Writers of Pro Football Prospectus 2008

Most Recent FO Features

FitzgeraldLar04.jpg

» Scramble for the Ball: The DVOA Schism

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

12 Dec 2007

Week 14 DVOA Ratings

by Aaron Schatz

Well, first of all, we sincerely apologize for all the problems over the last two days. The server situation got so bad yesterday that we had to start our server migration a day early, which meant no DVOA ratings article because we didn't want it to suddenly disappear when the migration started. We still have some things that need fixing, but we'll be catching up with all our columns over the next couple days (Every Play Counts is going to run Thursday afternoon) and this should mean a major improvement to our traffic issues, with an even better, more permanent solution planned for the off-season. Let's hit the numbers then a couple comments...

* * * * *

Here are the Football Outsiders team efficiency ratings through 14 weeks of 2007, 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 for opponent and 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.

To save people some time, we request that you 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 NE 60.5% 1 57.8% 1 13-0 48.2% 1 -7.9% 7 4.4% 7
2 DAL 38.0% 2 39.4% 2 12-1 30.3% 2 -7.0% 8 0.8% 13
3 IND 36.0% 3 30.6% 3 11-2 29.2% 3 -13.1% 3 -6.2% 31
4 GB 27.3% 5 29.6% 4 11-2 21.8% 4 -1.9% 15 3.6% 10
5 PIT 24.1% 4 16.6% 9 9-4 4.8% 13 -21.3% 1 -2.1% 20
6 JAC 20.4% 8 23.4% 5 9-4 16.5% 5 -0.2% 16 3.7% 8
7 TB 17.5% 6 16.4% 10 8-5 12.2% 8 -7.9% 6 -2.6% 22
8 SD 16.4% 7 18.7% 6 8-5 4.2% 14 -6.3% 9 5.9% 3
9 SEA 16.2% 10 17.6% 8 9-4 3.7% 15 -10.0% 5 2.5% 12
10 MIN 14.7% 9 17.9% 7 7-6 6.3% 12 -4.8% 10 3.6% 9
11 CLE 7.3% 11 12.5% 11 8-5 12.5% 7 12.3% 27 7.1% 2
12 CIN 5.9% 12 7.3% 12 5-8 13.4% 6 8.2% 22 0.7% 14
13 PHI 3.4% 13 0.7% 16 5-8 9.4% 10 2.3% 17 -3.7% 28
14 DEN 2.4% 20 2.4% 14 6-7 11.5% 9 5.9% 21 -3.2% 24
15 WAS 0.4% 14 -1.3% 18 6-7 -4.0% 21 -3.7% 12 0.7% 15
16 NYG 0.1% 15 0.3% 17 9-4 -1.8% 19 -3.4% 13 -1.5% 17
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 BUF 0.0% 18 4.4% 13 7-6 -2.6% 20 2.5% 18 5.1% 6
18 TEN -0.3% 16 -4.1% 19 7-6 -9.8% 23 -13.1% 2 -3.7% 27
19 NO -6.4% 22 0.9% 15 6-7 9.0% 11 12.4% 28 -2.9% 23
20 BAL -8.0% 17 -9.9% 21 4-9 -16.5% 27 -10.2% 4 -1.7% 18
21 CHI -9.1% 21 -5.8% 20 5-8 -19.2% 28 -2.0% 14 8.1% 1
22 ARI -9.4% 19 -11.0% 22 6-7 -1.8% 18 4.2% 20 -3.5% 25
23 HOU -9.5% 24 -11.7% 23 6-7 0.0% 17 12.0% 26 2.5% 11
24 DET -13.1% 26 -13.8% 24 6-7 1.6% 16 8.9% 23 -5.8% 30
25 NYJ -19.2% 25 -13.9% 25 3-10 -8.8% 22 15.5% 31 5.1% 5
26 KC -19.7% 23 -20.5% 26 4-9 -22.6% 31 -4.7% 11 -1.8% 19
27 ATL -26.4% 28 -26.9% 28 3-10 -13.9% 26 12.6% 29 0.1% 16
28 OAK -27.1% 27 -24.3% 27 4-9 -13.0% 25 10.5% 25 -3.6% 26
29 CAR -29.7% 30 -32.0% 29 5-8 -20.3% 29 3.0% 19 -6.4% 32
30 MIA -31.7% 29 -36.0% 31 0-13 -12.6% 24 16.7% 32 -2.5% 21
31 STL -34.5% 31 -33.6% 30 3-10 -20.5% 30 9.6% 24 -4.4% 29
32 SF -44.8% 32 -45.4% 32 3-10 -35.8% 32 14.9% 30 5.9% 4

  • 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.
  • 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 least consistent (#1, highest variance) to most consistent (#32, smallest variance).


TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR. RANK
1 NE 60.5% 13-0 58.7% 12.3 1 5.2% 3 -25.4% 27 12.0% 18
2 DAL 38.0% 12-1 39.6% 10.3 3 -0.1% 17 -13.0% 22 10.3% 22
3 IND 36.0% 11-2 35.8% 10.8 2 2.7% 12 -18.4% 24 17.6% 8
4 GB 27.3% 11-2 30.9% 9.5 4 -0.6% 19 -28.3% 28 10.0% 23
5 PIT 24.1% 9-4 23.7% 8.0 9 -0.5% 18 -11.0% 21 26.8% 2
6 JAC 20.4% 9-4 19.9% 9.3 5 1.2% 14 -6.2% 19 5.8% 32
7 TB 17.5% 8-5 19.7% 8.4 6 -4.5% 25 -50.5% 32 14.8% 13
8 SD 16.4% 8-5 10.3% 7.6 11 4.9% 6 -18.9% 25 22.1% 5
9 SEA 16.2% 9-4 24.2% 8.1 8 -9.6% 32 -32.0% 29 10.8% 20
10 MIN 14.7% 7-6 18.7% 8.2 7 -2.9% 23 -3.2% 18 20.4% 7
11 CLE 7.3% 8-5 7.5% 7.4 12 -1.2% 20 -19.5% 26 6.8% 29
12 CIN 5.9% 5-8 9.9% 7.7 10 2.4% 13 -34.6% 31 7.6% 28
13 PHI 3.4% 5-8 -3.2% 7.2 13 6.0% 2 15.8% 10 12.6% 17
14 DEN 2.4% 6-7 5.0% 6.5 16 0.6% 16 10.8% 14 37.7% 1
15 WAS 0.4% 6-7 -8.5% 6.4 18 4.8% 7 26.4% 6 9.2% 25
16 NYG 0.1% 9-4 1.3% 6.7 15 -1.4% 21 30.5% 3 6.6% 30
TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR. RANK
17 BUF 0.0% 7-6 -6.9% 6.9 14 7.3% 1 5.4% 15 12.6% 16
18 TEN -0.3% 7-6 -3.1% 6.4 17 0.8% 15 -1.4% 17 16.5% 9
19 NO -6.4% 6-7 -3.6% 6.1 19 -6.7% 31 -7.6% 20 13.9% 14
20 BAL -8.0% 4-9 -15.8% 5.5 22 4.0% 10 4.3% 16 15.3% 11
21 CHI -9.1% 5-8 -12.4% 5.6 21 3.3% 11 17.8% 9 7.7% 27
22 ARI -9.4% 6-7 -2.9% 5.1 24 -6.2% 30 -33.6% 30 26.7% 3
23 HOU -9.5% 6-7 -9.7% 5.4 23 -3.1% 24 29.4% 4 11.9% 19
24 DET -13.1% 6-7 -11.0% 5.9 20 4.6% 8 12.0% 13 21.0% 6
25 NYJ -19.2% 3-10 -21.2% 4.4 25 4.9% 5 20.3% 8 10.7% 21
26 KC -19.7% 4-9 -24.3% 4.3 26 4.9% 4 -16.3% 23 26.0% 4
27 ATL -26.4% 3-10 -20.3% 4.1 27 -5.2% 28 12.2% 12 6.1% 31
28 OAK -27.1% 4-9 -23.1% 3.3 29 -2.3% 22 36.4% 1 14.8% 12
29 CAR -29.7% 5-8 -24.4% 3.8 28 -4.5% 26 35.9% 2 15.6% 10
30 MIA -31.7% 0-13 -37.8% 3.0 30 4.2% 9 29.2% 5 13.8% 15
31 STL -34.5% 3-10 -29.7% 2.8 31 -4.9% 27 21.0% 7 9.8% 24
32 SF -44.8% 3-10 -37.9% 2.4 32 -6.2% 29 15.4% 11 8.3% 26

Best and Worst DVOA Ever Watch

BEST TOTAL DVOA
AFTER WEEK 14
  BEST OFFENSIVE DVOA
AFTER WEEK 14
  WORST TOTAL DVOA
AFTER WEEK 14
  WORST OFFENSIVE DVOA
AFTER WEEK 14
2007 NE 60.5%   2007 NE 48.2%   2005 SF -63.8%   2005 SF -45.4%
1999 STL 48.5%   2004 IND 45.1%   1999 CLE -52.6%   1997 NO -40.3%
2001 STL 44.5%   2002 KC 37.9%   2000 CLE -46.4%   2002 HOU -39.6%
2004 PIT 42.3%   1998 DEN 33.1%   2003 ARI -45.1%   2007 SF -35.8%
2004 NE 41.7%   2005 IND 32.7%   2000 CIN -45.0%   2004 CHI -35.0%
2005 IND 41.0%   2006 IND 31.8%   2007 SF -44.8%   2000 CLE -34.5%
1998 DEN 40.8%   2000 IND 31.3%   2004 SF -44.7%   2006 OAK -33.6%
2002 TB 40.8%   2007 DAL 30.3%   2000 ARI -44.4%   2004 MIA -32.8%
2004 IND 39.1%   2004 KC 29.8%   1998 PHI -43.4%   1998 PHI -32.6%
2007 DAL 38.0%   2007 IND 29.2%   2002 ARI -43.4%   2001 CAR -30.1%

How good are the Patriots? They beat one of the top five teams in the league by three touchdowns and their DVOA rating actually went down by one percentage point. OK, it isn't quite like that... The Patriots had a rating of 69.4% for this game; their overall dropped slightly because of changes in opponent adjustments. Pittsburgh, meanwhile, had a rating of 18.7% and barely saw their league-best defense drop despite giving up 34 points.

Meanwhile, while the undefeated Patriots are keeping up their high level of play, the winless Dolphins have completely dropped off a cliff. Three weeks ago, I wrote an article at ESPN.com saying that the Dolphins would not finish 0-16. They simply were not playing like the worst team in football. Well, in the last two weeks, that has changed. In two weeks, Miami's DVOA rating has gone from -16.2% to -31.7%. To show just how horrible the Dolphins have been in the last two games, I put together a DVOA week-to-week graph:

For most of the season, Miami was just a garden-variety bad team with poor luck. Their last two games have been two of the three worst games of the season according to DVOA. The only game with a lower rating was also played in Week 14. If you look at which teams moved up and down significantly in the ratings, you may be able to figure out which game it was. Since very few people actually watched the game, it's probably a bit of a surprise.

Worst Single-Game DVOA 2007
(through Week 14)

Team Opp. Week DVOA Score
KC DEN 14 -143.4% 41-7
MIA BUF 14 -104.3% 38-17
MIA NYJ 13 -101.7% 40-13
STL TB 3 -98.3% 24-3
SF CAR 13 -92.5% 31-14
DEN SD 5 -87.4% 41-3
CAR NO 12 -85.5% 31-6
SD MIN 9 -84.8% 35-17
ARI CAR 6 -81.5% 25-10
OAK GB 14 -81.4% 38-7

Kansas City managed a single touchdown against a below-average defense. They went 1-for-13 on third-down conversions and gained 2.5 yards per offensive play, including 16 yards rushing on 17 carries. They fumbled three times and gave up five sacks, while Denver had just one of each.

On to the housekeeping... all the team stats pages and individual stats pages are now updated, as are the playoff odds. New England now is listed with a 62.2 percent chance of finishing 16-0, while the Dolphins are still listed with less than a 25 percent chance of finishing 0-16. The premium database and Loser League will be updated later today; things are delayed a bit by the server nonsense. LOSER LEAGUE AND PREMIUM NOW UPDATED.

For (short) comments on every team, remember to look for DVOA on AOL, every Wednesday. (This will be linked on the FO Goes Mainstream page.)

Posted by: Aaron Schatz on 12 Dec 2007

190 comments, Last at 19 Dec 2007, 6:40pm by BDC

Comments

1
by Ch V Kalyan (not verified) :: Wed, 12/12/2007 - 2:55pm

blah blah on the main page... huh!

first!!

2
by Yosi Scharf (not verified) :: Wed, 12/12/2007 - 2:56pm

How does NE have the best rush offense in the league? (20.2 DVOA)

3
by Frick (not verified) :: Wed, 12/12/2007 - 2:56pm

And the crowds rejoice!

