Writers of Pro Football Prospectus 2008

Most Recent FO Features


» The Deep Ball Project

Guest columnist John Kinsley breaks down the tape of every deep pass in the NFL in 2017 and comes away with a shocking conclusion: even without Andrew Luck, the Colts had the best long-ball quarterback in the league.

27 Oct 2009

Week 7 DVOA Ratings

by Aaron Schatz

The Saints may have engineered a big comeback to stay undefeated this week, but they still take a huge hit in the Football Outsiders DVOA ratings. Although the Saints remain at number one, they lost more than ten percentage points off their actual rating and are now less than one percentage point ahead of the second-place team, the Indianapolis Colts.

The Saints weren't the other team to see their DVOA fall despite what looked like an impressive win. Those DVOA darlings, the Philadelphia Eagles, fall from second to sixth in the ratings despite a 27-17 win against Washington. In fact, the Eagles actually end up with a negative DVOA for the game (-5.5%) for a variety of reasons. Opponent adjustments are strong since the Redskins have fallen to 22nd overall. The Redskins fumbled twice on the quarterback-center exchange, which hurts their rating but doesn't help Philadelphia's. The team really stagnated after a hot first quarter, and opened the door for the Redskins to come back. (Unfortunately, the Redskins have no idea what an open door looks like these days.) The Eagles had 55.1% offensive DVOA and -81.5% defensive DVOA in the first quarter. In the last three quarters, they had -60.2% offensive DVOA and 11.6% defensive DVOA.

When it comes to ratings that differ from conventional wisdom, a couple of teams really jumped out at me this week. The first is Green Bay, which has now climbed up to fourth overall. It's obvious now that 2009 will go down as the Year of the Defensive Coordinator, and we've rightfully celebrated the amazing turnarounds orchestrated by Mike Nolan in Denver and Gregg Williams in New Orleans. But over the last couple weeks, without anyone looking -- probably because the opponents were Cleveland and Detroit -- Green Bay has moved into the top spot in our defensive ratings. Now, obviously there is an element of strength of schedule here -- as I just said, we're talking Cleveland and Detroit. DVOA is currently using 70 percent strength for opponent adjustments, so the Packers could get docked in future weeks. Still, while it may be easy to stop the Cleveland and Detroit offenses, the Packers did a better job of it than anyone else. (Against Green Bay, Detroit had its worst offensive DVOA of the year, by far; Cleveland's worst games of the year came against Green Bay and Baltimore.) The Packers only have one defensive performance this year with a positive DVOA, which came against -- oh, just kill me now -- Brett Favre and the Minnesota Vikings.

Anyway, check out the defensive top ten at this point and you'll see that the top eight defenses all have new coordinators. The bottom five defenses also all have new coordinators. Obviously, there's a difference between a new guy doing the same thing (Philadelphia) and a guy who brings in a whole new scheme (Denver), but still, this is weird. Offense is generally more stable than defense from season to season, but this year things are really out of whack. From 2000-2008, the year-to-year correlation of DVOA was .52 for offense and .45 for defense. The correlation of last year's DVOA to this year's rating through Week 7 is .63 for offense and just .18 for defense. Since 2000, there hasn't been a season where the year-to-year correlation of defensive DVOA was less than .31. Even if we take out the defense with the biggest turnaround, Denver, the year-to-year correlation in 2009 is lower than that (.29). There's no trend here, either; from 2007 to 2008, the correlation coefficient for defensive DVOA was .63. This defensive transformation across the league is unique to this season.

The other team that stands out in this week's DVOA ratings is Miami, which now holds the title of "best team with a losing record" by a healthy margin. The Dolphins have played a very difficult schedule (fifth by average DVOA) and have only been outscored by a combined six points. That suggests that the Dolphins have been unlucky, which isn't really true. They haven't lost particularly close games and they actually rank second in the NFL in "hidden" value on special teams, mostly because opposing kickers are just 12-for-17 on field goals with two missed extra points. The Dolphins' bigger problem is inconsistency, especially on defense, where they rank last in VARIANCE.

* * * * *

Housekeeping: All individual stats pages now updated. We've added a new feature on the 2009 stats pages: fumbles on receptions -- that's zero for almost all players, but we figured as long as we were adding the data to our own tables, we might as well add the data to the publicly posted tables. (We'll go back and add to past years in the offseason.) Team stats pages, playoff odds, and FO Premium database are also all updated.

If you are registered to receive Football Outsiders notification e-mails, you may have gotten e-mails irregularly or not at all in recent weeks. We're sorry about that; we're working on the issue. The Premium newsletters seem to be a little sketchy as well, so Premium readers who want to know when we post new picks each week can feel free to join my Twitter feed at FO_ASchatz. I also posted early DVOA ratings this morning, another little gift to those people who want to follow my ramblings.

* * * * *

These are the Football Outsiders team efficiency ratings through seven 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. Because it is early in the season, opponent adjustments are currently at 70 percent strength.

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

DAVE is a formula which combines our preseason projection with current DVOA to get a more accurate forecast of how a team will play the rest of the season. Right now, the preseason projection makes up 19 percent of DAVE for teams with six games and 7.5 percent of DAVE for teams with seven games. DAVE also uses WEIGHTED DVOA, so games from Weeks 1-3 are slightly discounted. Beginning next week, our table will switch from DAVE to WEIGHTED DVOA.

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>

1 NO 43.1% 1 33.2% 3 6-0 32.5% 2 -16.4% 5 -5.7% 27
2 IND 42.4% 3 40.8% 1 6-0 33.4% 1 -9.1% 8 -0.1% 15
3 NE 38.6% 4 38.1% 2 5-2 30.9% 3 -7.9% 9 -0.2% 16
4 GB 35.1% 8 27.5% 4 4-2 17.0% 8 -20.5% 1 -2.3% 23
5 DEN 34.7% 6 23.6% 8 6-0 17.1% 7 -20.4% 3 -2.8% 24
6 PHI 33.0% 2 27.4% 5 4-2 8.4% 16 -20.4% 2 4.2% 4
7 BAL 27.9% 9 25.2% 6 3-3 25.2% 5 -5.9% 11 -3.2% 26
8 MIN 24.6% 7 24.2% 7 6-1 11.1% 14 -3.9% 13 9.5% 2
9 NYG 22.8% 5 22.0% 9 5-2 13.2% 12 -9.4% 7 0.2% 14
10 ARI 22.0% 13 14.7% 12 4-2 4.2% 18 -17.4% 4 0.4% 13
11 DAL 21.5% 12 17.5% 10 4-2 26.8% 4 8.0% 22 2.7% 6
12 PIT 14.6% 11 15.6% 11 5-2 22.6% 6 1.1% 17 -6.9% 31
13 MIA 12.7% 14 7.1% 13 2-4 14.8% 10 1.1% 16 -1.0% 21
14 ATL 7.7% 10 3.0% 17 4-2 9.9% 15 3.1% 19 0.9% 11
15 CIN 5.1% 20 4.1% 16 5-2 15.0% 9 4.1% 20 -5.9% 28
16 JAC 3.5% 15 6.2% 14 3-3 13.3% 11 11.1% 25 1.2% 10
17 NYJ 3.4% 16 1.4% 18 4-3 -10.5% 20 -11.8% 6 2.0% 8
18 HOU -2.7% 18 -2.8% 20 4-3 12.7% 13 18.4% 27 3.0% 5
19 SEA -2.9% 17 -0.1% 19 2-4 -3.7% 19 0.0% 15 0.8% 12
20 SD -4.0% 24 4.1% 15 3-3 5.6% 17 9.3% 24 -0.3% 18
21 SF -6.6% 21 -9.6% 21 3-3 -10.9% 21 -6.1% 10 -1.8% 22
22 WAS -10.0% 22 -9.8% 22 2-5 -14.5% 22 -4.7% 12 -0.2% 17
23 BUF -12.0% 23 -13.5% 24 3-4 -17.6% 24 -3.8% 14 1.8% 9
24 CHI -18.7% 19 -11.2% 23 3-3 -19.1% 27 6.1% 21 6.5% 3
25 CAR -29.8% 27 -23.6% 25 2-4 -17.1% 23 2.8% 18 -9.9% 32
26 KC -31.6% 25 -29.5% 26 1-6 -22.7% 29 11.5% 26 2.7% 7
27 CLE -37.4% 26 -35.1% 28 1-6 -29.9% 31 20.5% 29 13.0% 1
28 TB -44.4% 29 -42.0% 29 0-7 -17.9% 26 25.7% 32 -0.8% 19
29 TEN -44.7% 30 -33.4% 27 0-6 -17.8% 25 20.3% 28 -6.7% 30
30 STL -45.3% 28 -42.3% 30 0-7 -20.4% 28 24.0% 31 -0.9% 20
31 OAK -49.7% 31 -47.2% 31 2-5 -38.0% 32 8.8% 23 -2.9% 25
32 DET -57.2% 32 -48.4% 32 1-5 -27.9% 30 23.4% 30 -6.0% 29

