
by randplaty (not verified) :: Fri, 12/25/2009 - 3:59pm

Signature wins? Wins at Dallas and against Phi, both of which are ranked significantly higher than SD in DVOA and both games were fairly solid wins by SD.

177
by jebmak :: Thu, 12/24/2009 - 2:57am

This is a small thing, but on the playoffs odds page, I wish that the teams with zero percent chance of making the playoffs were ordered by estimated wins instead of alphabetically. It just looks weird toward the end of the season when 2-12 teams are above 5-9 teams.

If you could find time to do that, I promise, that it would make my life a tiny bit less uncomfortable next December.

*crosses fingers*

181
by Danish Denver-Fan :: Thu, 12/24/2009 - 7:42am

Isn't it obvious? The Eagles get a boost for being in the, by far, toughest division in football the last decade. Other metrics of succes such as the binary "wins" or "making the playoffs" does not give you a bonus for playing a hard schedule...

182
by JimZipCode :: Thu, 12/24/2009 - 9:29am

Is it possible the Eagles style of play is consistent with what DVOA likes? They almost never run the ball: is it possible that the extreme-West-Coast version of offense they use looks better to DVOA than other offenses do? Higher yards-per-play because they're almost all pass plays, or something?

183
by nat :: Thu, 12/24/2009 - 10:16am

The short answer is "No. It is not possible."

In the team DVOAs there is no distinction made among types of plays - with the possible exception of choosing to go for it or to kick or punt on fourth down. An incomplete pass is equal in value to a zero-yard run. A five-yard pass is equal to a five-yard run. Etc. Team DVOA is all about results, compared to average results for the situation, adjusted for opponents' strength.

It is possible that the Eagles style of play produces better average results than other styles. But that would simply mean that the Eagles are playing better offense - on average. Produce the same results with a run-heavy offense and DVOA will 'like' your team just as much.

It is also possible that DVOA under or over-values certain results. But if it under-values a 40-yard run, it under-values a 40-yard pass to the same extent.

The problem with guessing that DVOA's method of assigning value to plays is badly broken is that VOA correlates very, very well with wins. And if DVOA's value system is only broken in a small way, the Eagles would have to be getting rare results on a lot of plays to affect their average in any meaningful way.

One last point: the Eagles are only 9th in team offense at this point in the year. It's their defense and special teams that are pulling them to the top of the DVOA list.

184
by t.d. :: Thu, 12/24/2009 - 10:54am

so, in other words, the eagles are doomed

194
by Mr Shush :: Thu, 12/24/2009 - 7:21pm

I'm not sure I agree with your conclusion. Plainly the outcome distributions for passes and runs are substantially different. That a pass and a run with the same outcome are scored the same by DVOA does not imply that DVOA might not "over-prefer" passing over running, if it's too keen on certain outcomes that are more likely with a pass than a run. It's also clearly true that different teams' schemes and personnel will lead to different outcome distributions. I really don't see why it's unfeasible that DVOA might be over-valuing something the Eagles do more of than other teams, and the most obvious difference between the Eagles and other teams does seem to be the run-pass ratio. I agree it's far from proved that this is going on, but I don't think it's prima facie implausible.

195
by nat :: Fri, 12/25/2009 - 2:15am

The point being that it's not passing that might be over-valued. It's some kind of result (yards in a particular situation) that is (a) unusual for most teams (b) common for the Eagles, and (c) badly valued by DVOA. Or so the theory must go.

No such theory has been put forward. Instead, the theory was that passing was over-valued, which plainly is not so.