4
by Trogdor (not verified) :: Wed, 12/12/2007 - 2:57pm

No need to apologize for the server issues. They happen. As long as it's a sign of the site growing (these servers used to be good enough but they can't handle the increased traffic) and not mis-management, I think we can all suck it up and deal.

Any thoughts as to why Miami's fallen off the cliff? The change in QB is probably part of it, but it can't explain everything. And it's notable that even with that dive, they're still slightly ahead of the Rams and significantly ahead of the 49ers. Which begs the question, how is San Francisco that freaking bad (and how did they win *three* games)?

5
by iowapatsfan (not verified) :: Wed, 12/12/2007 - 2:57pm

I love how Miami has 3 estimated wins and 0 real ones, while has SF has 3 real wins and only 2 estimated... The Pats #2 pick looks real good right now.

6
by B (not verified) :: Wed, 12/12/2007 - 2:58pm

2: They have a really good O-line plus teams are so afraid of their pass offense.

7
by apocalypse66 (not verified) :: Wed, 12/12/2007 - 2:58pm

First!

Yes, the Eagles' DVOA remains above-average despite another loss. Keep those long, consistent, red zone stalling drives alive.

Seriously, how do people think the Eagles' schedule has affected them this season?

8
by GlennW (not verified) :: Wed, 12/12/2007 - 3:00pm

"Printable version" is linking back to main page... thanks.

9
by Athelas (not verified) :: Wed, 12/12/2007 - 3:02pm

I'm stunned at how relatively good Miami's offense is--I would have thought their defense was better than their offense (but I admit I don't spend a lot of time watching them).

10
by chip (not verified) :: Wed, 12/12/2007 - 3:11pm

To lend further credence to the "there is no 'Grossman is our starter', there is no 'Greise is our starter', only a CHICAGO quarterback" theory, Grossman/Griese are ranked 33rd/34th in both DPAR and DVOA. The trifecta would be Orton listed at 35th by week 17. Interestingly, Adrian Petersen & Benson are ranked 47th/49th in DPAR and 50th/46th in DVOA. Maybe there's also the "CHICAGO RB" theory as well.

11
by David (not verified) :: Wed, 12/12/2007 - 3:19pm

So umm, wow. I was at the Broncos - Chiefs game, and while the Chiefs looked painfully inept, I don't think I would have pegged that as the far and away worst game played by anyone this season. How was the Broncos DVOA for this game?

12
by footballprofessor (not verified) :: Wed, 12/12/2007 - 3:21pm

Ooooh, look at the new security word images, super cool.

Anyway, let's talk about regression to the mean...if Miami just had their two worst DVOA games of the year, what do we think will happen in the ensuing weeks? They've got Baltimore this week and will lose by about two touchdowns. Then they'll play New England closer than everyone thinks they will and will have a shot at winning. If they lose to New England they'll beat Cincinnati in week 17...if they beat New England, they'll lose in week 17. No matter what, this team finishes 1-15 and New England finishes 15-1.

13
by Richie (not verified) :: Wed, 12/12/2007 - 3:24pm

#4 - I'm guessing that "starting at RB, Samkon Gado" is one factor.

14
by Ian (not verified) :: Wed, 12/12/2007 - 3:27pm

I guess I don't understand someting about how the system works. The cowboys DVOA is 2 pts better than the Colts, but their weighted DVOA is 9 points better, despite playing a weaker schedule. I thought weighted meant weighted by schedule?

15
by patriotsgirl (not verified) :: Wed, 12/12/2007 - 3:27pm

4: navin would be able to answer you more completely, but my take is that the Niners are completely offensively inept. Terrible O-line, terrible coaching, bad quarterbacking - they are such a complete disaster to watch that it's almost comical. The offense looks completely unprepared, and the defense managed to make Atlanta's passing game look threatening.

In terms of the Niners having 3 wins, keep in mind that they're in the division of suck (all of their wins are divisional wins) and completely have the Cardinals' number for some reason.

Oh, and Andy Lee is really, really good - and the rest of their special teams aren't bad. If Lee (and Moorman) don't go to the Pro Bowl, there's something seriously wrong.

16
by zenbitz (not verified) :: Wed, 12/12/2007 - 3:28pm

4. Obviously, it's the niners brilliant coaching staff that inspired them to win 2 extra games than expected.

The scuttlebutt on the niners boards is that it's the offensive total lack of ball control/time of posession that "tires out" the defense and makes it look worse than it actually is.

I kinda think DVOA accounts for this, at least partially... so I think they just have little talent and bad execution

17
by Jacob Stevens (not verified) :: Wed, 12/12/2007 - 3:29pm

No matter what, huh?

18
by Sid (not verified) :: Wed, 12/12/2007 - 3:29pm

Wooohooo! I guessed the worst game of the season correctly!

19
by deflated (not verified) :: Wed, 12/12/2007 - 3:30pm

In amongst the rest of the 'best/worst ever' action Denver is putting in a very strong bid to be the most inconsistent team to run onto a field. That 37.7% variance currently leads the field with only the 2005 Niners offering any challenge. Wouldn't have picked a Shanahan team to do that.

20
by Jacob Stevens (not verified) :: Wed, 12/12/2007 - 3:31pm

14 -- weighted measures how good they're playing "right now," weighing only the last 8 games -- or is it 10? -- with the most recent games being weighed slightly more.

21
by Scott de B. (not verified) :: Wed, 12/12/2007 - 3:33pm

I guess I don’t understand someting about how the system works. The cowboys DVOA is 2 pts better than the Colts, but their weighted DVOA is 9 points better, despite playing a weaker schedule. I thought weighted meant weighted by schedule?

No, it's weighted by recent performance.

22
by Sid (not verified) :: Wed, 12/12/2007 - 3:33pm

RE: 4

Any thoughts as to why Miami’s fallen off the cliff? The change in QB is probably part of it, but it can’t explain everything.

Cleo Lemon started most of the game against Buffalo and he's starting in Week 15. The QB situation is a very small part of it. The whole team is in disarray. Their best offensive player was probably Jesse Chatman, and he didn't play against Buffalo. Same for Marty Booker.
Their starting QB was Beck, their RB was waiver wire fodder Gado, and their best WR was Ginn, I guess. Combine that with the same bad offensive line and you have a historically bad offense. Don't let me get started on the defense and special teams.

23
by justin (not verified) :: Wed, 12/12/2007 - 3:34pm

any idea whats happened to Seattle's special teams, they were pretty good now they suck.

josh brown used to be automatic, now i think hes missed 5 of his last 9 or so

ryan plackermier used to boot it now he has like 3rd lowest average, and 2nd lowest net.

kick/punt return is pretty good

can we blame it all on our extremely shitty long snapper?

but even our coverage isn't the same as earlier this season

24
by louis (not verified) :: Wed, 12/12/2007 - 3:37pm

Well, first of all, we sincerely apologize for all the problems over the last two days. The server situation got so bad yesterday that we had to start our server migration a day early

Oh, it's hard to mind too much when the content on here is so good. And if nothing else, the FO outage did wonders for my productivity at work. (I guess that's a plus.)

At some point, you guys should really try to move beyond this WordPress system, though. It's probably really hurting your server load that every time someone posts, they have to load the entire page of comments over again at least twice. Moving to something more efficient might help alleviate the strain on the servers. (Although I guess you guys recognize this already, as it must be at least part of the reason the open game threads are implemented differently.)

The best thing in the world would be if someday you can go to a self-moderated web forum. I haven't seen one implemented perfectly -- even Slashdot (the original?) has drawbacks -- but I can't imagine a site better suited to the concept than this one.

On the other hand, the comments here tend to be so good, maybe it would be superfluous. But as you guys keep growing, who knows?

25
by Nevic (not verified) :: Wed, 12/12/2007 - 3:38pm

Ewww! 3 of the worst 10 games of 2007 were all in week 14. There were a couple games that aren't listed there that come ot mind as being pretty ugly too...like the first half of Chicago-Washington on Thursday night. Wait, if no one sees the game, does it exist?

26
by Dan (not verified) :: Wed, 12/12/2007 - 3:42pm

For some reason ,i LOVED the prospectus and buy it year after year and listen to it. A lot of this had to do with how accurate I thought that article was on rating Quarterbacks out of college. I thought it pinned down guys like Rivers (now I have changed my mind) and Campbell (big arm, where's that so called accuracy now?)... and I just thought using all these stats is the right way to predict football. Now I don't trust my own opinion...
The DVOA is so overrated. The fact that a team (Giants) can beat the Eagles TWO times in one season, including a game (This past week) where ANY viewer could easily see the Eagles had ONE GOOD DRIVE OF THEIR SCRIPTED plays, and then NOTHING...(they got the ball first and 8 at the 8 after a jacobs fumble, and spags stuffed them to a FIELD GOAL--pathetic), not to mention, jacobs fumbled again INSIDE the eagles 10 yard line. Did I mention that Dahl (undrafted free agent rookie) safety was on the field, and after re watching the tape I saw numerous times where Pierce had to push him to a spot before a play. And Michael Johnson a 7th round pick was our other safety, in only his 3rd game starting....

What does all that I just said mean?? One thing. These stats are crap. Stats can only take you so far. I would be more impressed if you guys formed opinions on all the games you watched just by that---watching more games than other people. I ALWAYS think the Eagles are the best team in the NFC, this has been a reoccuring thought of mine for the past 4 years. Always worry about them. BUT NOT THIS YEAR BOYS. They suck this year. They are GOD aweful...

Must give credit where credit is due...the Too Deep Zone article is the best analysis ANYONE on the web gives. If this site was dedicate to more aritcles like that, it would be amazing. Not to mention the quotes article is fun to read....

27
by Nevic (not verified) :: Wed, 12/12/2007 - 3:43pm

Speaking of the ugly, injury-plagued Thursday night game, has anyone noticed that teams playing on short rest (i.e. Thursday night, except for Pakcers-Cowboys who had a normal 7 days rest after Thursday games the previous week) have included a large number of injuries. Packers-Lions on Thanksgiving saw several Packers and a few Lions go out hurt. Bears-Redskins saw lots of injuries too. Is this just an issue of small sample size, or is the long season plus short rest really making a difference. If it is, the NFL should really re-think the Thursday night games.

28
by Tom (not verified) :: Wed, 12/12/2007 - 3:43pm

Re 14:

DVOA is already adjusted for schedule, weighted is adjusted to so early games count less.

29
by gustav (not verified) :: Wed, 12/12/2007 - 3:44pm

Dear Santa,

Please let us see what would be THE greatest Any Given Sunday *ever* the week after next: "Dolphins over Patriots."

I promise I won't ask for anything else!

30
by Johonny (not verified) :: Wed, 12/12/2007 - 3:48pm

Aaron's data really argues for removing the Miami Head coach, but I really can't see them attracting anyone to that position that is better. If the Dolphins beat the Patriots it will go down as the single greatest WTF in football history. As a Dolfan I can tell you that close to Christmas I got better things to do than watch that "game". I think the Raven-Dolphins game is the best chance at a Dolphin win this year. The Ravens team that showed up last week is a team that looked a lot like they weren't anymore interested in playing football than this Miami team is. Sundays game will be an ugly affair watched by only those that like "pain". Samkon Gado was hardly the worse thing on the field for the Dolphins last Sunday.

31
by mawbrew (not verified) :: Wed, 12/12/2007 - 3:49pm

I would have thought that the Dolphins players would be highly motivated to avoid being part of a winless season and that as the end of the season got closer their intensity/play would get stronger. Another theory dashed.

32
by admin :: Wed, 12/12/2007 - 3:49pm

Loser League is now updated.

33
by Jimi (not verified) :: Wed, 12/12/2007 - 4:00pm

#25 - Quick, somebody bring up the zlionsfan DVOA hate template.

34
by Ashley Tate (not verified) :: Wed, 12/12/2007 - 4:00pm

RE #25:

[These stats are crap. Stats can only take you so far. I would be more impressed if you guys formed opinions on all the games you watched just by that—watching more games than other people.]

In short you think FO should give up their competitive advantage and act like most of the 1.3 million other sports commentators in the world? Is that about right?

35
by patriotsgirl (not verified) :: Wed, 12/12/2007 - 4:02pm

25: I think DVOA has it dead on with the Giants, and corresponds almost perfectly with the somewhat crude measure of win-loss record.

Out of their nine wins, four have been against complete disasters: Mia (barely), SF, Atl, and NYJ. They have played four games against teams that currently have winning records: Dal (twice), GB, and Min. They have lost all four (and most were not very close). So, the Giants are 0-4 against good teams; 4-0 against terrible teams, and 5-0 against mediocre-to-bad teams.