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

1 NO 43.1% 6-0 47.9% 6.3 1 0.4% 14 -14.2% 31 6.4% 7
2 IND 42.4% 6-0 51.6% 5.8 2 -7.8% 26 4.4% 8 14.9% 18
3 NE 38.6% 5-2 39.1% 5.1 5 -3.9% 22 8.1% 7 18.5% 24
4 GB 35.1% 4-2 47.6% 5.0 6 -18.4% 30 -2.1% 21 15.3% 19
5 DEN 34.7% 6-0 38.4% 5.5 3 -3.7% 21 1.5% 15 4.4% 4
6 PHI 33.0% 4-2 48.4% 4.6 10 -17.5% 29 10.2% 5 21.7% 29
7 BAL 27.9% 3-3 31.9% 4.9 8 -0.7% 18 -1.8% 20 10.2% 13
8 MIN 24.6% 6-1 31.8% 5.2 4 -9.8% 28 -4.7% 25 3.6% 1
9 NYG 22.8% 5-2 21.4% 4.1 12 -7.0% 25 12.3% 4 14.8% 17
10 ARI 22.0% 4-2 18.8% 5.0 7 8.0% 7 -21.2% 32 28.4% 31
11 DAL 21.5% 4-2 24.1% 4.8 9 -5.8% 24 10.0% 6 9.3% 12
12 PIT 14.6% 5-2 29.1% 4.6 11 -18.9% 31 2.8% 11 4.2% 2
13 MIA 12.7% 2-4 4.2% 3.9 15 11.5% 5 -3.9% 23 17.4% 23
14 ATL 7.7% 4-2 12.6% 3.9 16 2.5% 11 0.5% 17 16.5% 21
15 CIN 5.1% 5-2 8.0% 4.0 14 7.6% 9 -12.2% 29 16.7% 22
16 JAC 3.5% 3-3 1.3% 4.0 13 -4.5% 23 -4.2% 24 21.3% 28
17 NYJ 3.4% 4-3 12.6% 3.6 17 -2.1% 20 2.6% 12 13.7% 15
18 HOU -2.7% 4-3 5.8% 3.6 18 -9.6% 27 3.8% 9 6.6% 8
19 SEA -2.9% 2-4 -6.0% 2.7 23 -0.4% 16 -10.9% 28 20.9% 26
20 SD -4.0% 3-3 -5.6% 3.4 19 1.2% 12 -6.3% 27 4.2% 3
21 SF -6.6% 3-3 -9.3% 3.0 21 0.5% 13 -3.7% 22 10.3% 14
22 WAS -10.0% 2-5 0.6% 2.9 22 -21.8% 32 14.5% 2 5.3% 6
23 BUF -12.0% 3-4 -13.1% 3.1 20 -2.0% 19 3.3% 10 15.4% 20
24 CHI -18.7% 3-3 -16.1% 2.6 24 0.3% 15 2.3% 14 21.2% 27
25 CAR -29.8% 2-4 -35.5% 1.8 25 -0.6% 17 19.3% 1 13.8% 16
26 KC -31.6% 1-6 -31.3% 1.7 26 5.9% 10 -1.2% 19 9.0% 11
27 CLE -37.4% 1-6 -50.5% 1.6 27 18.6% 1 -12.2% 30 22.5% 30
28 TB -44.4% 0-7 -43.4% 0.8 32 9.1% 6 13.3% 3 7.2% 9
29 TEN -44.7% 0-6 -57.1% 1.3 29 14.2% 2 0.8% 16 35.4% 32
30 STL -45.3% 0-7 -53.0% 1.5 28 12.3% 4 -5.1% 26 4.9% 5
31 OAK -49.7% 2-5 -56.4% 1.0 30 7.9% 8 2.3% 13 19.2% 25
32 DET -57.2% 1-5 -65.2% 0.9 31 12.7% 3 0.4% 18 9.0% 10

Posted by: Aaron Schatz on 27 Oct 2009

189 comments, Last at 30 Oct 2009, 3:04pm by Andrew Potter


by Anonymouse (not verified) :: Tue, 10/27/2009 - 6:14pm

"DAVE also uses WEIGHTED DVOA, so games from Weeks 1-3 are slightly discounted. Beginning next week, our table will switch from DAVE to WEIGHTED DVOA."

So...is the highlighted column Weighted DVOA (what it says) or DAVE (what it looks like)?

CAPTCHA: Sidonia washroom

by Aaron Schatz :: Tue, 10/27/2009 - 6:18pm

Oops. Fixed.

by sfckoski :: Tue, 10/27/2009 - 6:18pm

You said 9 of the top 10 in defensive DVOA have new D-Cors?

The 49ers are number 10 and Greg Manusky was the DC last year. Unless you're counting Nolan as HC/DC last year?

by Will Allen (not verified) :: Tue, 10/27/2009 - 6:20pm

I can't remember; is it unusual for 6 spots in the top ten DAVE (4-9 in this case) to be this close together?

by Anon1234 (not verified) :: Tue, 10/27/2009 - 6:26pm

The Saints have played 6 games and their estimated wins is 6.3??

by Will Allen (not verified) :: Tue, 10/27/2009 - 6:28pm

Teams that have had a bye have their numbers projected to assume a 7th game. Or something like that. I know what I am talking about. Really.

by Bobman :: Tue, 10/27/2009 - 6:56pm

How is that NOT in the boilerplate--it comes up every week as soon as byes start.

by Tundrapaddy (not verified) :: Tue, 10/27/2009 - 8:20pm

"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."

Yeah, one would think they'd include that bye week thing in the description of the forest index...

by Kevin from Philly :: Wed, 10/28/2009 - 11:36am

The extra 0.3 is obviously a Bourbon Street "Huricane adjustment".

by PatsFan :: Tue, 10/27/2009 - 6:27pm

Pats are #3 (and #2 in DAVE)?! Sure, they cruised the past two weeks, but over tomato cans.

I still think IND and NO are going to smoke NE big time (like 17+ or 21+ margins of victory) this year, especially since NE is on the road for both of them.

by Led :: Tue, 10/27/2009 - 6:42pm

Nice try for the reverse jinx, PatsFan. But nobody's buying it!

by Purds :: Tue, 10/27/2009 - 6:43pm

I had thought so, too, but perhaps NE is getting better, but we're just brushing off their improvements because of the teams they've been playing. If NE is improving, it would be hard to discern that, as the last two opponents were really weak (as were the Colts last two--only NO has played anyone with talent the last week). We might be using the first 4 weeks as too strong of an impression on New England.

by Rich Conley (not verified) :: Tue, 10/27/2009 - 6:53pm

The first 4 weeks include narrow losses to what is now a 6-0 Denver team and a Jets team that was playing very good ball at the time (jenkins getting hurt is huge)and dominating what looked like a good Baltimore team.

Yeah, the Buffalo game was a stinker, but it was Brady's first game back.

Also, this Pats defense is already better than its been in a couple years, and thats surprising seeing how theres a whole bunch of rookies getting a ton of playing time (Butler, Chung, Pryor).

by Tundrapaddy (not verified) :: Tue, 10/27/2009 - 8:24pm

I think the 'bunch of rookies' is a two-edged sword. They may lack NFL experience, but you know you're getting better read-and-react/response time than what you got with the ancient (NFL standards, anyhow) players they've replaced.

by Temo :: Tue, 10/27/2009 - 8:47pm

As I remember, they also outgained Buffalo by large amounts despite the close game so I'm sure DVOA liked that.

by bmerryman :: Tue, 10/27/2009 - 9:20pm

"dominating what looked like a good Baltimore team."

I think the DVOA numbers for that game were good for both teams. 25.9% NE and 21.4% BAL. Not really what I'd call dominating.

by nat :: Tue, 10/27/2009 - 9:39pm

When you compare two teams in the same game, you should use VOA, not DVOA.

by bmerryman :: Wed, 10/28/2009 - 8:16am

I don't have access to that info...I think. It just shows DVOA for a specific week. Where do you find VOA for a specific week? You'd have to think by any stretch of the definition of domination, even the margin in DVOA from one opponent to another would be significantly more than 4.5%.

by nat :: Wed, 10/28/2009 - 8:47am

I don't know where to get it, either, unless it shows up in an FO feature.

I don't knock your comment on "domination" - the Patriots earned that win, but a domination it was not.

But DVOA is the wrong stat. Danny DeVito in six-inch stiletto heels is not taller than Shaquille O'Neal in flip-flops.

by steelberger (not verified) :: Wed, 10/28/2009 - 9:30am

Agreed. Not sure how you can say New England dominated the Ravens. The Pats were outgained by nearly 50 yards and had Clayton(?) not inexplicably dropped a first down pass at the Pats 10 yard line that hit him squarely in the chest, the Ravens very well may have won that game.

by dryheat :: Wed, 10/28/2009 - 11:47am

Fourth down pass.

If football games were three quarters long, maybe the Pats dominated, but they let Baltimore right back into it at the end. I thought Baltimore was the best team in the league at the time, but I think some shine's come off the win at this point.

by steelberger (not verified) :: Wed, 10/28/2009 - 2:31pm

I know the pass took place on fourth down, I meant the result of the pass would have been a first down.

Still not seeing the domination angle. It was a 3 point game with 14 minutes left in the 4th quarter. The Ravens threw for more yards. The Ravens ran for more yards. The Ravens had more sacks. Both teams had 1 turnover. Both teams punted 3 times.

And before anybody gets their panties in a twist and says I "hate the patriots", I am by no means saying they didnt deserve or earn the win. I am just saying that it was by no means a "domination". That was about as even a game as we have seen this year.

by anotherpatsfan :: Wed, 10/28/2009 - 5:46pm

Of course you hate the Patriots...

However, I too have a hard time seeking that as "domination" by the Pats. Definitely a good win against at least a fairly good team in today's NFL...

by MJK :: Wed, 10/28/2009 - 6:53pm

Agree with anotherpatsfan--it's well known that you hate the Patriots. And that's OK. I'd rather have a discussion with an intelligent fan that hates the Patriots than with a Patriots fan who's not from/never lived in New England (and is also a fan of the Lakers, the Yankees, USC...you get the idea).

by Andrew Potter :: Wed, 10/28/2009 - 7:10pm

Being a Patriots fan who's never even been to New England (or really anywhere Stateside other than Walt Disney World/Orlando, for that matter), much less lived there, I'll choose to interpret that statement as "a glory-hunting Patriots fan". ;)

(Unfortunately, it's a little trickier on this side of the Pond to support one's local team since NFL Europe went the way of the Dodo.)

As for the Baltimore game, I watched that here and at no point did I feel comfortable - or that the Patriots were comfortable - as I would expect with a dominating win. I was at Wembley this past weekend - THAT was dominant. Tennessee the week before was dominant. Washington in 2007 was dominant. Baltimore? Not so much.

by MJK :: Wed, 10/28/2009 - 7:50pm

Yes, of course, if there is no local NFL team to root for, you're forgiven. :-) Or anyone who's been a Pats fan since the Pete Carroll years, or before.

And I didn't think the Baltimore game was "domination". I didn't get to see it...I just listened to it on the radio, but I was never comfortable. And I don't think Belichick was, either.

Incidentally, if you're going to come all the way to North America again, there are many, many better places to go than Florida. Don't you have a Disney World somewhere in Europe? But there are some amazing wonders scattered around this continent worth visiting...the Rocky Mountains, the amazing national parks in the great southwest, the redwoods of California, New England in autumn...and I'm sure there are a bunch more places I haven't been that are equally stunning.

by Andrew Potter :: Wed, 10/28/2009 - 8:22pm

I definitely pass the Pete Carroll test. In fact, I go beyond that to the 1-15 and 2-14 seasons of the early 90's. Until 1996, I was cursing my Dad for bringing me back a Patriots shirt from Boston instead of a shirt for one of the good teams. (My sister got a Dolphins cap because... umm... she liked dolphins. Safe to say Dad doesn't understand the game.)

Incidentally, if you're going to come all the way to North America again, there are many, many better places to go than Florida. Don't you have a Disney World somewhere in Europe? But there are some amazing wonders scattered around this continent worth visiting...the Rocky Mountains, the amazing national parks in the great southwest, the redwoods of California, New England in autumn...and I'm sure there are a bunch more places I haven't been that are equally stunning.

Again, blame my parents; I was 10 years old the first time and 18 the second, and both were family holidays. My best friend lives in Colorado and her family has a cabin in the Rockies, so I'm determined to make that visit. New England (obviously) and Seattle are both on my list of places to see, and I've good friends in Arkansas, Tennessee, Oklahoma, and Michigan.

Now if I could just get this danged money tree to take root...

by Raiderjoe :: Wed, 10/28/2009 - 8:41pm

Yosemite park also good. Another place to go is Norht Carolina outer banks. Black Hole could place to vidsit. Maybe could go there in 2010 becuause if Raiders and Pates both finish in 1st place this year then in 2010 Raoders going to play Pates at Oakland. Washington dc crappy unless US history biff. Las Vegas good .Never wnet to Grand cannon but would like to. also want to go to Seattle Oregon areas

by MJK :: Thu, 10/29/2009 - 12:10pm

Heh heh. So is your sister now a Dolphins fan?

I'm not really one to criticize (although I do)...I have variously been a fan of teams other than the Patriots. Though it's not my fault!

I was ins second grade in 1986, and knew nothing about football. For some strange reason, I was fond of the colors teal and orange, so my aunt, who also knew nothing about football, bought me a Miami Dolphins sweater. I loved that thing and started wearing it all the time once it got cold...which of course, was exactly when the Patriots went up against their hated division rivals, the Dolphins, in the AFC Championship Game. Needless to say, the experience was a little traumatic. Immediately thereafter, my dad decided to teach me about the game so I wouldn't make such a mistake again, and I've been a football fan since.

However, the late 80's and early 90's Pats were so bad, they were never on TV. I don't know if that was because their games were getting blacked out, or if it was just because my dad refused to watch them suck so much, but all I ever remember seeing on TV in those years were the Bills, who were pretty fun to watch back then. So for about four years, I'm ashamed to say, I was a hardcore Bills fan. Once I got old enough to make my own football watching decisions (and once the Pats stopped being blacked out) I properly became a fan of the hometown team (it helped that Kelly and Thomas retired and the Bills started to suck). So I was a bandwagon fan for a while, but of a different team.

by PatsFan :: Wed, 10/28/2009 - 8:45pm

Or anyone who's been a Pats fan since the Pete Carroll years, or before.

Glad to hear someone who's been a fan since 1976 isn't a bandwagoner :)

by Anonymous1 (not verified) :: Wed, 10/28/2009 - 12:21pm

Baltimore had a nice first drive. From that point until their final drive of the game they were completely dominated and essentially leveraged two successful plays (a strip sack returned for a TD and a poorly defended Ray Rice draw play) to still be in that game. Teams as badly outplayed as Baltimore was will usually be down by 2+ plus scores at the end. The fact that the game ended close is not an accurate reflection of how each team performed.

by MJK :: Tue, 10/27/2009 - 8:06pm

Possibly true.

Brady has certainly looked better each week, and I'm reminded of Manning last year. Of course, it's easy to look better versus the feared defensive tandem of TEN and TB, and he certainly had more time to throw, which helps, but more significantly, he hasn't been blatantly missing receivers anymore (maybe cutting Galloway helped that).

On the other side of the ball, it's more or less a completely new defense with a lot of young guys. New defenses often take a few games to gel, so I could buy an improvement there.

Still, I think the Colts and New Orleans are better teams right now. I think the Pats have maybe a 40% chance against either of the, and only a 16% chance of beating them both.

by Rich Conley (not verified) :: Tue, 10/27/2009 - 10:31pm

"Still, I think the Colts and New Orleans are better teams right now. I think the Pats have maybe a 40% chance against either of the, and only a 16% chance of beating them both"

I would agree that both the Colts and NO are better teams right now, but I have no idea if that will be the case in a couple weeks.

Indy will be a tough tough game, and I dont think the chances are good, but I think they beat New Orleans.

Both sides of the ball are improving rapidly, and its pretty clear at this point that the Patriots of week 1 are not the same team as the Patriots of Week 7

by FooBarFooFoo (not verified) :: Wed, 10/28/2009 - 2:57am

I think the Saints will run out of steam come end of November or December. If they play more "winning experienced" (stupid term) teams which play them physically, they look much less good on offense pretty easily. No way they will beat the Pats by 20 points. I even think that the Colts Pats game will be close, with a slight edge for the Colts (unfortunately).

Good teams win ugly in September and October - I am a Pats fan but not saying they are a real contender this year - just like the Pats did against the Bills. Winning ugly means a lot of room for improvements plus a lot of humble pie from the HC. But they should have won one of the two losses (Broncos, Jets). Teams that look sharp in September always fall apart at the end of the season. As much as I would like to see the Saints succeed, they won't make it to Miami again. I think they are pretty much playing at their limit. Looking at the Pats Bucs game, the Pats are either not playing near their limits or are a pretty crappy team. That game was not good.

by ukwhodat29 (not verified) :: Wed, 10/28/2009 - 8:47am

So the Eagles and Giants aren't "winning experienced" or physical teams then? I seem to remember the Saints giving both of them a good kicking.

And Miami were pretty physical this week too, so I'm not sure you can use that line of thought (unless, like most non-Saints fans, you still think of them as a 'finesse' team).

Really looking forward to the Saints 2 MNF games, wins in both might even make a few more people realise they're for real this season.

by FooBarFooFoo (not verified) :: Wed, 10/28/2009 - 3:35pm

The point is that the Saints are playing near their limit, that it is impossible to do that throughout a whole season, that it is downright stupid to play well at the start of the season, that near the end of a season other teams get increasingly better at throwing bricks into offenses which looked well-tuned at season start. And that smart teams and smart coaches know that the real season starts in November. Throw your crappy power rankings out the window. Waste of time.

And exactly, the Dolphins almost beat the super-duper invincible Saints.

We can talk again after the Saints lost their first playoff game. CU. Until then, you can bury your money by betting on the Saints to win the Super Bowl.

by Andrew Potter :: Wed, 10/28/2009 - 8:50pm

The point is... that it is downright stupid to play well at the start of the season... And that smart teams and smart coaches know that the real season starts in November.

A good-to-great November/December doesn't mean a thing if you don't do enough in September/October. The real season starts in September, no matter how much anybody shouts otherwise. I very much doubt you would get a single NFL coach, smart or otherwise, who either says or thinks any differently.

And exactly, the Dolphins almost beat the super-duper invincible Saints.

We can talk again after the Saints lost their first playoff game. CU. Until then, you can bury your money by betting on the Saints to win the Super Bowl.

Bitter, much?

by FooBarFooFoo (not verified) :: Thu, 10/29/2009 - 10:09am

There is a slight difference between doing "enough" and being on the top of every power ranking, every super bowl prediction and trying to break the scoreboard in each game during the irrelevant part of the season.

The first may be smart, the latter is stupid.

by Andrew Potter :: Thu, 10/29/2009 - 10:47am

Ah, yes, I see what you mean now; it didn't work out very well for either the 2006 Colts (scored 30+ points in 4 of their first 8 games) or the 2007 Patriots.

by Mr Shush :: Fri, 10/30/2009 - 6:11am

Ok, but there was this team a few years back that started the season 6-0, blowing out all and sundry.

Week 1: BAL W 27-10
Week 2: bye
Week 3: ATL W 35-7
Week 4: @CIN W 38-10
Week 5: SF W 42-20
Week 6: @ATL W 41-13
Week 7: CLE W 34-3

Yup, those 1999 St. Louis Rams sure regretted peaking too early. What were they thinking?

by Eddo :: Fri, 10/30/2009 - 1:17pm

I don't know if the 1999 Rams are a great example; they very nearly lost in both the NFC Championship and the Super Bowl, and their offense didn't play all that well in either.

by Andrew Potter :: Fri, 10/30/2009 - 3:04pm

Well I was being sarcastic in bringing up the 2006 Colts (Super Bowl winners) and 2007 Patriots (AFC Champions). The argument that "dominating in the first half of the season is stupid" just doesn't make sense; the 1999 Rams are the perfect example of that. Even the 2005 Colts with their AFCC loss back up the idea that dominating is dominating whenever it happens - it took a dynamic, outstanding defense putting on a performance for the ages AND a huge slice of luck (Vanderjagt's missed FG) to sink that team, just as it did the 2007 Patriots (Manning-to-Tyree).

Plus, we all know how random the playoffs can be - even going one-and-done in the playoffs doesn't mean you're not as good as advertised. The Saints are tearing it up; you'd be perfectly justified in claiming they're the best team in the league right now (personally I still think that's the Colts, much as it pains me).

by Sophandros :: Wed, 10/28/2009 - 9:04am

Nice use of cliche's, brah.

Anyway, you may want to read the article on Guts vs. Stomps...