Which means, to me, that they are almost by definition an average team.

36
by chad w (not verified) :: Wed, 12/12/2007 - 4:03pm

Re 26:

There are always going to be anomalies in any season ... to use that as an example that DVOA stats suck is ignorant. Yes, the Giants beat the Eagles twice, but those games did not happen in a vacuum ... all the other games have to be considered as well. Like the games against the Packers and Bears in which the Eagles completely outplayed their opponent and got unlucky to lose. Or the domination against a Detroit team the Giants struggled to beat.

Every system has its flaws, but it's still better than looking at yards gained and yards allowed, which is relatively meaningless. In any given season, the sample size is so small that there can be apparent inaccuracies. But all DVOA is telling us is that if these two teams played ten times, they'd probably both win 5 times, which I would agree with. It just so happens the Giants won the two games that were actually played during this season.

37
by patriotsgirl (not verified) :: Wed, 12/12/2007 - 4:04pm

Er, instead of "win-loss record," I meant "win-loss" breakdown.

38
by slo-mo-joe (not verified) :: Wed, 12/12/2007 - 4:05pm

#12, 29:
OK, I am a Pats fan, and I do understand the pervasiveness (and, to some extent, the origins) of Pats-hate, but I find it absolutely befuddling that people would so openly root for a fluke loss by the Pats. Losing to the Colts, Cowboys or Steelers, sure, you may actually like them better, and think they can pull it off. But the Dolphins or Jets?

Here you are, with the privilege and fun of watching one of the best teams in history do things no other team in history has done before, beat a season win record that a very good but not truly outstanding team has secured and kept for decades thanks to circumstances beyond their real worth, and you wish that a sloppy field and a couple fumbles rolling the wrong (or right) way would cause the Pats to lose to an abysmally inferior opponent. Just for kicks and gloating.

Frankly, if the Colts (or some other team I dislike: Jets, Oakland, etc) were in the same situation, I would hope they would lose to the Pats and no one else. Just strange.

39
by Jason (not verified) :: Wed, 12/12/2007 - 4:06pm

According to your playoff odds, the odds of the Patriots winning their divisional round playoff game are: 72.8%

The odds of the Patriots winning the conference game if they have already won their divisional round game are: 67.9%

How can there be so little difference? Isn't NE likely to face a much stronger opponent in the championship game?

40
by David (not verified) :: Wed, 12/12/2007 - 4:10pm

Maybe the dolphins look over at the niners and see them with 3 wins and think 'you know, all this trying hard and not playing that badly hasn't been working... let's do what they do instead, and maybe we will win, too.' That's my only guess for why they have fallen off so badly.

41
by Boston Dan (not verified) :: Wed, 12/12/2007 - 4:11pm

Printable link just brings us back to the front page.

42
by Jimi (not verified) :: Wed, 12/12/2007 - 4:11pm

#39: The difference between New England and Indianapolis (likely AFCCG opponent) is greater than the difference between Indianapolis and, say, Jacksonville or San Diego (likely AFC Divisional Round opponent).

43
by Eric J (not verified) :: Wed, 12/12/2007 - 4:12pm

39 (Jason),

Not really. The Pats would probably face the Jags (5th in Weighted DVOA) or the Chargers (6th) in the divisional round, followed by either the Colts (3rd) or the Steelers (9th) in the AFC Championship. There's not that much improvement there.

44
by James, London (not verified) :: Wed, 12/12/2007 - 4:12pm

Re Miami:

Is there a variable in the DVOA formula for "we hate the coach, we quit"?

Because that seems to be the most logical explanation for the utter collapse of the Dolphins over the last fortnight. To echo Johonny, Miamis' biggest issue in the offseason will be finding a credible Head Coach. I'd love Parcells or Marty, but really, do they want the pain of a 4-5 season rebuilding effort?

45
by doktarr (not verified) :: Wed, 12/12/2007 - 4:14pm

Since there are currently 4 3-10 teams, there's a good chance that strength of schedule will determine who gets the second through fifth picks.

Going by current strength of schedule (linked in my name) the Jets are the runaway leader in this category, and are all but assured the #2 pick unless they win another game. (It seems the stronger schedule wins this tie breaker).

The Falcons, 'Niners, and Rams are in a virtual dead heat, with the Rams and Falcons opponents just one win ahead of the 'Niners. This will probably change, but if it ends in a dead heat, the Falcons pick third with the worst conference record (2 currently) and the Rams pick ahead of the 49ers (both have 3 conference wins, but the 49ers have all 3 in the division, which is the next tie break).

So basically, the pick the Patriots own is likely to end up as the 5th pick, unless one of the Falcons, Rams, or Jets manage to win a game. Things could change, but right now the 49ers are the losers of the tiebreakers.

46
by B (not verified) :: Wed, 12/12/2007 - 4:16pm

30: Bobby Petrino will be looking for a job in a few weeks. Maybe they can get him.

47
by Harris (not verified) :: Wed, 12/12/2007 - 4:17pm

The Eagles are 13th in DVOA, 8th in conventional offense, 10th in conventional defense and I believe they've outscored their opponents. Even with their struggles in the red zone, they should be solidly in the playoff hunt. Yet, they're circling the drain and I'm afraid villagers bearing torches and pitchforks may descend upon McNabb's home at the next full moon. What the hell is going on?

48
by B (not verified) :: Wed, 12/12/2007 - 4:19pm

47: Really bad special teams play seems to be disproportionately affecting the Eagles. And it's a part of the game that's not usually reflected in traditional stats.

49
by Sven (not verified) :: Wed, 12/12/2007 - 4:24pm

45: You're wrong, your link goes to SOS LAST year not for this year, Pats likely pick #2, if I did it right see link in my name for current order and Strength of Schedule

50
by louis (not verified) :: Wed, 12/12/2007 - 4:24pm

#38:

I agree, and I have to say, even as a die-hard Pats fan, I was sort of rooting for the Colts when they had that chance to go undefeated in 2005. In fact, I never really cared much for Dungy until he got up there and said, "We're going to try for 16-0." (Beats watching Sorgi, for one thing.)

And I completely agree: If that Colts team had done it, much better them than the 1972 Dolphins, who were very good, but not historically great.

51
by SteveW (not verified) :: Wed, 12/12/2007 - 4:27pm

Just a methodology question - how are teams treated in garbage time. For example, how are Colts rated on plays where they benched their starters? (or the Ravens on those plays)

52
by slo-mo-joe (not verified) :: Wed, 12/12/2007 - 4:29pm

#45:
Interesting how divergent the DVOA and win percentage strength of schedule values are. Top five hardest by DVOA (win/loss rank):
BUF (24), PHI (10), NE (25), KC (6), NYJ (28)
Top five easiest by DVOA (win/loss rank):
SEA (30), NO (3), ARI (19), SF (23), ATL (16)

53
by Bobman (not verified) :: Wed, 12/12/2007 - 4:30pm

#39 slo-mo-joe This Colt fan agrees. If we didn't beat youse in week 8, then I wanted (still do) you to run the table so we get another crack in week 20. The downside of that, for me, is having to listen to so much Pats love from some ignorant folks and the media. (If I ever read Simmons any more ths would be even worse. It would be worth it for Miami to steal some of the glory I am hoping the Colts will capture just to see "Simmons face" when it happens.

But otherwise, I am so greedy for my own team to win accolades and glory, that I want all their opponents to appear invincible and the ONLY team capable of beating them would be Indy. So good luck. Until week 20. Then... may your team's Gatorade be laced with laxatives.

54
by fish shure (not verified) :: Wed, 12/12/2007 - 4:31pm

47: It's the curse of Terrell Owens. Look at the 49ers... that is your future, Eagles and Cowboys fans.

55
by turbohappy (not verified) :: Wed, 12/12/2007 - 4:31pm

Re: 38

This Colts fan is rooting for the Pats to go 17-0 and then lose to the Colts in the AFCCG :o)

56
by slo-mo-joe (not verified) :: Wed, 12/12/2007 - 4:32pm

#49:
Duh, I completely missed the "2006", that explains the divergence in #52.

57
by David (not verified) :: Wed, 12/12/2007 - 4:34pm

45 - I'm pretty sure that the weaker Strength of Schedule drafts higher (meaning earlier). Since draft order is based on the amount of failure from the previous season, it makes sense that the team with the same number of wins against inferior competition would get to pick before the team that was able to 'succeed' equally against better competition.

58
by BadgerT1000 (not verified) :: Wed, 12/12/2007 - 4:34pm

Since extraordinary individual and/or team achievements are more often generated in unique environments, I am curious what is special about 2007 with respect to the NFL.

For example, when baseball expanded in 1961 all kinds of players had extraordinary power numbers including players like Norm Cash and Jim Gentile who never repeated those performances.

So in 2007 we have the Patriots bid for perfection, the Cowboys quietly having one of the best DVOA seasons in recent memory and the Dolphins stinking it up like month old cheese sitting in the sun.

What's the deal Camille??

59
by sam (not verified) :: Wed, 12/12/2007 - 4:35pm

RE: 19 - shanahan/variance

Maybe, just maybe, this can put a dent in the argument that Variance is first and foremost a product of coaching. Mike Shanahan didn't forget how to coach this year. He may not be the best coach in the whole league, and he's not a stellar GM, but he's a good coach (even if it's time for him to move on from Denver). People pointed to Jacksonville's variance last year as a sign that Jack Del Rio was a terrible coach.

This year, the Jaguars are THE most consistent team in the league. Does that make them the best team, or him the best coach? No. Has he improved, yeah, but not from 32 to 1. I would imagine that injuries, QB changes - really, any personnel changes which the team might need to adjust to, and the players just not playing well in some weeks might have more to do with it.

Is Randy Moss's proclivity for getting called for OPI a reflection of Bill Belichick's ability to coach? No. Penalties and Variance are not, IMO, the best measuring sticks for coaching ability.

60
by slo-mo-joe (not verified) :: Wed, 12/12/2007 - 4:35pm

#53:
may your team’s Gatorade be laced with laxatives.
Eh, that would be a fluke too, wouldn't it? Have you so little faith in your own team?

Regardless, you are better off feeding Purds boiled potatoes and rice for a week. His intestinal regularity is the real reason the Pats keep winning. Ask him.

61
by Mikey (not verified) :: Wed, 12/12/2007 - 4:36pm

Is it possible to see what the dvoa ratings are just for the last 4 weeks?

Just wondering where Minnie would rank.

Lotta people have got DAL-GB carved in stone but it wouldn't surprise me at all to see Minnie in the NFC Championship Game.

62
by Bobman (not verified) :: Wed, 12/12/2007 - 4:36pm

RE #53, I should have referenced #38. My bad.

63
by TomC (not verified) :: Wed, 12/12/2007 - 4:37pm

To lend further credence to the “there is no ‘Grossman is our starter’, there is no ‘Greise is our starter’, only a CHICAGO quarterback” theory, Grossman/Griese are ranked 33rd/34th in both DPAR and DVOA. The trifecta would be Orton listed at 35th by week 17. Interestingly, Adrian Petersen & Benson are ranked 47th/49th in DPAR and 50th/46th in DVOA. Maybe there’s also the “CHICAGO RB” theory as well.

Much of this has to do with the fact that all of these guys have to play behind the "CHICAGO OL."

64
by Bobman (not verified) :: Wed, 12/12/2007 - 4:39pm

#60, Hey, a win is a win. I'd prefer it to be clean and honorable and all that, but you know, your team is pretty darn good and wishing injury on them is beneath me. Besides, laxitives are funny. Plus you may need help with the running game.

Sorry, vile joke. Couldn't resist. And you don't need help anyway.

65
by Andrew (not verified) :: Wed, 12/12/2007 - 4:40pm

Harris #47:

The 2007 Eagles record is what happens when you lose a bunch of 3 and 4 point games (Packers, Bears, Patriots, Seahawks, Giants) where you fail in the redzone at the end of the game or fail at a field goal at the end of the game or muff a punt at the end of a game or play prevent and let the opponent march 97 yards in 57 seconds.

In other words, change around half a dozen or so close plays, and the Eagles end up 10-3.

But the real problem is that the Eagles have had just 2 wins by 10 points or more. What they really needed this year was not just a few breaks to make them a flukey 10+ win team, but a consistently better presence in the redzone to make most of their games not be close. Their defense played well enough to allow this, but their offense has not, and it was not helped by a complete lack of turnovers on defense and pitiful special teams for returns.

66
by gustav (not verified) :: Wed, 12/12/2007 - 4:43pm

Re: 38

C'mon, man--wanting the Jets to win this week (for non-Jets fans) would be spiteful Pats-hate. But wanting the Fins to win next week? How often do you see 0-14 beat 14-0? More unlikely than a team going 16-0, I submit!

That would be historic!