Sports talk radio and sports message boards are the killing fields of intellectual discourse.

by DeltaWhiskey :: Wed, 10/28/2009 - 9:13am

I can only hope that "brah" was meant to be ironic.

by Rich Conley (not verified) :: Wed, 10/28/2009 - 10:29am

The guts vs stomps article is a great example of misinterpreted stats.

by roguerouge :: Wed, 10/28/2009 - 11:08am

How so?

by RickD :: Wed, 10/28/2009 - 11:57am

Nice use of apostrophe'''''''s

by RickD :: Wed, 10/28/2009 - 11:59am

Seriously, when did people start using apostrophes with plurals all the time? I feel like I'm becoming an old fart, but when I was a kid, I don't remember seeing this happening all over the place.

by Temo :: Wed, 10/28/2009 - 12:56pm

When did people start replying to their own posts? When did people start asking rhetorical questions?

by RickD :: Wed, 10/28/2009 - 1:14pm

Rick here again. Sorry, but we had a minority voice in the brain that felt it must be heard.

by M :: Wed, 10/28/2009 - 1:21pm

I've become very confused as to the correct use of apostrophes with regard to plural nouns. I always thought they should be used (and only used) in conjunction with abbreviations (The DVOA's of the Lions and 49ers are surprisingly similar this decade). However, I've already contradicted myself by using "49ers" vs. "49er's". Any thoughts on what is correct? I've had two English-major types say apostrophes are ALWAYS wrong, but I'm not so sure...

by tuluse :: Wed, 10/28/2009 - 1:22pm

Maybe you mean acronyms and not abbreviations.

by M :: Wed, 10/28/2009 - 1:35pm

Thanks. I knew somehow I intellectually sabotaged myself with my post, I just didn't know where.

by dryheat :: Wed, 10/28/2009 - 1:33pm

I would suggest that DV'sOA might be the correct plural.

by Omaholic (not verified) :: Wed, 10/28/2009 - 5:32pm

Just like R'sBI?

by zlionsfan :: Wed, 10/28/2009 - 1:40pm

Here's a guideline that seems to spell it out fairly well, courtesy of grammar.ccc.commnet.edu:

We use an apostrophe to create plural forms in two limited situations: for pluralized letters of the alphabet and when we are trying to create the plural form of a word that refers to the word itself. Here we also should italicize this "word as word," but not the 's ending that belongs to it. Do not use the apostrophe+s to create the plural of acronyms (pronounceable abbreviations such as laser and IRA and URL*) and other abbreviations. (A possible exception to this last rule is an acronym that ends in "S": "We filed four NOS's in that folder.")

Based on that, you would say "The DVOAs of the Lions and 49ers are surprisingly similar this decade." (Also note that 49er was originally an abbreviation and you were talking about acronyms. If we were still using the abbreviated form, it should be '49ers.)

by Eddo :: Wed, 10/28/2009 - 1:47pm

As an addition to your point, zlionsfan, in the Chicago Style Manual, I believe, it suggests that using apostrophes to pluralize single letters is OK, but not necessary, unless single letter + "s" would be a valid word.

That is, you can say either "I got all B's this term" or "I got all Bs" this term, but you must say "I got all A's", since "as" is a valid English word.

by RickD :: Wed, 10/28/2009 - 11:41pm

Here's another guide, in cartoon form.


Or rather, "a simple guide to some basic grammar, you illiterate morons" as
Bob the Angry Flower says.

by RickD :: Wed, 10/28/2009 - 1:16pm

"the feared defensive tandem of TEN and TB".

Wow. Things change fast in the NFL, don't they? TEN was indisputably elite as recently as last season, and it wasn't much longer than that for TB, was it?

by PatsFan :: Wed, 10/28/2009 - 2:16pm

Well, at least when they played NE, TEN had three of its starting DBs injured and out of the game.

by FooBarFooFoo (not verified) :: Wed, 10/28/2009 - 3:40pm

If your players quit on your coach (TEN not TB), you can field 11 genetically engineered, first ballot hall of faming-orang-utans with Peytons head (and smarts and aerobic moves) and will lose any game by 40. Sorry, but stop talking about Tennessee. They should return the paychecks they received for the Pats game (just like Irvin said). That was disgusting.

by RickD :: Wed, 10/28/2009 - 11:44pm

Indeed. I have watched a lot of football and I haven't seen a team quit like that since, well, actually only since last December when the Cardinals did the same thing
in New England.

Buck up, Titans fans! There's a Super Bowl appearance in your future!

by Big Johnson (not verified) :: Tue, 10/27/2009 - 10:00pm

yeah i completely agree. when new england smashes tomato cans they get high dvoa for it. When san diego smashes tomato cans nothing happens. i think they moved up 2 spots for beating a team by 30 points. pretty sweet!

by RickD :: Wed, 10/28/2009 - 11:54am

The Pats handled both Baltimore and Atlanta, and those are both very good teams.

I would be very surprised if the Pats were "smoked" by anybody. Their defense has improved quite a bit and their offense, while not at the level of 2007, is still excellent.

by Anonymous1 (not verified) :: Wed, 10/28/2009 - 12:23pm

I still think IND and NO are going to smoke NE big time (like 17+ or 21+ margins of victory) this year, especially since NE is on the road for both of them.

Not a chance. NE will win at least one of those games and the other will be close.

by bravehoptoad :: Wed, 10/28/2009 - 2:37pm

Nice analysis. Love the way you break these things down.

by Anonymous1 (not verified) :: Wed, 10/28/2009 - 3:00pm

Yes, because every post on here breaks out charts to prove their positions. I certainly am humbled seeing your legwork.

by Pats fan.... (not verified) :: Thu, 10/29/2009 - 3:05pm

Gee, I don't know which to believe:

2. What PatsFan thinks.

What a tough choice.

by ugarte (not verified) :: Tue, 10/27/2009 - 6:31pm

How is it possible to have more expected wins than games played? I understand that the bye week means that the Saints have had 7 weeks since the start of the season but no matter how well they played they couldn't have had 7 wins.

by Will Allen (not verified) :: Tue, 10/27/2009 - 6:34pm

Aaron is in depths of a peyote binge.

by Bobman :: Tue, 10/27/2009 - 6:58pm

Depths? Heights? Either way, LMAO.

Now you're just being mean.

by Gruntled (not verified) :: Tue, 10/27/2009 - 8:48pm

Definitely heights. It's been 40 years, but the memory lingers.

by BroncosGuy (not verified) :: Tue, 10/27/2009 - 9:40pm

Will has become mean since the season-ending injury.

by Bobman :: Wed, 10/28/2009 - 12:13am


Though I think even joking about peyote gets you a suspension from Goodell, so he better not cross the line. Remember, Will, one minute it's all fun and games and the next minute your picture is hanging in the post office over Hunter S. Thompson's rap sheet.

by Will Allen (not verified) :: Wed, 10/28/2009 - 12:53am

I am unconcerned, due to the talents of my 300 pound Samoan attorney, and really, until the jar of "E" is opened, it's all good. Of course, there is nothing more helpless and irresponsible than a man in the depths of an ether binge.

by dryheat :: Wed, 10/28/2009 - 8:11am

I'd hate to advocate drugs, alcohol, violence, or insanity to anyone, but they've always worked for me.....

This is exactly what I need to get me through a morning of dreadful meetings. Mahalo.

by loki (not verified) :: Tue, 10/27/2009 - 6:36pm

Presumably it's just extrapolated to 7 games for ease of comparison.

by ChrisH :: Tue, 10/27/2009 - 8:01pm

Just to quote the description of estimated wins:

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

So, New Orleans, and everyone else, has estimated wins with a maximum of 7 at the moment, no matter how many games they have played. I looked at the NO thing and thought the same thing, but I went and read the description instead of posting a comment about it.

by loki (not verified) :: Tue, 10/27/2009 - 6:35pm

With the Variance measurement, does that take work out the variance of each phase and then add the values together? Or is it just variance in total DVOA?

by C (not verified) :: Tue, 10/27/2009 - 6:36pm

The Giants have a higher win projection than Philly but are rated lower by DVOA...
The Packers have a higher win projection than Minnesota but are rated lower by dvoa

Washington at 22?
Really? Because they lost to 2 good teams ( and scored some garbage TD's to make it look closer), and are almost .500 against really really crappy teams?

Glad I own NFL futures and win total props on the GB OVER...

by Dales :: Tue, 10/27/2009 - 6:45pm

"The Giants have a higher win projection than Philly but are rated lower by DVOA..."

Likely caused by Philly's higher variance.

Variance should be a good thing for bad teams and a bad thing for good teams.

As for Washington, it does seem too high until one looks at who is below them. The only one I think is one to ponder is Chicago.

by JasonK :: Tue, 10/27/2009 - 7:42pm

I think it's more likely caused by the Giants being 5-2 and the Eagles being 4-2 right now. All else being equal, it's easier to get 5 wins in 9 remaining games than it is to get 6 wins in 10 remaining games.

by Dales :: Tue, 10/27/2009 - 7:56pm

Except that estimated wins are based off of 7 games right now for both teams, and estimated wins does not consider at all how many wins a team already has.

by nat :: Tue, 10/27/2009 - 8:11pm

He carefully said "win projection" - from the playoff odds page, not "estimated wins" from the DVOA page. The Giants still have the higher "mean wins" projection. That's done by simulating the remaining games in the season, and assumes your current record is what it is.

by Dales :: Tue, 10/27/2009 - 8:53pm

I stand corrected. I thought he was referring to the estimated wins and my comment was regarding that.

by Dales :: Tue, 10/27/2009 - 8:54pm

I stand corrected. I thought he was referring to the estimated wins and my comment was regarding that. Of course, if I had read carefully, I would have seen that Phi has more estimated wins and would have known I was missing something.

by nat :: Tue, 10/27/2009 - 8:07pm

My guess would be that it is caused by winning one more game already.

The Giants now trail the Eagles in DAVE, DVOA, mean (remaining) wins, and estimated wins (even with PHI's higher variance). But, with their extra win in the bag, the Giants' mean wins for the season is 10.3 to the Eagles' 10.0. To put it another way, DVOA says they average 5.3-3.7 for their last 9 games, with the Eagles averaging 6-4 for their last 10.

With a few weeks until opponent adjustments reach full strength, I'd say the Eagles have the edge when they meet (barely), but the Giants have the edge for winning the division. It's enough to make me root for the Redskins. Nothing could make me root for the Cowboys.

by wr (not verified) :: Tue, 10/27/2009 - 6:45pm

I was wondering if NO wouldn't take a DVOA hit with they way they played (despite winning). Looks like it was significant.

[mischief]And I am shocked, just shocked, that Philly didn't take over the top
spot this week [/mischief]

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 10:00am

"I was wondering if NO wouldn't take a DVOA hit with they way they played (despite winning). Looks like it was significant."

Similarly, I thought Minnesota might not suffer much because they played so well in Pittsburgh, but nope.

by countertorque :: Wed, 10/28/2009 - 10:37pm

Yeah, well, the Steelers pay off Aaron too, to make sure opponents get unfair DVOA rankings.

by Aussie Bengal (not verified) :: Tue, 10/27/2009 - 6:51pm

Glad to see the Bengals finally getting a little bit of DVOA love here. 4 estimated wins, a middle-of the road DVOA and high variance looks to fit this team to a "t". I'm a little surprised, though, that one STOMP could lead to a 10% DAVE swing. Is there something I'm missing? I'm assuming GB's ascent to the top 5 helps, as does the improved special teams (new long-snapper, fewer bad snaps). But one week?

by Bobman :: Tue, 10/27/2009 - 7:02pm

Hey, whatever happened to the school of thought that went "Denver's early slate was easy, but their last 10 games are tough..." because as I read it, 15 ain't so bad for a remaining sked. Hell, I'd take it. (I know, it's dynamic based on changing DVOAs and not on last year's records, like traditional measures. But suddenly, Den is very hard for the doubters to dismiss.)

by Pio (not verified) :: Tue, 10/27/2009 - 7:25pm

I think folks were saying more that Denver's first 4 games were easy but they were screwed after that - certainly they were generally expected to lose to NE and SD. At this point I don't think anyone is expecting Denver to do any worse than 50-50 on the remaining 10 games, and IND is the only near-sure loss on the list (partly since Denver has tended to have trouble with Indy for the past half decade). Personally I would guess Denver ends up somewhere like 12-4 or 13-3, with the losses coming to Indy and some combination of SD, Pittsburgh, the Giants, and maybe some fluke game.

by t.d. :: Tue, 10/27/2009 - 9:33pm

What's Peyton's record against Denver when they're playing a 3-4?

by Raiderjoe :: Tue, 10/27/2009 - 8:46pm

dnever broncos going to fall apart in second half of season. Raiders going to catch them. may not seem lielylt o you now butr you just wait and see. It will happen. Raiders 31sth in DAVE standings but have better record thsn 6 teams so when start winning Raiders going to get higher with DAVE.

by HAL 9000 (not verified) :: Wed, 10/28/2009 - 10:32am

I'm sorry Dave. I'm afraid I can't do that.

by bravehoptoad :: Wed, 10/28/2009 - 10:47am

In the name of all that's holy, why hasn't anyone said this before about DAVE?

by M :: Wed, 10/28/2009 - 1:28pm

"...Raiders going to get higher with DAVE."
Maybe THAT explains why 4 of their losses have been by a combined 134-16. I think this DAVE character is a bad influence on them.

by Otis Taylor89 :: Tue, 10/27/2009 - 7:11pm

AZ defense #4, offense #18.

If they can get Bennie to do what he do then they will be trouble for anyone.

by rk (not verified) :: Wed, 10/28/2009 - 12:21am

I thought Bennie was on the Jets.

by Kevin from Philly :: Wed, 10/28/2009 - 3:39pm

That guy was cut for having too many uniform violations: electric boots, mohair suits, etc. I read that in ESPN the maga-za-een.

by jayinalaska :: Wed, 10/28/2009 - 4:20pm

Nicely played!

by Keith (1) (not verified) :: Tue, 10/27/2009 - 7:24pm

I am a huge Packers fan, so I will take the top-ranked defense, but as a guy that watches them pretty much every week, I cannot support the idea that they are a top-ranked defense. They get beat along the line quite often and in their secondary. It always seems like their safeties are in the wrong place and if their line does not get an immediate push, they can be beaten.

Maybe I am just too pessimistic...

by Arkaein :: Tue, 10/27/2009 - 8:16pm

As another Packers fan, I don't think the #1 ranking will hold up (full oppoennt adjust would probably knock them down to #5, considering they're basically in a three-way tie for first with two other teams close). However I think you are being a bit pessimistic.

I don't think they get beat along the line that much. They haven't genearted a ton of sacks, but were very productive against Chicago and Detroit and had decent pressure against Cleveland. The run defense has mostly been very solid apart from the Cedric Benson missed tackle debacle.

I also think concerns about our safeties might be more fears due to lack of depth rather than actual substance. GB has given up the following big pass plays:

- Hester and Knox against Chi
- flea flicker and ridiculous screen pass on 3rd and forever against Cincy
- a couple of goofy FB plays against the Rams
- deep pass to Harvin and communication breakdown to Berrian against Minny

That's about it. The Rams plays were really LB coverages, and the Berrian play was due to a backup who is now back on the bench behind Bigby. The screen pass was all due to horrendous tackling. Favre definitely diced apart GB's coverages, but it was mostly intermediate stuff.

Most of the problems with the new zone coverages have been giving up underneath stuff, Harris in particular doesn't have the knack Woodson does for jumping short routes, but he still is terrific covering go routes.

They also haven't given up any big running plays I can think of except to Benson.

GB won't likely finish the year #1 in DVOA, but I think they have a chance for real improvement as players continue to learn the scheme and Raji and Matthews continue to improve.

by Rich Conley (not verified) :: Wed, 10/28/2009 - 10:34am

"They haven't genearted a ton of sacks, but were very productive against Chicago and Detroit and had decent pressure against Cleveland."

To be fair, aside from Joe Thomas in Cleveland, those lines are all pretty much terrible.

by Arkaein :: Wed, 10/28/2009 - 12:33pm

Absolutely, which is why my optimism is on the cautious side, but it's still a lot better to look good against poor opposition than to look bad.

Also, with regards to Joe Thomas, Cullen Jenkins actually got his sack directly against Thomas, so there's at least one data point in favor of GB's D-line being able to handle high quality opposition.

This week's game against Minnesota should show us more. They've proved they can beat down inferior offenses, and they proved they can stop a great RB with good run blocking line if they fully commit to it. We'll have to see if they can be successfully playing a more balanced defense as well, assuming they modify the approach used in the first meeting.

by t.d. :: Tue, 10/27/2009 - 9:39pm

The Packers have the same DVOA special sauce problem the Eagles have for the last two years, at least. I also think the Packers not only beat up on a terrible Detroit team, but they beat up on a Detroit team playing with their third string quarterback, as well as beating up on a Cleveland team that has committed to a low percentage, big play quarterback who had recently traded away their best big play receiver (shitty as he was). There defense is pretty good, but not #1 good.

by Sean D (not verified) :: Tue, 10/27/2009 - 7:25pm

Just a cursory count gives me 17 teams with new defensive coordinators (that number is more likely to be low than high because I only count the ones I was sure of). So, its not that unlikely that you could end up with the top 8 and bottom 5 all with new ones. Especially, given that defenses are volatile and that the bottom defenses are likely to have the same bad personnel with a new DC. The top 8 is more interesting, but could just be a coincidence given the amount of DC turnover.

by Mr Shush :: Tue, 10/27/2009 - 7:39pm

Not a total coincidence at either end, you would think - teams with great defenses are more likely to lose co-ordinators, as they get hired to be head coaches elsewhere, and teams with terrible defenses are more likely sack d-cons. That's clearly not the whole story (it doesn't explain Denver at all), but it's certainly a part of it.

Also, I'm pretty sure your 17-count is indeed too low.

by Mr Shush :: Tue, 10/27/2009 - 7:33pm

I'm not the only one who's noted that there appears to be an unusual plethora of god-awful teams this year. I think the number is ten. DVOA thinks it's eight. The two teams present in my terrible ten but absent from DVOA's dreadful decuria (yes, a decuria was eight men. No, I don't know why either) are Buffalo and Washington, both of whom DVOA thinks are better than Chicago, and indeed only bad in a fairly ordinary way. At the risk of bringing down the FOMBC on one or more teams about which I could not care less, DVOA is, um, wrong about this. Probably, this is the result of some combination of 8 of their 14 games to date having been against fellow sewer denizens - I think DVOA still has problems with really extreme schedules - and Buffalo having played two teams (New England and the Jets) at times when they were weaker than they have generally been this season (Brady's first game back and injured Jenkins/Sanchez meltdown). In any case, this season's DVOA #25, Carolina, would have been 30th or 31st in each of the past five seasons, and in three DVOA era seasons would have been dead last. The Lions are on course to eclipse the 2005 49ers as the worst team in DVOA history. If the season ended tomorrow, five of the eight worst teams of the last sixteen years, including nos 1 and 3, would have played this season. Two more, incidentally, played in 2008. This is simply staggering. There has to be more to this than random fluctuation, but the reasons for the suckitude seem quite varied - meddling senile ownership leading to perpetual suck (Oakland); teams under construction following years of incompetent management (Detroit, KC, St Louis); seemingly terrible new coaching hires (Tampa); When Good Quarterbacks Go Bad (Carolina); whatever the hell is going on in Tennessee . . .

Any suggestions? Anyone? Is it really just about the cap rising too fast to retain meaning (especially in a depression), or is there something else going on here?

by Otis Taylor89 :: Tue, 10/27/2009 - 7:55pm

Mr Shush,

I have disagree with you on BUF as they have played hard in every game, save CLE and had freak loss to #3 NE on the road where they outplayed them. Yeah, their offense sucks, but D doesn't put them in terrible range.

Does anyone feel the lack of a rookie salary cap is really hurting the weak teams? There have been so few sure fire superstars going in the top 10 of the draft recently, however the bottom teams have to PAY them as superstars when teams outside the top 10 get similar talent, at much lower costs.

by Mr Shush :: Tue, 10/27/2009 - 8:32pm

As I implied above, I don't think NE was #3 or anything like it when they played Buffalo. They've got it together since, but they were not a good team that day, and I don't think it's just because Buffalo played well on defense. I saw most of both the Tampa and Cleveland games, and my subjective opinion was that the Bills sucked hard. You're right that they haven't quit, unlike Tampa and the Titans, for example, and that probably does make them better than a lot of the teams ranked below them. It doesn't mean they aren't a severely crappy team, whose only chance of beating an actual NFL side is a quarterback meltdown.