67
by Kaveman (not verified) :: Wed, 12/12/2007 - 4:46pm

deflated, #19:
It has been a roller-coaster year for Broncos fans. I wonder what the variance is, broken down by offense, defense and special teams.

slo-mo-joe, #38:
Goes to show just how much dislike Pats fans have raked up, hmm? Unless you think that all the Pats-haters, as you call them, are irrational, you have to recognize that a good chunk of that hatred has been stimulated.

68
by sam (not verified) :: Wed, 12/12/2007 - 5:35pm

67:

That data is available on the individual offense/defense/st pages.

DENVER:
10.1% variance on offense (2nd most inconsistent)
12.2% variance on defense (Most inconsistent)
3.4% variance on special teams (Most inconsistent)

69
by doktarr (not verified) :: Wed, 12/12/2007 - 5:41pm

RE: 49 et al,

I was looking at both the 2006 and 2007 SoS lists, and mistakenly put in the wrong link. The 2007 link is in my name on this post.

Moreover, that 2006 link was actually the PROJECTED 2006 SoS based on the 2005 records, and as such is meaningless. It also confused me into thinking that better SoS leads to picking higher, which is incorrect. So the Jets are basically locked into the 5th pick unless somebody wins a game.

That said, I did have one important thing right about this year. As I said, the 49ers are in a virtual dead head for SoS with the 49ers, Falcons, and Rams, and the 49ers lose subsequent tiebreakers due to winning more conference/division games. At the end of the season, it's extremely likely that the Rams SoS will be higher due to higher W/L in remaining opponents. So it basically comes down to the Falcons against the 49ers for the second pick. Assuming neither wins another game, of course.

70
by Yinka Double Dare (not verified) :: Wed, 12/12/2007 - 5:42pm

Hey, remember when we all thought the Bears had the easiest schedule in the league back before the season started?

Whoops.

71
by B (not verified) :: Wed, 12/12/2007 - 5:45pm

68: Wow, Denver is off the charts in inconsistency. Not only are they 1st or 2nd in all three phases, they've got an 11 point lead on the #2 team.

72
by Bronco Jeff (not verified) :: Wed, 12/12/2007 - 5:50pm

The huge variance doesn't surprise me at all for the Broncos.

I was one of the few that did see the Denver game last week, and the San Diego game which is listed on the worst DVOA games of the year list.

This team can be one of the top 5 teams in the league...seriously. But they have just flat out underperformed this year, to my great disappointment. The defense in particular has been poor, and once again poor field position has hurt the offense.

They still have a chance at the playoffs, but it ain't much of one.

73
by Nicky P (not verified) :: Wed, 12/12/2007 - 5:55pm

For any of you bettors out there - there is snow and major wind expected in Foxboro on Sunday.

The Pats should win, but it might not be the revenge blowout everyone seems to be expecting.

Also, as a Pats fan, I'm scared to death of them going 16-0 and then losing in the playoffs. Just being honest.

74
by chip (not verified) :: Wed, 12/12/2007 - 5:59pm

# 63. I know. As a season ticket holder, it's painfully obvious where the rebuilding effort lies. I didn't think anyone could be worse than Freddy Miller until Terry "The Turnstile" Metcalf started in place of Rueben "One-Armed" Brown.

75
by thestar5 (not verified) :: Wed, 12/12/2007 - 5:59pm

38

That would be because most football fans HATE the Pats. Its not like the Colts, no one hates or hated them. But many of us despise the Pats. I personally root against them in every game. And I take no joy in seeing a team I despise winning like they do. Thats the answer to your question, we don't get any fun out of watching them. In fact, many people will probably be Pats haters for a long time, just like how people hate the Broncos. I know I will be loving it when the Pats are going 7-9 again. I'll enjoy embarassing the Pats fans who are probably the worst in football around here. So its just that a team that has so many people sick of them is not gonna be enjoyed by most people except Pats fans.

As far as the Dolphins, isn't it just cause they lost their QB (Green), RB (Brown), and top WR (Chambers? That seems like the obvious answer to me. The only reason we haven't seen this decline earlier is because two of their games after those guys were gone were the MUD games (for lack of a better term) against the Giants and Pitt. The niners look like a much better team right now to me.

76
by citizen jason (not verified) :: Wed, 12/12/2007 - 6:03pm

Looking at the offensive/defensive charts, it just occured Minnesota-Seattle playoff game would be a pretty strange game: a great passing team (Seattle) v. a team with a great run defense (Minn), and a great runnning team (obviously, Minn) v. a team with a great pass defense. Sort of mirror images of each other. I'm not sure how that would all play out. I'd be tempted to say the running team wins, but I'm not sure I'd take Jackson over Hasselbeck in a game in Seattle ...

77
by sam (not verified) :: Wed, 12/12/2007 - 6:04pm

75:
I hate the Colts.

Also, regarding Variance:
The other thing with Variance is that it can actually penalize good coaching, insofar as a team which plays badly to begin the season but improves as the season wears on will suffer from a poor variance rating. That's why the week-to-week DVOA charts are so much more informative IMO than VAR.

78
by mawbrew (not verified) :: Wed, 12/12/2007 - 6:05pm

Re: 75

So I'm guessing you would disagree with #38's characterization of watching the Pats as a 'priviledge'? Unless, or course, you're in prison, in which case any TV at all would be a priviledge.

79
by Kurt (not verified) :: Wed, 12/12/2007 - 6:12pm

(quoting what at the moment is post #35 - patriotsgirl)

25: I think DVOA has it dead on with the Giants, and corresponds almost perfectly with the somewhat crude measure of win-loss record.

Out of their nine wins, four have been against complete disasters: Mia (barely), SF, Atl, and NYJ. They have played four games against teams that currently have winning records: Dal (twice), GB, and Min. They have lost all four (and most were not very close). So, the Giants are 0-4 against good teams; 4-0 against terrible teams, and 5-0 against mediocre-to-bad teams.

Which means, to me, that they are almost by definition an average team.

The first problem with this is that DVOA is apparently under the impression that the Eagles are a good team, and the Redskins are averageish. You can't just flip back and forth between DVOA and raw record whenever it suits your needs.

Secondly, average teams generally don't go 5-0 against mediocre to bad teams. Beating *all* of those teams is evidence of goodness.

Thirdly, the Eagles are 1-4 against good teams, 1-3 against averagish teams and 3-1 against bad teams. So what does that make them?

Let's say it makes the Eagles mediocre to bad, which is convienient when you want to dump on the Giants' strength of schedule. That would put the Giants at three spots *below* mediocre to bad. Does that sound right? Not to me it doesn't.

80
by ammek (not verified) :: Wed, 12/12/2007 - 6:16pm

Methodology question. Did you adjust DVOA for the Noah's ark game in Pittsburgh? If not, what would happen to the Dolphin and Steeler numbers if you eliminated that game - treating it as a complete fluke? Would the Steelers' be less variant? Wouldn't the Dolphins DVOA look more like that of an 0-16 team?

81
by mrh (not verified) :: Wed, 12/12/2007 - 6:18pm

I think the '72 Dolphins WERE a historically great team in that they were part of the '71-'73 run of three straight Super Bowl appearances, something almost no team has done. The '90-93 Bills had four, I don't think any other team has had three staight appearances, and of course the Bills didn't win any. The Dolphins won two of three, dominating the two they one (although they were equally dominated in the one they loss).

I think the '73 Dolphins were actually better. The '72 team benefitted from a very easy schedule but suffered from playing their back-up QB most of the year. At that stage of Morrall's career, I'd say it would be something like Testaverde playing for the Pats this year.

BTW, I was a Dolphin-hater (Colt fan) at that time - the Fins had been forced to give up a draft pick for tampering with Shula. But they were a great team; don't let their current annoying behavior obscure that.

82
by Will Allen (not verified) :: Wed, 12/12/2007 - 6:20pm

I go back and forth between whether I'd rather have the Vikings, should they not blow their playoff opportunity, play the Bucs or Seahawks. Assumimg Garcia is healthy, I guess I lean towards Seattle. Cedric Griffin seems to have made significant improvement in recent weeks, and the Vikings pass defense in the red zone has never been all that bad. The Vikings offense would have better success against Seattle than they would Tampa, I guess, and I think a hot day in Tampa, which is not impossible in January, would hinder the Vikings more than a bad weather day in Seattle.

Of course, the Vikings could go out get beat by the Bears on Monday, which would render these judgements rather less relevant.

83
by Will Allen (not verified) :: Wed, 12/12/2007 - 6:24pm

Yeah, I think the '73 Dolphins team was significantly better than their undefeated team.

84
by Mark Glickman (not verified) :: Wed, 12/12/2007 - 6:28pm

Re 26, etc (Giants are underrated/no they're not thread): I'm a Giants fan, but overall I find it hard to take exception with the Giants middling ranking. However, I am a bit skeptical of the standard rationale that the Giants have feasted only on mediocre teams this season, and just barely several times. Indeed, Giants have gone 0-4 against teams in the top 10 DVOA, and 6-0 against the bottom third of the league. But shouldn't they receive some credit rather than scorn for beating all the teams they should have? Aside from NE, Dallas, and Indy, how many other teams can claim this (even GB has a loss at home to Chicago)? Not that DVOA works this way, but I find it unconvincing when people single out unimpressive Giants wins while cherry picking other teams' wins in comparison, while neglecting others' weak performances. After all, ultimately they play 14 of the same 16 opponents as everyone else in the division, and given the widespread mediocrity in the NFC, probably a similar strength of schedule as everyone else in the conference.

85
by kevinNYC (not verified) :: Wed, 12/12/2007 - 6:32pm

Looking at the Giants, I see a team that has yet to play what we could call it's "A game". With only 3 weeks left, it appears they're incapable of having that type of performance. Their biggest problem is having to overcome all of their self-inflicted mistakes in order to win games. In their last 3 wins, the Giants have turned the ball over SIX times inside the opponents 30 yard line. Their other TWO turnovers occurred inside their 20 yard line. Add to that, 3 FGs attempted from the 10 yard line or closer, the Giants are constantly trying to snatch defeat from the jaws of victory.

No Giants fan could argue that they've beaten a good team. However, I think too much has been made of the Minnesota loss due to the fact it was the perfect storm of total team ineptitude. If we're to take that game at face value, then we (as a Giants fan) must also believe Minnesota is better than Dallas or the Packers since they beat the Giants MUCH easier than either of those 2 teams.

86
by vanya (not verified) :: Wed, 12/12/2007 - 6:33pm

#75,
sorry to disappoint you, but a lot of people DESPISE the Colts, and that's not just Baltimoreans.

In Texas Patriots hatred really doesn't seem very widespread. Most Pats hatred is limited to AFC cities. Anywhere people really like the game of football you'll find people who appreciate the way the Pats approach the game. I think sports writers hate the Pats a lot more than the general fan base does.

87
by Pat (not verified) :: Wed, 12/12/2007 - 6:37pm

Secondly, average teams generally don’t go 5-0 against mediocre to bad teams. Beating *all* of those teams is evidence of goodness.

Not really. If you assume average-average is a coin flip, 5-0 will happen fairly often. Maybe once every year or two.

The Giants are clearly not top-tier, they're clearly not bottom-tier, and the NFL doesn't have nearly enough games to distinguish between teams in the same tier.

88
by deflated (not verified) :: Wed, 12/12/2007 - 6:38pm

59 - sam on variance:

I wasn't thinking variance reflects coaching quality, more impressions of coaching style. My image of Shanahan is an upper-tier coach (excellent with single game plans, good preparing for an entire season) who is known primarily for one thing: the most consistent running game in the last 20 years. Its the lazy 'insert new 1000 yd RB here' trope from most fantasy forecasts.

I would have thought that would be a great start for a consistently low variance in DVOA since 1998; turns out it ain't this year and or a couple of other years. Makes me furiously ponder just what is the main factor in consistency (player age/experience? injuries to starters? scheme changes?).

89
by Joe (not verified) :: Wed, 12/12/2007 - 6:42pm

This will answer why the Pats are best at rushing DVOA:

The Patriots have run for 105 first downs, second-most in the league behind Minnesota (108), and are tied with Miami with 28.8 first downs per carry. Oh, and Tom Brady has nine first-down runs on third down out of 10 tries, a league-best 90-percent success rate. -- Cold, Hard Football Facts

90
by Dr. Wayne Pitcher (not verified) :: Wed, 12/12/2007 - 6:43pm

Re: Top draft picks

Question--does the 49ers' head-to-head loss to the Falcons enter into the tiebreaking procedure for determining draft position?

91
by Hemlock (not verified) :: Wed, 12/12/2007 - 6:45pm

75:

"Wah, I don't get any enjoyment, they're cheaters, wah."

Brady just wins, baby! But I do agree with you, pats fans suck. Especially the ones on this site, smriking bastards.

92
by GlennW (not verified) :: Wed, 12/12/2007 - 6:51pm

> Yeah, I think the ‘73 Dolphins team was significantly better than their undefeated team.

I don't know about "significantly". Still, I think because of all the conflicted media attention over the years the pendulum has swung unwarrantedly on the 1972 Miami Dolphins to the point that they're now only "very good". I don't know what it takes for a team to be "historically great", but regardless of the relatively weak schedule, being undefeated and both #1 in scoring offense and #1 in scoring defense of 26 teams is a pretty good start. In spite of a couple close calls, in the day and age where the running game was prevalent and games were closer overall, I'll bet DVOA would have loved the Dolphins' consistency in breaking the NFL rushing record while also shutting down opponents defensively.

93
by Will Allen (not verified) :: Wed, 12/12/2007 - 7:00pm

Well, Glenn, I do think my judgement is clouded by my childhood memories, in which a 7-7 Vikings team all but had the '72 Dolphins beat, but a better Vikings team the following year just got hammered by the Dolphins. Of course, that is a small sample size, and I really didn't mean it as criticism of the '72 Shulanistas, but more as praise of the '73 squad.

94
by patriotsgirl (not verified) :: Wed, 12/12/2007 - 7:02pm

79, 84: I didn't cherry pick anything; or, indeed, mention the Eagles at all. I chose to use conventional stats to indicate why DVOA is supportable; if I were to use DVOA to classify the Iggles as a "good" team despite their record, then I would be using the exact method that the original poster questioned.

I agree with Pat in 87 that the Giants (along with the Eagles and Skins) are a middle-tier team. I think part of what really hurts the Giants is that when they played top teams (like GB and Min), they not only lost, but played quite poorly.

And 75: I also hate to disappoint you, but lots of fans "hate" the Cowboys too. (Look at Romo's hat, fer crissake!) And, unlike the comparatively recent Pats' "hate", anti-Cowboy sentiment has been prevalent for decades. I've heard the following reasons: 1) resistance to the America's Team moniker; 2) envy/dislike of the 90s Cowboys and their "lack of class"; 3) the numerous bandwagon-jumpers the team has attracted over the years; 4) the fact that Cowboys fans are often seen as obnoxious. Etc. Sound familiar?

Much of it is, of course, completely irrational and fueled as much by emotion, division rivalry, and/or media blathering as by substance.

I will admit that Pats fans are often obnoxious, but disagree (obviously) with 91's opinion that the ones on this site suck.

95
by citizen jason (not verified) :: Wed, 12/12/2007 - 7:10pm

Will Allen:

Just curious what you thougth about the Vikes game vs. SF. I know the route was on pretty early, but the numbers (box score, drive logs, etc) are pretty ugly for both teams. Did the Vikes just shut things down at halftime and coast?

96
by ammek (not verified) :: Wed, 12/12/2007 - 7:11pm

Looking at the defensive stats, I wonder if we're finally entering a period where rushing becomes merely a sideshow. FO's been talking about this for a while, but for as long as there have been LDT, Edge, Alexander and others, it has seemed rather futuristic.

DVOA says that fully three-quarters of NFL teams now stop the run better than 'average' (average being 2004, iirc). Only the league's three astoundingly bad teams have a defensive run DVOA of more than plus-5.

Conclusion: are defenses now so quick and powerful that running is a waste of time? In other words, has the impetus for change come not from the liberalization of passing rules, but rather from the globalization of run-stopping defense? And if that's the case, why are teams still playing eight in the box?

97
by B (not verified) :: Wed, 12/12/2007 - 7:12pm

I'm a Patriots fan, and I hate the Cowboys. I really can't stand them. Ever since they were successful in the early 90s, I've hated them, and I hate them now. Other successful teams, like the Niners I liked. I even liked the Broncos championships, despite the fact that Denver whupped the Patriots for about 20 straight years.

98
by GlennW (not verified) :: Wed, 12/12/2007 - 7:13pm

Will, the last part of my commentary was more in response to statements like this one (and yes, your Vikings gave the '72 Dolphins their closest brush with defeat, with the Dolphins scoring to take the lead only in the final minutes of that game):

"And I completely agree: If that Colts team had done it, much better them than the 1972 Dolphins, who were very good, but not historically great."

No offense, but I've heard the same or similar statements being thrown around on Boston talk radio of late, and that opinion just doesn't hold up to the facts. I don't consider the '72 Dolphins to be the greatest team of all time, but they certainly weren't your "very good", marginally-above-average Super Bowl winner...

99
by David (not verified) :: Wed, 12/12/2007 - 7:16pm

89 - I saw that stat on CHFF as well, and there's something I just don't get. "28.8 first downs per carry"? What on earth? I don't even know what they're trying to say there.

100
by Jimmy (not verified) :: Wed, 12/12/2007 - 7:18pm

#74, 63

I have been despairing about Miller since about ten games into last year, and Brown looked crap all this year. I had fan-boy hopes that Metcalf would be the answer, but the amount of times he is stood watching the running back get tackled on an inside running play is really starting to get me down. While the Bears are giving Orton a go to see if he can play, why not try Beekman to see if he is any good too.

101
by Wanker79 (not verified) :: Wed, 12/12/2007 - 7:21pm

Re: 26

If you're going to make asinine comments the least you could do is follow the zlionsfan template. Aaron posts that at before the numbers every week for a reason.

Philly had more first downs (18-15).
New York had 12 total net yards more than Philly (318-306).
Both teams kicked off 4 times and punted 6 times.
Philly had 1 turnover, NY had 2.
Each team had 1 TD.
Each team had 3 FG attempts (the only miss of the game was the final play).
NY went 0-for-3 in the Red Zone. Philly went 1-for-2.
NY went 0-for-3 in Goal-To-Go (including being stopped 4 times inside the 3 yardline and settling for a FG). Philly went 0-for-1 in Goal-To-Go.
Philly held the ball for 5 more minutes than NY.

Can you please share some of your tremendous wisdom and explain to me exactly how you can watch that game and come to the conclusion that Philly sucks in comparison to the Giants?

102
by Oswlek (not verified) :: Wed, 12/12/2007 - 7:24pm

Just so you know, the "printable version" link just brings me back to the FO main page.

103
by Frankie (not verified) :: Wed, 12/12/2007 - 7:27pm

RE: Pats hatred. I don't understand why Pat's fans take such umbrage at the thought that people hate them and their team. HELLO MCFLY! It comes with the territory of winning games. You should relish in the fact that people hate your team. BTW, to suggest that its "a privilege" to watch them play is only going to increase the anger directed your way.

As a Colt's fan, I enjoy the fact that more people now hate the Colt's more than at any time in their Indianapolis history. The ironic point being that more people also LOVE the Colts now more than at anytime in their Indianapolis history. In fact, I bet there is a direct correlation to the number of fans that LOVE a team to the number that HATE them. What you really want to avoid is the dreaded apathy. For example, does anybody really HATE the Arizona Cardinals? How about the Atlanta Falcons? You really don't want apathy towards your home team.

104
by Dr. Wayne Pitcher (not verified) :: Wed, 12/12/2007 - 7:31pm

99 & 89-

I think the "first downs per carry" stat is a percentage. NE has 105 first downs in 365 rushing plays, which is 28.76%. Or 0.288 first downs per rush.

105
by AlexDL (not verified) :: Wed, 12/12/2007 - 7:32pm

I posted this earlier. I think it applies to this thread.

QUICK READS WEEK14
POST #63

It may be my masochistic nature, but I’ve been trying to slog through the FOXmb, for some golden nuggets of wisdom. (where is DVOA? you have an obligation to feed our addiction for stats and charts and rankings, you know? I hope Aaron is okay.)

What I noticed: There have been no comments on there addressing DVOA, statistics, football plays or any other topic that really relates to the majority of discussions on this board.
Can anyone come up with a simple term that can be called on when the discussion denigrates to FOXmb levels?
foxing it up?
you foxed up?…..?

I’m not sure why, but this has been my favorite so far from the FOXmb:

“wow wow wow

A dallas fan talking about:

Arrogance and lowclass. You should look at the mirror.”

…I don’t now why looking at the mirror would give anyone any insite? maybe it’s a zen thing.

:: AlexDL — 12/12/2007 @ 12:55 pm

Joe T. posted:

POST #65

*foxing it down
*letting the fox out
*opening up a can of fox
*breaking em off something real fox-like
*stupid like a fox

:: Joe T. — 12/12/2007 @ 4:40 pm

106
by Will Allen (not verified) :: Wed, 12/12/2007 - 7:34pm

Jason, I think the Vikings got out to a 27 point half time lead on a very, very, bad offensive football team, and pretty much coasted the whole second half. I have no idea of how damning an indictment that is of the Vikings.

I am not surprised that the Vikings have won four in a row, but I am very surprised that they have won four in a row by large margins. Of course, three of those opponents are pretty bad. Luckily, none of the remaining opponents are very good, either, which isn't to say that the Vikings are very unlikely to lose to any of them. If the Vikings can just manage to beat the Bears and Redskins at home, while the Saints manage to lose one more game, that would be nice, because I really don't want the Vikings to have to win in Denver on the last Sunday to secure a playoff spot, even though the Vikings offense matches up very well with the Broncos defense. Hopefully, Denver will have been eliminated by then.

107
by GlennW (not verified) :: Wed, 12/12/2007 - 7:37pm

I think it's a fundamental tenet that otherwise impartial fans aren't going to support a team that 1) wins multiple championships in a short period, and 2) pounds teams into submission in doing so. The Cowboys met both criteria in the early 1990's, and the Patriots are really only meeting criterion #2 as of this season.

All ethical judgments around running up the score aside, blowouts are just not that interesting or appreciated (at least not after halftime). I didn't really need to see the Patriots successfully throw on every down late into the 4th quarter last week, or to see Randy Moss score another fish-in-a-barrel TD against a deflated and defeated opponent in order to "appreciate" the Patriots' greatness (which is why I was glad Moss dropped that rather easy TD catch, to forestall further showboating). Now the Patriots games versus the Colts, Eagles, and Ravens games-- those were "interesting", without the Patriots sacrificing much in the way of their reputation. I'm not suggesting that the Patriots shave points or anything, but as a sports fan, viable competition is the goal, and teams which lap the field are always going to be resented as a natural response, especially in the absence of any humility around that success (which also appears to be a natural response these days).

108
by Mark Glickman (not verified) :: Wed, 12/12/2007 - 7:37pm

Re 85, 94: not to beat a dead horse, since most people on this issue seem to be saying much the same thing, but how many "impressive wins" have NFC teams compiled this year once you get past GB and Dallas? Chicago beat Green Bay in GB. Philly and Washington beat the hell out of Detroit when it looked like the Lions might be good. Seattle beat TB 3 months ago, and TB beat Tennessee when the Titans were playing well. All these teams have losses to worse teams than do the Giants. My only point is that you break down any of the middling NFC teams this year, and just about any team below the top 4 DVOA, and you draw a similar conclusion -- these are very flawed teams.

109
by Temo (not verified) :: Wed, 12/12/2007 - 7:38pm

As a cowboy fan, I've always liked Bill Belichick (I think he's the best football mind in the world right now), and still do... but I'd love to see the jets or dolphins beat the Pats. Not because I hate them, but more or less for hilarity's sake.

In any case, I think most Pats fans on this site are fine, but the couple who are pretty annoying (*cough*RichConnley*cough*) are very visible.

110
by Temo (not verified) :: Wed, 12/12/2007 - 7:41pm

Oh, and as much as I like and admire Patriot players like Brady, Wilfork, Sammuel, and Seymour (the four pats players I'd love to have on my team)... I despise Rodney Harrison. Like I really hate that HGH-using arrogant m-fer, and I'd love to see his face if Kellen Clemens or Cleo Lemon beat him for a TD.

111
by Rich Conley (not verified) :: Wed, 12/12/2007 - 7:43pm

108.

Wow Temo, insulting me before I'm even on the thread? Don't you belong on the fox board?

Probably 20% of my posts say positive things about the Pats. I think I'm more critical of them than most people (brady can't throw the deep ball, Kazcur is a turnstyle, etc).

112
by Rich Conley (not verified) :: Wed, 12/12/2007 - 7:44pm

109.

Hold on, you hate Rodney for being arrogant, but you like Samuel, who has "GET PAID" tattoo'd in giant letters across his chest? Thats not arrogant?

113
by B (not verified) :: Wed, 12/12/2007 - 7:46pm

Rodney Harrison, whom I have the utmost respect for, is kind of a dick, and his reputation for being a dirty player is mostly deserved. At least he doesn't have a rule named after him, though.

114
by GlennW (not verified) :: Wed, 12/12/2007 - 7:50pm

> At least he doesn’t have a rule named after him, though.