Why I'm saying this when my team is playing them on Sunday is beyond me. It's like I'm actually begging the gods to strike down Schaub with lightning/swine flu and set the over-under on Rex Grossman's first pick six at forty-seven seconds, followed by further return touchdowns off a Slaton fumble and a Jacoby Jones punt muff while Bad Lung Boy Andre Johnson dies of Consumption on the sidelines. I'm going to bed before I do any more damage.

by Rich Conley (not verified) :: Tue, 10/27/2009 - 10:37pm

"As I implied above, I don't think NE was #3 or anything like it when they played Buffalo. "

Agree. Strongly. Brady was still missing wide open receivers, Galloway was still running the wrong routes, and defenders still didn't know where they were supposed to be.

When week 15 rolls around, bet the over, and give the points. Its gonna be ugly.

by Mike B. In Va :: Wed, 10/28/2009 - 11:02am

Keep in mind it won't be the same Bills team, either, given the number of injuries they've had, especially at LB. I'd bet the over, too (I don't see the world through homer glasses), but give the Bills some credit - they played well in that game, much better than I thought they would. It wasn't just NE being out of synch.

If only they had an offensive coordinator that would run something other than a Pop Warner offense...

by Tundrapaddy (not verified) :: Tue, 10/27/2009 - 8:34pm

Mr Shush -

I have to disagree with you on the idea that 'decuria was eight men'. Everything I've seen indicates it was a unit of ten men. Which would account for the name ('dec' indicates 10 - deciliter, decibel, decathlon, etc.).

You many now continue the topics relevant to football.

by ebongreen (not verified) :: Tue, 10/27/2009 - 9:39pm

I've noticed the same thing, but I have no good explanations. As a Packer fan, we've watched a playoff game (@ the Vikings) followed by two scrimmages (vs. Det, @ Cleve), to be followed by another playoff (vs. the Vikes) this week. It's almost like the Packers had two weeks playing on Bizarro-world against the squads of Wayne State and Case Western Reserve.

Much like the Raiders, one wonders how these mind-warpingly poor clubs actually qualify as NFL franchises. Does it take more than mere incompetence to be that catastrophically bad? How do you do that in a league that is designed to promote parity so well?

My major theory, like yours, is poor ownership/managerial decisions. These aren't necessarily coaching issues, they're franchises that are FUBARed from the top down: bad owners (Oak, Det, KC, StL). Giving Delhomme a $20mil-guarantee extension this summer after last year's playoff disaster? (Car) IMO, the Bucs lack any discernable young talent on defense AND they're missing a quarterback, so what does Raheem Morris have to work with?

As for Tennessee, that's just an aberration. Jeff Fisher has been so good for so long, maybe he gets a bye this year. For everyone else, a combination of management & personnel suck has created a wasteland out of a quarter of the league's teams, and that's a pity.

Captcha: thiamine Brattleboro

by t.d. :: Tue, 10/27/2009 - 9:53pm

Surely Jacksonville, Seattle, and even Chicago are fully capable of playing truly dreadful football in a given week, as well. If they haven't yet joined the bottom feeders, they are likely to, once it becomes apparent the season is a lost cause. Furthermore, it seems like the same thing happened in 2004, when the whole NFC except Philadelphia stunk. I think it just goes in cycles.

by Scott de B (not verified) :: Tue, 10/27/2009 - 10:42pm

Actually, the Roman military unit consisting of eight men was the contubernium, or "tentful," but you are correct they were commanded by a decurion.

by Still Alive (not verified) :: Wed, 10/28/2009 - 12:08am

I think a problem with lots of these types of things is that we are talking about over a thousand years. Look at how much English has changed in the past 1000 years, and that is with mostly superior methods of standardization.

There are other odd wordings like this in Roman history where the say the "80th man" seems an odd name for a man in a group of 300, but there was presumably some point previously where the grouping was smaller in unrecorded history and the phrase for a position survived the change in facts.

I was a "typsetter" at a job where I did no typesetting.

by RickD :: Wed, 10/28/2009 - 12:12pm

Yeah, but they could never handle the zone blitz.

by dryheat :: Wed, 10/28/2009 - 12:32pm

Robo-Decurion could.

by Football Man (not verified) :: Wed, 10/28/2009 - 1:59pm

My theory is that it has something to do with the probable upcoming labor strife. If you were the owner or GM of a bad team, and you anticipated that an upcoming season was likely to be cancelled, perhaps the rational thing to do is not to start your four year plan to rebuild, and not do anything to improve until football is being played again.

by Mansteel (not verified) :: Wed, 10/28/2009 - 3:12pm

Support for your claim that there are exactly 10 godawful teams: There is a group of ten teams (exactly the ten you say) that are a combined 2-45 against the rest the league (discouting games they played against each other). The only wins were Philly's disaster in Oakland and Mark Sanchez's 5-INT implosion vs. the Bills.(who still needed OT to win, BTW). I'm too lazy to look it up, but I'd wager that college 1-AA teams (yeah, FCS, I know) have fared better against I-A teams this year.



by jayinalaska :: Tue, 10/27/2009 - 7:34pm

Man, the Rams are ranked 30th in total DVOA and 5th in variance. They just consistently suck this year.

by Otis Taylor89 :: Tue, 10/27/2009 - 7:44pm

Steve Slaton, the RB with the worst DYAR running the ball, but the best RB DYAR receiving - how is that possible? Probably the 5 fumbles. His running is worse than LJ or LT. Is that a major reason why HOU has the worst DVOA of any winning team?

by Mr Shush :: Tue, 10/27/2009 - 8:24pm

Combination of the fumbles and the fact that both the Texans' starting guards are on IR. Slaton's as dangerous as ever in the open field, but the inside zone play just is not there in the way it was last season. And yes, he's fumbling like crazy, and it is hurting the team.

by chemical burn :: Tue, 10/27/2009 - 7:47pm

I find it amazing that the Eagles' passing game is clearly the weakest part of the team (and DVOA really bears that out), but the Philly faithfully aren't making any noises about benching McNabb. Is the World Series really that much of a distraction? Or will the calls for McNabb's head come after Philly drops both games at home on Sunday?

I thought the offense looked putrid last night, I'm worried that DVOA agrees...

by Todd S. :: Tue, 10/27/2009 - 9:14pm

I think McNabb just one-hopped another wide receiver. If you could somehow combine McNabb's worm-burners with Eli's overthrows...

by chemical burn :: Tue, 10/27/2009 - 7:49pm

Interesting to see the top 3 defenses knotted up at 20.4%-20.5% DVOA. I imagine Philly and GB will continue to get dropped as opponent adjustments rise in strength (obviously, moreso with Philly), but those three teams are amazingly close...

by ChicagoRaider :: Tue, 10/27/2009 - 8:00pm

I am looking for evidence that Nnamdi Asomugha exists. I can't find it.

Unless he is covering tight ends.

by Whatev (not verified) :: Tue, 10/27/2009 - 10:51pm

Didn't he get poked in the eye or something? Have people been making catches on him?

by Mike B. In Va :: Wed, 10/28/2009 - 11:03am

He's been out some, yeah.

by c_f (not verified) :: Wed, 10/28/2009 - 12:08am

The only Raiders football I've seen this year has been "highlights" of Russell mistakes, so here are some guesses:

Have the Raiders been playing more strict Right/Left corner rather than matchups?

Does Asomugha not follow #1 receivers into the slot?

Has Asomugha not been assigned #1 receivers exclusively? (That is, have the Raiders put Johnson/etc. on #1s and rolled coverage?)

by Whatev (not verified) :: Wed, 10/28/2009 - 2:10am

The thing is that I've watched at least part of every Raiders game that I get, and I can only recall seeing him targeted once, a deep left pass (I think it was from Carr to Manningham late in the Giants game) where the receiver caught the ball but went out of bounds.

The camera follows the ball on telecasts, so I assume the fact that I very rarely see him on screen implies that he's not being targeted very much.

by M :: Wed, 10/28/2009 - 1:33pm

I read before the Giants game that the Raiders use him to cover one side of the field instead of a specific receiver. Hence a #1 receiver can have a huge day just by lining up opposite him most of the day.

I believe this tactic was used by past employers of Deion Sanders, but are there any other current CB's used this way?

by tuluse :: Wed, 10/28/2009 - 1:55pm

The Bears generally play corners based on side, rather than follow receivers. It allows the corner to perfect his technique.

by The Hypno-Toad :: Wed, 10/28/2009 - 3:22pm

Champ Bailey was used that way for large portions of the last three or four years. When your number 2 CB is Dominique Foxworth or Dre Bly, it doesn't take an opponent's coordinator long to line up preferred targets on the left. But all that scheming was kind of unnecessary because the Broncos couldn't stop the run or cover rb's, te's or wr's coming across the middle of the field.

by rk (not verified) :: Wed, 10/28/2009 - 10:49pm

I think most teams have left and right corners rather than #1 and #2 corners.

by joenamath :: Tue, 10/27/2009 - 8:03pm

1.5 wins for the rams where does that come from

by t.d. :: Tue, 10/27/2009 - 9:58pm

well they did take Jacksonville to overtime, and they only lost by two to Washington. So chin up, Ramsfan, it could be worse.

/not accusing you of being a Rams fan

by jmaron :: Tue, 10/27/2009 - 8:29pm

Ranking the schedules by taking a simple average of the teams played doesn't really give a real sense of the difficulty of the schedule.

Compare Minnesota's Schedule with Indy

both have played 3 teams with positive DVOA's

Minn - GB (4th - 35.1), Balt (7th - 27.9) and Pitt (12th - 14.6)
Indy - Arz (10th - 22.0), Mia (13th - 12.7), and Jack (16th - 3.5)

they played one mediocre team each

Minn - SF (21st - -6.6)
Indy - Sea (19th - -2.9)

and some stiffs

Minn - Det, StL, Cle
Indy - StL, Tenn

Looking at those schedules I would say that Minnesota's is quite a bit tougher because GB, Balt, Pitt is a lot tougher bunch to play than Arz, Mia, Jack

by Purds :: Wed, 10/28/2009 - 10:38am

I completely agree, and I've always argued this on FO. If one elite team A plays 5 average teams, team A is likely to win 5 times. If elite team B plays 5 opponents that average out to "average," but two of the are the best in the league, and the other 3 are the worst, then it's likely that team B will go 3-2, or maybe 4-1. But, by FO standards, they've played the same schedules. A deceptive stat.

by dbostedo :: Wed, 10/28/2009 - 2:28pm

That's exactly why SOS ratings in college football are an even bigger problem. And why the college metrics on FO use expected wins against a particular schedule game by game for some baseline team.