Rodney Harrison came along before his time. The Kimo-Palmer Rule would have been named the Rodney-Green Rule had the NFL been more proactive at the time; Rodney's hit was much more egregious and deliberate. Although it did indirectly produce a Rams' Super Bowl win...

115
by Rich Conley (not verified) :: Wed, 12/12/2007 - 7:51pm

112

Oh, hes definitely dirty, but I think ALL safties are dirty. I've seen Bob Sanders spear a prone Randy Moss, Roy Williams try to break people in half, etc. Its just part of the position.

But arguing that you dislike him because of Arrogance, and then saying you like Samuel? Thats inconsistent at the least.

116
by James, London (not verified) :: Wed, 12/12/2007 - 8:28pm

Will, how are you feeling about your promise to buy scotch for half the posters on this board? (Not me, I missed the cut-oof *sigh*)

The Vikings look pretty good for a playoff spot right now.

117
by Will Allen (not verified) :: Wed, 12/12/2007 - 8:35pm

Yeah, James I'm rather less confident than I was a few weeks ago, but two playoff wins, on the road, is far different than getting to the playoffs. I will say this, however; if they end up in a close game, in the fourth quarter, in Dallas or Green Bay come mid-January, I will have a rather bizarre mix of emotions. I should have hedged with a bet on the Vikings to succeed in making the conference finals! Perhaps it isn't too late, albeit with far less favorable odds than was the case several weeks ago.

118
by Waverly (not verified) :: Wed, 12/12/2007 - 9:04pm

How can Tampa Bay have a future schedule strength of -50.5%, when there are no teams in the league at -50% or worse?

Or do I not understand the description for how the future schedule metric is calculated?

119
by otbricki (not verified) :: Wed, 12/12/2007 - 9:07pm

The Pats' DVOA suffered considerably at the hands of the Ravens, who were clearly playing at a level far above any other effort they put forth this year. Their next game was a pathetic mail-in in comparison.

It seems that this extreme fluctuation in weekly play might not be properly accounted for in the DVOA calculations.

Any thoughts?

120
by Unshakable Optimist (not verified) :: Wed, 12/12/2007 - 9:37pm

#2 (regarding Pats run offense DVOA):

The Patriots have run for 105 first downs, second-most in the league behind Minnesota (108), and are tied with Miami with 28.8 first downs per carry. Oh, and Tom Brady has nine first-down runs on third down out of 10 tries, a league-best 90-percent success rate. -- Cold, Hard Football Facts

121
by Kurt (not verified) :: Wed, 12/12/2007 - 9:39pm

Secondly, average teams generally don’t go 5-0 against mediocre to bad teams. Beating *all* of those teams is evidence of goodness.
---------
Not really. If you assume average-average is a coin flip, 5-0 will happen fairly often. Maybe once every year or two.

No. A coin coming up heads 5 times in a row happens once in 32 tries. So it's *possible* that the Giants just happened to come up heads 5 times in a row, but not very likely. I did say "evidence", not "irrebuttable proof". Also remember that this started when patriotsgirl used their 5-0 record against them, in saying that a 5-0 record against average to bad-but-not-terrible teams is evidence of averageishness. Clearly it's not.

The Giants are clearly not top-tier, they’re clearly not bottom-tier, and the NFL doesn’t have nearly enough games to distinguish between teams in the same tier.

Agreed up to a point, though I (unlike some? not sure) think there are more than three tiers.

The NFL is the way it is, though. Channeling Herm, you gplay to win the game. When one team is four games ahead of another team in the same division, it's fair to presume that that team is better, unless there's some compelling reason to think otherwise. Some find DVOA more compelling than others.

122
by Sid (not verified) :: Wed, 12/12/2007 - 9:43pm

RE: 27

I have noticed that actually, and it's one of the reasons I'm against Thursday games.

123
by Sid (not verified) :: Wed, 12/12/2007 - 9:52pm

RE: 45

Going by current strength of schedule (linked in my name) the Jets are the runaway leader in this category, and are all but assured the #2 pick unless they win another game. (It seems the stronger schedule wins this tie breaker).

Absolutely wrong. The better draft pick goes to the team with the weaker strength of schedule. That's the tiebreaker.

124
by Bob Coluccio (not verified) :: Wed, 12/12/2007 - 9:53pm

Yeah, I wouldn't particularly relish the Packers having to beat the Vikes for a third time this year, even at Lambeau, where the conditions could actually favor Minnesota's run/stop run game. Not to mention Minnesota's confidence would be sky-high, because they would presumably have already beaten Seattle and Dallas to get there.

125
by Bob Coluccio (not verified) :: Wed, 12/12/2007 - 9:56pm

To boldly split where no infinitive has been split before. . .

126
by Sid (not verified) :: Wed, 12/12/2007 - 9:57pm

For those curious, the order right now would be:

1)Miami
2)New England
3)Atlanta
4)St. Louis
5)Jets

This is based on strength of schedule for the entire season.

127
by patriotsgirl (not verified) :: Wed, 12/12/2007 - 9:57pm

117: Heh, way to take a portion of a sentence and run with it. I said the following:

So, the Giants are 0-4 against good teams; 4-0 against terrible teams, and 5-0 against mediocre-to-bad teams. Which means, to me, that they are almost by definition an average team.

That isn't "using the 5-0 record against them"; it's saying that a 5-4 record against teams that aren't complete disasters says, to me, that they're pretty firmly in the middle of the pack.

I have no axe to grind, and I don't care that much about the NFC East one way or the other; just making an observation.

128
by slo-mo-joe (not verified) :: Wed, 12/12/2007 - 10:06pm

#78:
So I’m guessing you would disagree with #38’s characterization of watching the Pats as a ‘priviledge’?
I didn't characterize watching the Pats as a privilege, I characterized watching one of the very best teams ever a privilege. It just so happens that that team is the Pats. If it were the Colts, I'd say it's a privilege to watch them play, as much as those whiny, sore-loser, self-righteous mo-fo's turn every fiber in my body to alternative fits of violent rage and dry heaving. (Not really, I am just trying to match the rhetoric here. ;) )

Let me make another example: I am Italian, and as much as I like football, for me soccer is far more intellectually interesting and emotionally significant. Still, I consider myself privileged at having seen Maradona play in his prime, and would pay a good chunk of money to be transported in time to Mexico City in 1970, to see Pele's Brazil beat Italy in the WC final.

As for Pats' fans being obnoxious, I have seen on this board Pats haters openly wish for major *intentional* injuries to key NE players, repeatedly insult Pats fans, etc. I can understand Patriots fans' gloating may be grating to supporters of losing teams, but I would bet my left nut that all of those who are now pouring vitriol on the Pats and their fans would do the same or worse in their position. Meanwhile, I cannot remember any Patriots fan here wishing career-ending injury on Manning, either last year or in 2004.

In my modest opinion, obnoxious fans are those that put team allegiance and sheer envy of others' success and skills above any appreciation for the game and the athletes who play it. Gloating, teasing, and assorted bullshitting are just part of being a fan.

129
by Tim Kirk (York, UK) (not verified) :: Wed, 12/12/2007 - 10:13pm

re: Denver/Variance

I wonder how much of this is due to Denver's Line problems...

Their OL has been hurt by losing two long term starters through injury (Hamilton never even started the season...) and with 2 more of last years starters leaving in the off season there is now only LT Matt Lepsis (who missed much of last year injured) from last seasons' starters left.

The DL has been rebuilt (again) and the defensive scheme has changed. With the loss of Al Wilson behind them, on leadership and simply MLB experience as much as anything else, there must be ample cause for inconsistency.

Jay Cutler is also a relatively inexperienced QB, there is no veteran leadership on the WRs (Smith out and Walker missing much of the season), and with the nature of his off-the-field distractions I have trouble seeing Travis Henry as a veteran leader type...

The fact that a team in fairly obvious flux has managed some very good performances gives me some hope that they may gel more next season. They certainly have failed to fall apart totally, despite several occasions when I expected it to happen, and I think that must say something positive about the performance of the coaching staff. This looked to me like a potential rebuilding year anyway, and I'd expect next year to be similar but hopefully showing more consistency as both lines will likely change more.

The QB, WR, TE positions look secure. There are some options at RB (personally I'd love to see the back of Henry - he seems another example of Shanahan the personnel man giving Shanahan the coach trouble), the line can certainly be servicable as it is, but I would not be surprised to see some changes.

Defence seems to have some promise at DE, more or less passable LBs and a real problem in the middle at the front and the back...

It seems as though if the defense can not blow up badly then Denver can compete, and even begin to dominate some teams. But if the offense gets pressured into the 'we have to score on every drive' mentality then mistakes start to happen. Shanahan's gambling tendancies probably don't help here, but may make him at least able to empathise with his players who might go down in flames trying to much rather than the just trying not to lose too badly style of CYA football. There does seem to be a 'win or bust' attitude that sometimes turns a likely close loss into a win, but also can turn a close loss into a blowout loss, without the team actually giving up, just with it appearing as though everything has fallen apart... (it also sometime makes it seem impossible to come back from a bad start - like San Diego game - some teams would try to rebuild slowly, but Shanahan seems to like to try to grab some big momentum from somewhere and if it isn't there things just get worse - but it doesn't seem to affect the team so badly after the game...).

I still like Shanahan as a coach, but think that the lack of really solid foundation on either side of the line couple with his nature is a recipe for a wildly inconsistent season. It might not be as historical as some records that may go this year, but (as a Denver fan) I hope at least it has provided some entertainment for the neutral audience!

130
by Pat (not verified) :: Wed, 12/12/2007 - 10:19pm

No. A coin coming up heads 5 times in a row happens once in 32 tries

Psst... and there are how many teams in the NFL again? 32 times 1 in 32 is... one per year. Cut it down to "only mediocre teams" and you'll probably get one every other year or so, on average.

Agreed up to a point, though I (unlike some? not sure) think there are more than three tiers.

Dividing them up into strict tiers I just did to make a clear point.

Mathematically speaking, what you're essentially talking about is the uncertainty in an ordering of the NFL teams based on just wins and just losses.

In other words, if I come up with a rating system based on wins/losses which puts the Giants 12th, how far do they move if I change things within the uncertainty of whatever the ranking constant is?

From experience, I can tell you that that uncertainty is about a third of the total distance from the best team to the worst team. (On average - probably not this year, thanks to the Patriot insanity and 49er/Dolphin suckitude).

I will say: why do fans have this bizarre desire to have their team declared "better" than another team just because they won? Winning is its own reward - the Giants are getting to the playoffs regardless of whether or not they're a "better" team.

From the Platonic 'true strength' perspective, though, you can't say with confidence whether the Giants are 'better' than the Eagles. They're both middling teams, capable of looking like good teams some weeks, and bad teams other weeks.

131
by Andrew (not verified) :: Wed, 12/12/2007 - 10:24pm

The scariest thing about the Vikings for any NFC teams in the playoffs is their propensity to score defensive/special teams touchdowns. They already have 9 this season in 13 games and lead the NFL in that category. They had 7 last year (3rd in the NFL) and 5 in 2005 (2nd in the NFL). 25% of their touchdowns are off returns.

They are especially good at returning interceptions, and Eli Manning, Hasselbeck and Romo are all interception prone.

132
by MJK (not verified) :: Wed, 12/12/2007 - 10:30pm

Pat,

Where the desire to declare a team "better" comes in is when the question is asked "if Team A faces Team B at some point in the future, is it more likely or not that Team A will win, based on their past performance?" (Of course, this is exactlyw hat DVOA tries to do. And why fans of teams that don't get rated well via DVOA don't like it). That is why some fans have an obsession with having their team declared better.

E.g. Giants fans don't want to hear that their team is "average" this year, because average teams don't progress very far in the playoffs, and to hear that your team is average, even though they have a good W/L record and are playoff bound, means that someone is trying to take away your hope.

OK, enough philosophical rambling.

133
by MJK (not verified) :: Wed, 12/12/2007 - 10:41pm

The Pats’ DVOA suffered considerably at the hands of the Ravens, who were clearly playing at a level far above any other effort they put forth this year. Their next game was a pathetic mail-in in comparison.

It seems that this extreme fluctuation in weekly play might not be properly accounted for in the DVOA calculations.

Any thoughts?

otbricki,

The problem is, any sort of metric that measures how good a team is over a season is by definition a measure of how good the team is on average, not how good they are "on a good day" or "whenever they play an opponent who is performing according to their average performance".

The Pats played sloppy against the Ravens, and simultaneously got unlucky to draw refs that didn't call a game in a manner that fits the Pats's style of play. Yes, they don't play sloppy all that often, and don't draw such refs all that often, but IT IS POSSIBLE. A team that plays at a level of, say, "100" 95% of the time, and at a level of "60" the other 5% of the time, is, on average, a team that plays at a level of "98", not "100". In other words, the fact that the Pats exhibited that they are capable of playing sloppily, or that their style of play suffers when they happen to draw certain refs, should count for something in attempting to measure their average performance.

Similarly, even though the Ravens only performed as well as they did against the Pats, they still demonstrated that they are CAPABLE of playing at a higher level, so even if they usually don't, they still should get some bonus for playing that well against the Pats.

Of course, the problem is that the season is far too short to find out if a team that plays really above themselves in 1 game is a team that has a 5%, 15%, or even 50% chance of playing above itself. I think this is what Pat means by "uncertainty" and "bias". One uncharacteristically good (or bad) game indicates that the tail of a team's playing level distribution extends out that far, but with a limited number of measurements, we really don't know if that game lies out in the tail, or if a team's mean was really closer to that level than we would have otherwise thought.

DVOA does a better job than W/L, exactly because it looks at plays, not games, and hence has a much larger sample set, but even this isn't perfect, because plays are not independent events. I.e. all the plays in a given game will be affected by the same set of biasing circumstances, and success at some plays will affect how other plays are measured. So you're right--DVOA is not perfect. But I'm absolutely clueless how to make it better, at least to address the problem you raise.

My apologies to any statisticians who get what I'm talking about and are apalled at how I just butchered an explanation of statistical phenomena, and to any non-statisticians who have no clue what I'm talking about and are completely confused because I just explained things so badly...

134
by Kurt (not verified) :: Wed, 12/12/2007 - 10:54pm

From the Platonic ‘true strength’ perspective, though, you can’t say with confidence whether the Giants are ‘better’ than the Eagles. They’re both middling teams, capable of looking like good teams some weeks, and bad teams other weeks.

True. If they played again next week, on a neutral field, the Giants would probably be slightly favored, but obviously an Eagle win would hardly be shocking.

The problem is, as I understand it DVOA describes past performance, not current strength or likely future performance. So DVOA is saying that the Eagles have played better than the Giants this year. Which is preposterous.

And no, I don't really care. As you suggest, I'd much rather they be 9-4 with a lousy DVOA than 4-9 with a great one. I do admit I find the cult of DVOA here a bit creepy.

135
by Kurt (not verified) :: Wed, 12/12/2007 - 10:58pm

E.g. Giants fans don’t want to hear that their team is “average” this year, because average teams don’t progress very far in the playoffs, and to hear that your team is average, even though they have a good W/L record and are playoff bound, means that someone is trying to take away your hope.

For the record, while I think they're clearly above average, I absolutely don't expect the Giants to get very far in the playoffs. Whether or not they beat Tampa or Seattle, they're not beating Green Bay or Dallas. I have no hope to take away.

136
by Kurt (not verified) :: Wed, 12/12/2007 - 11:13pm

That isn’t “using the 5-0 record against them”; it’s saying that a 5-4 record against teams that aren’t complete disasters says, to me, that they’re pretty firmly in the middle of the pack.

Sure, if you completely ignore (a)the 4-0 record against the "disasters", which is better than plenty of teams (incl. San Diego, Pittsburgh, Minnesota, Cleveland, Denver, Cincinnati) have managed, and (b) those disaster teams are part of the pack, too. If the Giants were ranked in the middle of the non-disaster pack (ie 12th or so) this thread would be half the size.

You could just as easily slice it another way and say they're 0-3 against top 4 teams and 9-1 against everyone else, so...16th! Of course!

137
by Yaguar (not verified) :: Wed, 12/12/2007 - 11:33pm

130: DVOA is meant to be predictive. If you want a measure of how happy you should be with your team so far, look at the standings.

And I'll take it over any other measure as a predictive tool, and I'll buy its crazy results. Look at the week 8 DVOA rankings thread. It had a 2-5 Minnesota ranked at 16th, and a 5-2 Detroit ranked down in the mid-20s. Essentially, it claimed that Minnesota was the much better football team. Look where those teams are now.

138
by Kurt (not verified) :: Wed, 12/12/2007 - 11:37pm

130: DVOA is meant to be predictive.

But only in the sense that it measures a "true" value of past performance, right? There's not actually an element of projection built in, is there.

(not rhetorical or smart-alecky - I'm honestly not clear on this.)

139
by Andrew (not verified) :: Wed, 12/12/2007 - 11:39pm

Giants - 9-4
Points for 290
Points against 270
Pythagorean wins 7.05

Eagles - 5-8
Points for 271
Points against 262
Pythagorean wins 6.76

Both these teams are slightly better than average 7-6 teams by how they have played. The Giants have simply gotten lucky and won 2 extra games, while the Eagles have gotten unlucky and lost two extra games. DVOA pegs them right in this range 13th and the Giants 16th in the league.

Now look at common opponents. The Eagles have a clear advantage here.

NFC Central - advantage Eagles in 3 of 4 games
Detroit - Eagles win blow out, Giants lose blow out
Packers - Eagles close loss, Giants lose blow out
Vikings - Eagles win, Giants lose blow out
Bears - Eagles close loss, Giants close win

AFC East - advantage Eagles in 2 of 2 games so far
Dolphins - Eagles win, Giants close win
Jets - Eagles win, Giants win
Patriots - Eagles close loss, Giants ???
Bills - ???

NFC East - Advantage Giants in 3 of 4 games, 2 games need second result
Redskins - Eagles split, Giants win once
Cowboys - Eagles lose blow out, Giants lose twice
Head to head - Giants win, Giants close win

The Giants main advantage

140
by Yaguar (not verified) :: Wed, 12/12/2007 - 11:45pm

134: At this point, yes, there's no projection. It's just a measure of "true" past performance.

At the beginning of the year there's a projection system that estimates a DVOA for each team for the year. As the teams start playing more games, eventually the projection disappears and the rating takes over. That means the system was skeptical of Houston, even though they started 2-0, and it didn't panic on San Diego despite their poor start.

At this point it's entirely based on past performance for the season. It would be a crappy predictive tool for a team if that team suffered a huge injury, because it wouldn't adjust for that. So some common sense has to be involved.

141
by Tony C (not verified) :: Wed, 12/12/2007 - 11:54pm

106. I strongly disagree with the notion that the Patriots don't sacrifce anything in their reputation by having close wins. It's actually quite the opposite, suddenly they went from being a juggernaut to being considered very beatable by the pundits. Many still argue the merits of their dynasty based on the fact that they won each Super Bowl by "only" three points. Throw in the discussion of Spygate and suddenly margin of victory becomes a major factor in the measurement of their greatness.

142
by Cogito (not verified) :: Thu, 12/13/2007 - 12:01am

91: I'm a Pats fan, and guess what? I'm "smriking" at you right now.

143
by thestar5 (not verified) :: Thu, 12/13/2007 - 1:43am

127 interception prone,

Ehh, Hassellback is ranked 18th in INT's, and would be lower if more QB's threw as many passes as he did. Romo is pretty high, but the same principal applies to him. Also 5 were in one game (the Buffalo meltdown) and other than that he has never thrown more than one in a game. He also only has one more INT than Peyton, and I don't think anyone would call him interception prone. While the Vikings did give the Boys a good game, that was also the only game where Romom only had one TD, and I don't expect that to happen again. All in all, yeah there's some teams I'd rather face. But i think I'm more worried about Seattle or GB than the Vikes.

Also, regarding the Eagles/Giants debate. I know this may sound dumb, but put it in college ranking terms. Can you honestly rank a 5-8 team ahead of a 9-4 team that has already beaten it twice? That seems like more of a fault of DVOA to me, regarding its liking of long, piece-by-piece drives even if they stall out (which is the Eagles MO). Not trying to take shots or anything, DVOA is great!!! Just maybe one of the flaws that it has (as any ranking system will).

144
by Shawn (not verified) :: Thu, 12/13/2007 - 3:38am

Now that FO is so popular, it would be helpful to have a message board that breaks down comments from big articles like the weekly DVOA numbers into smaller pieces devoted to certain teams or aspects of the week. (Congrats on the explosive growth.)

Was the Chiefs game the worst in DVOA history? Is there any way FO can make at least a small contribution to getting Herm fired? KC fans are easily in the top 5 most passionate in the game and Herm is personally destroying their collective will. It's one thing to be terrible. Fans in KC can fight through that. It's quite another to be so depressingly, mindbogglingly, uninterestingly bad.

Really, even the smallest contribution to the ouster of Carl and Herm would be greatly appreciated.

145
by Shawn (not verified) :: Thu, 12/13/2007 - 3:40am

Oh, and for Lions fans, did Detroit beat Dallas on a DVOA basis as it appeared watching the game?

146
by tunesmith (not verified) :: Thu, 12/13/2007 - 5:38am

The Broncos went up six slots while KC went down three. No other team moved more than three slots unless I missed something. So based off of that, it seems that it was more an insanely good game by the Broncos, rather than a bad game by the Chiefs. Which would also help contribute to the Broncos wild variance. Their variance wasn't so bad last week, because they hadn't had such a good game yet.

That's a paragraph of opinions posted with almost no backup sources, so now I'm wondering how many times I'm wrong. Hooray lazy writing.

147
by Temo (not verified) :: Thu, 12/13/2007 - 6:05am

143. But that's the function of DVOA... to tell us things that normal statistics wouldn't tell us. Sure, the Giants have a much better record, but DVOA says that the Eagles are better than their record indicated. Now obviously, DVOA might be wrong, it's very possible, but that doesn't mean DVOA is broken just because it puts a team with a worse record ahead of one with a good record.

148
by Not saying (not verified) :: Thu, 12/13/2007 - 8:54am

Re: 146

it seems that it was more an insanely good game by the Broncos, rather than a bad game by the Chiefs.

If you look at the table at the bottom of Aaron's comments, you can see that the Chiefs actually played the worst game of any team this season. I think the Chiefs were already bad, so had less far to go.

Also, KC's variance went up by as much as Denver's. Denver already had the second highest last week. KC moved from 14th to 4th. Denver went up 9.5% in variance, while KC went up 11.1%.

So, basically, I'd say your wrong every sentence except the first two. I think that your variance just shot up, too, since your comments usually add a lot.

149
by BDC (not verified) :: Thu, 12/13/2007 - 10:29am

Re 137:
"130: DVOA is meant to be predictive. If you want a measure of how happy you should be with your team so far, look at the standings.

And I’ll take it over any other measure as a predictive tool, and I’ll buy its crazy results. Look at the week 8 DVOA rankings thread. It had a 2-5 Minnesota ranked at 16th, and a 5-2 Detroit ranked down in the mid-20s. Essentially, it claimed that Minnesota was the much better football team. Look where those teams are now."

The problem is that for everyone of these "successes", there are an equal number of failures. Look at the article; there is a section devoted to Miami. FO is apparently shocked, SHOCKED to find out that Miami, which DVOA has consistently ranked as a pretty bad but not absolutely terrible team (just unlucky), is in fact actually pretty terrible. This would only come as a shock to an FOer though, as pretty much everyone else has known this for months. ESPN (boo, hiss) for instance has ranked them at\near the bottom since I think week three.

150
by Temo (not verified) :: Thu, 12/13/2007 - 10:36am

149. Only this Miami team, which got blown out by two sub par teams, is not the same Miami team that started the year with Trent Green and Ronnie Brown and a defense that didn't seem like it was on its last legs. The Dolphins have gotten much worse recently; that does not mean they were always this bad.

151
by Wanker79 (not verified) :: Thu, 12/13/2007 - 11:25am

Re: 146/148

Also, looking at the rankings instead of the ratings isn't necessarily the best way to go about things.

Denver's actual Total DVOA rose by 10.30% since last week (best in the league; next was Indy with an increase of 4.90%) and KC's Total DVOA fell by 8.00% (worst in the league; next highest was Miami with a decrease of 6.60%).

Weighted DVOA tells the same story. Denver rose by 11.30% (best in the league; next was Indy with an increase of 6.50%) and KC fell by -11.30% (worst in the league; next was Miami with a decrease of 8.60%).

I'm not exactly sure how that agrees or disagrees with your overall point, tunesmith, but to me the only thing the rating differences tells me is that it's much easier for a run-of-the-mill bad team to climb into the realm of mediocrity than it is for a run-of-the-mill bad team to drop down to the pit of suckitude that is Atl/Oak/Car/Mia/StL/SF.

152
by GlennW (not verified) :: Thu, 12/13/2007 - 12:24pm

> I strongly disagree with the notion that the Patriots don’t sacrifce anything in their reputation by having close wins. It’s actually quite the opposite, suddenly they went from being a juggernaut to being considered very beatable by the pundits.

All washed away after a couple subsequent blowout wins, or even just good solid wins. I see what you're saying, but this reputation for being beatable given a certain formula only sticks if an opponent actually follows through on it; otherwise the notion was merely illusory as the Patriots recognized and overcame it. If the Patriots finish undefeated with a few close calls along the way, they're still going to be generally recognized for overall dominance.