To apply it back to the example above, team A's and team B's schedules would both be looked at from the point of view of an imaginary baseline team...and how many expected wins that team would get from each schedule. That way you account for game-to-game differences more than an averaging technique would.

The problem there is one of which baseline you use...

by DrewTS (not verified) :: Wed, 10/28/2009 - 3:37pm

Minnesota has played 3 of the "tomato can" teams this year to Indy's 2. That could, and should, lower their SOS. I don't think it's nearly so clear that Minnesota has had a harder schedule.

The overall point is still valid though. Playing Jacksonville 16 times is different than 8 against New Orleans and 8 against Detroit. A good team would prefer the former, while a bad team would prefer the latter.

by Rick A. (not verified) :: Tue, 10/27/2009 - 8:41pm

Keith 1--Of course the Pack does not have a top defense in reality. After 6 games DVOA usually starts becoming reflective of reality--but not this year. Another aberrant thing is this No 1 ranking of the Saints. Miami showed the top teams how to get on top of them and those other teams will finish the job that the hapless Dolphin defense and offense could not...

by t.d. :: Tue, 10/27/2009 - 10:01pm

Miami got like three turnovers early against Brees, and took advantage. Not sure that's a replicable gameplan.

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 10:14am

True, but look at the Dolphins possessing the ball for well over 40 minutes against Indy. I know Miami lost, but that's the way to beat prolific passers. New Orleans looked very vulnerable last week.

by Still Alive (not verified) :: Wed, 10/28/2009 - 2:27pm

Actually I am not sure that helped them at all, and i think it had more to do with IND scoring quickly than MIA holding the ball particularly well.

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 3:21pm


I guess we'll sort of find out when (not if) New Orleans loses. Unless it looks particularly flukey, we should be able to see how opponents should attack the Saints and whether it bears resemblance to Miami's gameplan against them and Indianapolis.

by MilkmanDanimal :: Tue, 10/27/2009 - 9:25pm

A dozen years watching Tampa consistently have an excellent defense, and the year after Kiffin leaves they're last in defensive DVOA. Last! There's "rebuilding" and there's "train filled with syphlitic puppies wrecks into nuclear waste site". Gaaaah. And the third-toughest remaining schedule. They either beat Seattle or they're winless.

Football just isn't as fun this year . . .

by Rick A. (not verified) :: Tue, 10/27/2009 - 10:37pm

t.d.--of course it's a replicable game plan. The three turnovers were not flukes, Miami pressure on the N.O. O line and on their, not the NFL's greatest, receiver corp., created them. You'll see more of it start happening now. In fact, Atlanta will play them very well on Monday. An outright would not surprise. Once a weakness is made visible in the NFL it's open season...

by Rick A. (not verified) :: Tue, 10/27/2009 - 10:39pm

P.S. Shockey is a break even proposition at best--i.e. his pluses versus his minuses. Speaking of on the field, not off.

by t.d. :: Tue, 10/27/2009 - 11:29pm

Agreed about Shockey, but I think if the formula is, 'make Drew Brees throw picks' I'm just saying that it's odd that they'd have gone away from that after the first quarter and a half.

by B-Rick (not verified) :: Tue, 10/27/2009 - 11:07pm

A question on DVOA.

I understand that fumble recoveries are considered random events and that a team with a higher rate of recoveries is considered to be on a lucky streak. Generally I agree with this logic, except in one important category: fumbles by a QB when he is sacked. These plays are the result of great defensive plays and seem to usually be recovered by the defense.

I saw that Denver is a variable when considering special teams dvoa, due to the altitude. I know Denver's special teams have varied between below average to horrific over the years, but could Denver being a variable artrifically lower their DVOA?

by Arkaein :: Tue, 10/27/2009 - 11:42pm

Actually you're wrong, I believe sack-fumbles are about an even chance to be recovered by either team.

I don't have the link handy, but a few years ago I remember FO grading fumble recovery by the offense something like this:

aborted snap = 60%
sack = 50%
fumbles just past LOS = 40%
fumbles farther downfield = 30%

These numbers are from memory and may not be exact, but you get the idea. I believe the thinking is that the farther downfield the fumble occurs the fewer offensive players are close enough to attempt a recovery. On a sack fumble there's the QB, the lineman who the defender just got past, maybe a RB in protection, and any other linemen who happen to be near the ball. Downfield it's often a single WR surrounded by 3 or more DBs and LBs.

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 10:16am

"and seem to usually be recovered by the defense."

Unless Ray Edwards is being mr. cutey-pants-twinkle-toes.

by Duke :: Wed, 10/28/2009 - 12:39am

This week, Detroit had its worst offensive DVOA of the year, by far...

Well, BYE has always had a stout defense.

by Rick A. (not verified) :: Wed, 10/28/2009 - 1:19am

t.d.--what happened in the game was that the press coverage and blitzing by the Fish was working, Brees could not get the ball to Colston or Henderson and hurried into some mistakes. But, in the third quarter Will Allen went out with a bad injury and all the Fins have behind him are two rookies. Furthermore, the Saints were adjusting and called upon Mr Shockey. As the world saw, he had one of his good games and between that and that injury to the Dolphin cornerback, things began opening up on the outside for the Saints. Then, besides that, what I saw was that the Saints are in better condition than Miami. It's usually the road team that wilts in the heat of Miami, but Sunday it was the Dolphin pass rushers that wilted. Joey Porter appears to not be in top football condition, certainly not up to where his mouth alleges he is. Once all that was happening, forget it, the Fish have nothing in the league of the Saint barage. But they showed the other teams how to play it. If Shockey would have been stopped, if the injury didn't happen, if the reality of the Dolphin pass rushers were up to their mouths, they would not only have beaten New Orleans but embarrassed them.

by Whatev (not verified) :: Wed, 10/28/2009 - 2:01am

Arizona? Fourth in defense??

by Joe in Seattle (not verified) :: Wed, 10/28/2009 - 1:07pm

Bill Davis has done a great job with that defense, I think Clancy Pendergast was too smart for his own good sometimes.

The Cardinals run defense is very good and their secondary has too much talent to be as bad against the pass as they were to start the season. Alan Branch has been playing very well this season and their defensive line rotation is pretty solid.

by Spielman :: Thu, 10/29/2009 - 7:00pm

They've been playing the run first all year, making their run stats look better. I think that's smart, not necessarily because their secondary is so brilliant, but because they've got Rolle and DRC who are excellent ballhawks, so their chances of getting turnovers against an air attack are better than most, and because Adrian Wilson is at his most valuable playing up, so you might as well take advantage of that. They're doing okay on getting INTs so far this year, and I expect their fortunes there to actually improve over the remainder of the season, while some of the teams up near the top in that category regress a bit.

by huston720 :: Wed, 10/28/2009 - 8:29am

It's interesting to me that the bottom 6 teams all have past schedules in the top 8 for difficulty (with KC the 7th worst, with the 10th hardest schedule). Is it possible there is some causation here, and if so which way does it go. Is it a case where these bottom teams are so bad that they have really boosted the play of the teams they have already played, even with opponent adjustments. Is it just a random coincidence, since these teams haven't played each other much yet. Or is there a chance some of these teams will show some improvement as their schedules get better, with the exception of poor TB whose schedule is actually projected to get harder.

I'm especially hopefull as a Browns fan since their schedule has been brutal. Only one game against a bad team (a win against Buffalo) with the rest against some of the top teams in the league. (vs Minn, Cinci, GB, @ Balt, @ Pitt, @ Den, all of whom rank in the top 15 in DVOA) So maybe there is some hope for the Browns, assuming Derek Anderson can remeber how to at least occaisionally throw an accurate pass.

by Rich Conley (not verified) :: Wed, 10/28/2009 - 11:58am

"Or is there a chance some of these teams will show some improvement as their schedules get better, "

OF course thats the case. DVOA does pretty good adjustments, but it doesn't work well when schedules are really far from the mean.

Playing a string of good teams always drives a team's DVOA down, and playing a string of bad teams always drives it down. It makes its adjustments assuming linear performance difference, when its most likely a complicated curve.

Still, its better than pretty much anything else at this point.

by Still Alive (not verified) :: Wed, 10/28/2009 - 2:25pm

Also if you are a horrible team it is impossible to play yourself. So by that very fact alone you will have a more difficult schedule than your opponents on average. The differences between schedules are not so great that it overwhelms the trend of the best team in a division having the easiest schedule and the worst the hardest.

by huston720 :: Wed, 10/28/2009 - 4:59pm

Good points, but it also confirms my suspicion that perhaps we don't have as many horrible teams as thought since the bottom dwellers have seemed to play the hard parts of their schedule early. My point also was more about overall strength of schedule, not compared to their opponents. I went back and looked at the final strength of schedule for the bottom teams in years past on the team efficiencies pages and it looked like the varied quite a bit.

Basically my point is that while it appears that there are 7 or 8 really bad teams in the league, which would be a high number compared to past years, pehaps this is a bit of an illusion, and some of these teams will move up from terrible to just bad and below average teams by the end of the year. It looks best for StL and Cle who both have had rough schedules so far, but have much easier schedules going forward. On the flip side as I mentioned it looks bad for TB, and also Car who goes from an average schedule so far to the toughest schedule from here on out. Of course the caveat being that the past and future schedule ranks will adjust with more info.

by Mathias (not verified) :: Wed, 10/28/2009 - 8:45am

A question to jets fans: How good is Darrele Revis? From what I've seen so far this season, he looks like the new Nnamdi Asomugha. Am I right?

by MJK :: Wed, 10/28/2009 - 11:38am

I've only see him play versus the Patriots this year, but he looked insanely good in that game. He more or less shut down whichever of Randy Moss or Wes Welker he was covering--with no safety help.

Ryan came out and said that he thought Revis was the best CB in the NFL. Now, some of that is Ryan talking up his own player, but it at least implies that he's the best CB on the Jets.

by Mediawatch (not verified) :: Wed, 10/28/2009 - 11:54am

Wes Welker didn't play in that game.

by MJK :: Wed, 10/28/2009 - 12:08pm

Well, that explains why Welker was shut down that game. :-)

But Revis is why Moss more or less was.

by RickD :: Wed, 10/28/2009 - 12:21pm

Moss was also playing with a sore back that week. He wasn't 100%.

I will say that Revis is very good, but I wish we had more to go on than the word of Rex Ryan and the NY Media.

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 3:27pm

Revis also shut down Andre Johnson, so regardless of Moss's health, he's no scrub.

I didn't watch the Saints game, but Colston was held to two catches for 33 yards. The leading receivers for the team (4 receptions each) were Pierre Thomas (46 yards) and Jeremy Shockey (34 yards). Brees had 190 yards on the day and no TDs. I'm not sure how much Revis contributed, but if the Saints' passing game wasn't clicking, something good is happening.

captcha: papoose FERON

by RickD :: Wed, 10/28/2009 - 12:18pm

He also looked excellent in the Raiders game.

Watching Jamarcus Russell try to thread the needle and beat Darelle Revis was high comedy!

by apk3000 :: Wed, 10/28/2009 - 10:04am

REDSKINS are clearly ranked TOO HIGH because DAN SNYDER IS A IDIOT. Ranking teams based on STADIUM FAN SIGNAGE POLICY is way better than this. FIRE VINNY!!!11

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 10:23am


On a more serious note, I often find myself puzzeled at the results of statistical analysis, but find ways that the numbers can make more sense upon further reflection. I've become somewhat adept at this process (work experience including statistician and data analyst) and have, unfortunately, been known to jump to incorrect conclusions that I would not have entertained but for the surprising analysis.

I'm wondering if the staff ran the formula, saw Green Bay with the #1 ranked defense and thought, "What? That can't be right. Did I fail to carry the 3?" and then, with deep introspection "I suppose the failed defensive play against opponent x could be argued away due to these somewhat extenuatiing circumstances, and..." on and on.

Is that possible?

by Chip :: Wed, 10/28/2009 - 2:17pm

GB's defense is solid, but they heavily benefited from an early soft schedule. They racked up INTs against poor competition - INTs that score highly in DVOA. However, interceptions themselves while descriptive in nature, are not predictive. Brian Burke has two great posts on how "QB's throw interceptions" more than "DBs catch them" (here and here ).

The in-season (1st half to 2nd half) correlation of defensive interceptions is a whooping 8%.

Take a look at GB's weekly Defense DVOA
CHI -34%
CIN -4%
STL -3%
MIN 24%
DET -73%
CLE -50%

Opposing QBs in the CHI, DET and CLE games had complete meltdowns throwing 4, 3, and 1** respectively interceptions each. This is 8 of 11 defensive interceptions on the season, which is 4th in the league. Taking a simple DVOA average, which includes 3 opposing QB meltdowns, and presto…. you have the #1 Defensive DVOA.

** CLE also had 3 fumbles. Burke didn't create a Def. Fumble rate, but given that Off. Fumble rate correlates at ~50% rate in-season (1st half to 2nd half), it's a somewhat persistant skill that the opposing offense takes with it from game to game. Thus it's seems that GB benefited from the schedule here too, driving the -50% game DVOA.

by M :: Wed, 10/28/2009 - 2:56pm

Interesting stat - it's really a correlation coefficient of .08 for 1st half/2nd half interceptions?

Also, you wouldn't have access to what the correlation on offense would be, would you? I suspect (for obvious reasons) that it is a LOT higher.

Another hypothesis - I'm guessing the coefficient would be even lower or negative if you could control for the effects of intra-divisional games where defenses usually play the same QB twice. That of course relies on the belief that interceptions are somewhat dependent on the QB but mostly random for defenses.

If the offensive coefficient is high (say .55), this could support the idea that choosing fantasy defenses should be done on a matchup basis and not riding one defense for the whole year - unless you happen to luck into the 2000 Ravens or 2002 Bucs.

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 3:31pm

"could support the idea that choosing fantasy defenses should be done on a matchup basis and not riding one defense for the whole year"

I've been riding Green Bay all year, interestingly enough.

Although I'll admit the reason I took them was because they (along with the Cardinals) had a complimentary schedule against crappy offenses. So there you go.

by M :: Wed, 10/28/2009 - 4:42pm

I'm thinking of loading up on 4-5 fantasy defenses on my roster after the bye week, and each week taking the most attractive matchup. I would be easier if I could identify 2 RB's from my current pile of shit to start every week:

Tomlinson (6 times stuffed from 1-2 yard line; yeah, I drank too much of the LDT kool-aid this offseason).
Jacobs (should be #1, but obviously isn't right)
Bradshaw (who's better - Bradshaw or Jacobs?)
Ray Rice (As far as I can tell, the only keeper of the bunch)
J. Stewart (Carolina should run 85% of the time; if they did, they might have a winning record given their early schedule)

Other possibilities - Beanie Wells, Shonn Greene (NYJ), Correl Buckhalter (sigh)

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 3:33pm

Great analysis.

And I agree: I think Green Bay has an excellent defense for fantasy football, I just doubt that this translates to solid defense in REAL football.

People are high on the Packers ascension after beating on a couple of scrub teams, but this week should show whether there's a race for the NFC North or the Vikings own the division.

by Arkaein :: Wed, 10/28/2009 - 4:03pm

I'm not sure your argument about interceptions follows the actual events on the field.

GB intercepted Palmer twice, which you've left out of your stats. So 2.67 INTs per game against CHI, DET and CLE, and 2.0 INTs per game against CIN. Considering that Cincy looks quite strong and CHI hardly counts as a scrub team (500 record, with a win against Pittsburgh), I'm not sure that the interception distribution across good versus bad opposition is much different than what any team would expect. Great defenses like Baltimore and Pittsburgh tend to rack up their best stats against inferior opposition as well.

Also, the GB has been pretty consistent at making picks since at least last season, and Charles Woodson (4 of GB's picks) has been very good at it his entire time in GB. He probably won't keep up his rate of 0.67 INTs per game over the whole season, but it's not impossible given that he had 8 INTs just a couple of years ago. GB's CBs also tend to rate well in passes defended, which I believe may be a better predictor of future interceptions than interceptions themselves, so a big dropoff certainly seems possible but not really likely.

by bravehoptoad :: Wed, 10/28/2009 - 2:48pm

Isn't that kind of the point of these statistics: to give clues about what's been important? To aid insight?

by peterplaysbass (not verified) :: Wed, 10/28/2009 - 3:36pm

Yes. I don't expected the stats to tell the whole story, but I think that sometimes the stats are confusing and people conform their thinking to fit the numbers. If the formula's right and it adds NEW insight, then that's terrific. If the formula's flawed and sends analysts down the wrong rabbit hole, then that's tragic.

Either way, it's an interesting exercise. I love this site, and its comment threads.

captcha: $250-million froth ...excuse me while I go buy a powerball ticket.

by Yuri (not verified) :: Wed, 10/28/2009 - 10:46am

How come that (on the offense DVOA page), the PHI total DVOA and VOA are lower than either of the component DVOA or VOA (pass or rush)? Is this an error of some sort?

by Pat (filler) (not verified) :: Wed, 10/28/2009 - 4:23pm

High number of penalty plays (the predictive ones, not the random ones).

Please note that the combination of pass offense and run offense will appear to be higher than total offensive DVOA. This is because offensive DVOA also accounts for false starts and delay of game penalties, which are all negative plays but are not considered as either passes or runs.

Right on the team offense page.

by The PAC (not verified) :: Wed, 10/28/2009 - 3:20pm

I'm just glad to see that the Seahawks had another dominating shutout agains BYE last week. I hope they can keep that up.


by jmaron :: Wed, 10/28/2009 - 4:32pm

GB has given up 25 sacks this year in 6 games worst in the league (11.7% sack rate). But what is really interesting is that they’ve faced only one team in the top 18 in sack rate (Minnesota 4th)

GB Sacks
Total vs GB pct ag GB Ranking
Chic 14 4 0.285714 24
Cinn 17 6 0.352941 19
Stl 12 2 0.166667 27
Minn 24 8 0.333333 4
Det 15 5 0.333333 18
Clev 12 0 0 28
94 25 0.265957

GB has 6 games left with teams with a decent pass rush (Dall, Balt, Arz, Sea, Ptt, Minn). They gave up zero in their last game so maybe they've fixed the problem but I suspect they will really struggle in the 6 games I've listed).

by jmaron :: Wed, 10/28/2009 - 4:34pm

sorry for the messy table - does anyone know how to keep the excel table format in place

by Arkaein :: Wed, 10/28/2009 - 5:59pm

The pass protection has definitely been GB's greatest weakness, and it certainly could hurt them later on. However there are some positive indicators for the remainer of the season:

- Barbre was miserable in his first start against CHI. He's settled down since then, at least to the point where no single DE has manhandled him.

- Antwaan Odom and Jared Allen both took advantage of LG Daryn Colledge playing out of position at LT. GB should have Clifton back soon, and will probably be start the rookie Lang at LT in the short term, where he played well against Cleveland and prevents the need for interior line shuffling.

- Mark Tauscher should be ready to play at RT soon if Barbre does break down in the future. He may not have much left to provide in the running game, but has always had great pass blocking technique and should provide consistency.

A lot hinges on health, which has been a problem so far (especially if you count Tauscher as a starter who has missed all 6 games so far this season). If Clifton and Tauscher can both recover from injury and play at a reasonable level I think GB can rise to at least average pass blocking ability.

by jmaron :: Wed, 10/28/2009 - 11:11pm

We've already debated the how much is it Rodger's fault so I won't bother rehashing that argument. Minnesota's safeties have been their weakness - they are giving up a ton of long plays on passes over the middle. It seems both GB and Minnesota are very balanced teams with one major weakness.