Your point about the Super Bowl result is well taken. Using the final score of a single game, even the ultimate game of the season, to demonstrate much of anything beyond the W-L result doesn't make much sense (which doesn't mean that it's not done, of course). The Super Bowl itself may be the exception to the rule in establishing a team's legacy...

153
by Andrew (not verified) :: Thu, 12/13/2007 - 1:14pm

thestar5 #143:

It has nothing to do with the level of attempts. Romo and Hasselbeck are both pretty average in that regard.

Romo has 792 career attempts and 27 interceptions. That's 1 interception ever 29.3 attempts.

Hasselbeck is doing okay this year on this, but in 3 of the previous 5 years he has been in the low 30's in attempts per interception. Moreover, in previous meetings he played badly to Minnesota, throwing 2 interceptions each in a 2003 blowout and in a 2004 close win despite himself, and getting himself injured in a 2006 blowout. I can't imagine he has a ton of confidence going against Minnesota - just like Favre with his 0 for career records in Dallas and Philadelphia.

154
by Rich Conley (not verified) :: Thu, 12/13/2007 - 2:07pm

"That seems like more of a fault of DVOA to me, regarding its liking of long, piece-by-piece drives even if they stall out (which is the Eagles MO)"

This is a huge misconception.

DVOA doesn't "Like" anything. It gives points (or higher ratings) for things that are more likely to lead to points. Teams that have methodical long drives have a higher DVOA because teams that have methodical long drives score more points.

Basically, the giants offense is fluky.

155
by Wanker79 (not verified) :: Thu, 12/13/2007 - 2:21pm

And jumping all the way back to the second post of the thread...

I really don't understand why it's so hard to believe the NE's running game has been so successful. Click my name for a chart with the difference between each team's pass offense DVOA and their run offense DVOA. When a defense has to devote so much time and energy into defending the passing game, does it really surprise anyone that they're so successful?

I don't have any objective data to back this up, but intuitively, you'd assume that NE's running backs haven't faced (relatively) very many situations where the defense is stacking the line of scrimmage. And the o-line stats seem to fit that assumption. NE has the highest Adjusted Line Yards in the league, they're 4th in Stuffed Rank, and dead last in 10+ Yards Rank.

156
by hwc (not verified) :: Thu, 12/13/2007 - 2:51pm

I don’t have any objective data to back this up, but intuitively, you’d assume that NE’s running backs haven’t faced (relatively) very many situations where the defense is stacking the line of scrimmage.

According to Belichick, the Pats run game has faced a somewhat different problem. They've seen quite a few opponents gameplan a heavy dose of all-out blitzing. For example, Belichick says that Pittsburgh blitzed on 90% of the snaps last Sunday.

Although intended to hurry Brady, the constant blitzing is disruptive of the running game because you've got the constant threat of defenders shooting gaps into the backfield and stuffing the run before it even gets started. If you are going to take a bunch of negative running plays from blitzers in the backfield, you might as well just throw the ball over the blitzers heads, especially on those pesky WR screens and short slants.

If it were a predictable outside rush (like Freeney), you could still run inside. But, teams like Philly, Baltimore, and Pittsburgh come from all over. You never really know where they will overload your o-line. Every time you guess wrong, your runner is stuffed and you are on your way to needing to convert a 3rd and long.

157
by Diana (not verified) :: Thu, 12/13/2007 - 3:08pm

you should that avgguyphd person.

158
by Wanker79 (not verified) :: Thu, 12/13/2007 - 3:29pm

Re: 156

I can see the point about NE facing alot of blitzing, but if that's true it doesn't explain why they get stuffed less than any team in the league. If they really were under a constant blitz I'd guess that they'd be more boom-or-bust. But the o-line stats say that their running game is the polar opposite of boom-or-bust. Every running play (not literally "every" running play, but you know what I mean) picks up decent yardage, they never get stuffed, and they never break any long runs. They're just consistently good, which is very valuable, but hardly unexpected given the situations NE's passing game puts them in.

I haven't watched many Pats games this year (blowouts are boring), and I certainly haven't focused my attention towards their running game. I am in no way trying to pass off any of my analysis as fact. Just wanted to get that out of the way before you know who jump down my throat.

159
by Eddo (not verified) :: Thu, 12/13/2007 - 3:37pm

154 (Rich Conley):
DVOA doesn’t “Like” anything. It gives points (or higher ratings) for things that are more likely to lead to points. Teams that have methodical long drives have a higher DVOA because teams that have methodical long drives score more points.
I'm not so sure that's true. DVOA doesn't necessarily indicate what will lead to more points, but rather what will lead to a similar offensive performance. Basically, a team that has long, sustained drive, even if they stall in the red zone, is more likely to produce more long, sustained drives. Conversely, a team like the Giants, that has more quick-scoring drives and also more flameouts, is harder to predict. Of all things, DVOA values consistency (did Joe Morgan come up with DVOA? :-P) very highly, and a long, sustained drive is, by definition, more consistent.
Now, one could argue that an offense like the Giants is more valuable. Will I? Not really - I tend to think consistent drives are a better indicator of future performance, which meshes with DVOA. However, this is the one area where DVOA strays from game outcomes - take this example:
Team A (the Largemen) vs. Team B (the Predatorbirds). Each team has 10 offensive possessions.
Due to special teams/defense/whatever, Team A starts every drive at their own 20. On three drives, they take 4 plays to go 80 yards for touchdowns. On two more drives, they take 1 play to go 60 yards, then go 3-and-out and kick 37 yard field goals. On the remaining five drives, they go 3-and-out and punt. They will have had five scoring drives and have 27 points.
Due to special teams/defense/whatever, Team B starts every drive at their own 20. On five drives they take 12 plays to move 60 yards, stall, due to the way their offense is designed, then wind up kicking 37 yard field goals. On the other five drives, they take 6 plays to move 30 yards, again stall, and wind up punting. They will also have five scoring drives, but will have only 15 points.
Now, DVOA will most likely favor Team B (we're only talking offense, here). Yet Team A wins handily. Which offense is actually better? That's hard to say - if Team B has trouble finishing drives (and scoring touchdowns is definitely the goal of the offense), then their offense probably isn't really as good as DVOA indicates.

160
by B (not verified) :: Thu, 12/13/2007 - 3:56pm

158: The heavy blitzing is a relatively new phenomenon for defenses opposing the Patriots. In previous games, they saw a lot of nickel type defenses, which was when they started to get highly successful runs. Then, when teams started blitzing, they stopped inside running and switched to screens/slants.

161
by hwc (not verified) :: Thu, 12/13/2007 - 4:25pm

RE: #160

The Pats saw a lot of blitzing in the first few games of the year. The Jets blitzed a ton on opening day.

After week 3 or 4, teams got frightened by Moss and went exclusively to deep zones and nickle packages. It was this middle stretch of the season when Brady could eat lunch in the pocket because teams were only sending two or three guys on the rush.

Philly, having seen enough game film of Brady picking apart zone defenses, flipped the script back the other way and went to blitzing defense with physical man coverage on the receivers. That's been pretty much the approach followed by Philly, Baltimore, and Pittsburgh.

RE: #158

The Pats run when teams are in the nickle guarding the deep zones. Over the course of the year, they have thrown a lot on first down, run on second down. Or when a team just says "to hell with it" and goes nickle, the Pats will run the ball down their throats for a series to bring the safeties back up. It's all situational.

Also, watching Belichick over the years, he clearly believes that the purpose of the running game is:

a) never produce a negative play
b) convert short yardage situations for first downs or touchdowns.

I really don't think he's particularly looking to break long running plays in his offense (not that he would complain, if it happened). But, with Tom Brady and those receivers, he's got other ways of moving the ball downfield in big chunks. He's looking for his backs to convert 3rd and 1s, or pick up four yards and set up a 2nd and 6 or a 3rd and 2 -- down and distance situations where the defense is in real trouble.

Keep in mind that the Pats were evenly balanced run/pass until they hit the Philly/Baltimore/Pitt stretch of games. I think they just realized that trying to run against those particular defenses (big strong front 7s) is a fool's errand.

162
by Digit (not verified) :: Thu, 12/13/2007 - 5:12pm

re: 161

Doesn't stop idiotic Patriot fans from screaming this means Maroney is a bust because he's not getting carries.

I did wonder why the Patriots weren't trying to run- even Antowine Smith got more carries and attempts in 2001. Then I thought about who was throwing and who was receiving, and stopped wondering.

163
by Wanker79 (not verified) :: Thu, 12/13/2007 - 5:20pm

Re: 161

If Belichick really does think that way, he must be disappointed in his running game. NE's ranked 14th in Power Success (and that includes Brady's ridiculous success rate on QB sneaks), but that's besides the point.

I agree with everything you just said. But the point I was trying to address was the that NE's running game having the highest DVOA in the league isn't much of a surprise even though conventional wisdom would suggest otherwise.

By the way, I was playing around with the data intending to look at what kind of running offenses historically great passing offenses were paired with. I'll post that in a little while, but if you click my name it'll give you a little bit of perspective on the passing game.

164
by Wanker79 (not verified) :: Thu, 12/13/2007 - 6:14pm

I didn't really find anything interesting looking at how the run offenses of the top passing offenses since 1998 ranked. The top 12 pass offenses were all paired with top 10 run offenses except for the 2003 Titans (28th).

The same thing happens when you look at the best run offenses since '98. Of the top 15 run offenses, the only one that wasn't paired with a top 10 pass offense were the 2000 Steelers (21st).

I don't think anyone is really shocked that most good offenses are balanced. I'm not sure what I was expecting, but I figured since I put in the work I'd share the (unremarkable) findings.

What was interesting was that this year's Pats have the best passing offense since '98, the 11th best running offense since '98. But even though they are historically great on both sides of the ball, if you compare each team's pass offense to it's run offense, this year's Pats still rank as the third most pass-heavy offense since '98 (behind the '03 Titans {10th best Pass offense; 255th best Run offense} and the '04 Colts {2nd best Pass offense; 46th best Run offense}).

165
by MJK (not verified) :: Thu, 12/13/2007 - 7:08pm

I think the reasons for the Pats run DVOA are very simple.

(1) Most teams, even ones with a good passing game, HAVE to run from time to time because if the other team KNOWS that a team will only pass, going all out to stop the pass will usually stop the pass.

However, the Pats pass offense is so good that they don't really ever HAVE to run (as they demonstrated in their past two games). Hence they can and will run ONLY when it is optimal to do so--they are never in a "must run" situation, even facing 1st and 10 from their own 1-yard line, or 4th and 1 at midfield. Hence they generally only run when they see a defensive alignment that favors it, and not even then, always. (I think one reason for the struggles in the Baltimore game was that the Pats weren't trying to run even when it was favorable to do so--like when the Ravens had just one down lineman and brought an overload bliz all on one side, with nothing on the other side).

(2) Keep in mind the loss of Sammy Morris. Morris was proving to be a very effective runner. His power running style went very well with the Pats pass offense--defenses were so scared of Moss and Brady that they either (1) would go nuts with the blitz, or (2) play nickel and dime. Either way involves putting smaller, faster players (either pass rushing LB's and D-ends, or covering DB's) on the field at the expense of big, powerful, run-stuffers. Hence a power runner like Morris could bowl over people and get 4-5 ypc. Maroney, on the other hand, is a faster runner who relies more on his agility and cuts, and early in the year at least was injured so that he was running too upright with little power (according to Will Carrol). Hence he had less power, and it's a lot harder to succeed relying on your speed as a runner when the oppsing defense has it's fastest guys on the field to deal with the pass.

I bet the Pats run DVOA since Morris went down is significantly worse...

166
by Peter (not verified) :: Thu, 12/13/2007 - 7:46pm

Will Allen,

I agree with you about the Broncos being alive and the Vikings needing a win in week 17, but I kind of hope that this scenario DOES occur because I've got tickets for the game. I think Minnesota could stomp on Denver in a game meaningful for both teams, and I'd love to see it. I've always thought the Broncos won too easy of a SuperBowl the year the stupid Falcons denied a better 15-1 Vikings team the appearance. My psychologist says it'll be good for me.

Anyhow, you're right about Minnesota's offense matching up well with Denver's defense, and I also think that the Vikings pass rush could rattle Cutler and the Williams boys can certainly stuff the likes of Selvin Young, Travis Henry, Cecil Sapp, etc. It COULD be a very fun game to watch.

167
by Not saying (not verified) :: Thu, 12/13/2007 - 7:59pm

Re: 159/Eddo

It's hard to say which team DVOA would favor without knowing the actual plays that got them down the field. Does team B have negative rushing plays followed by 3rd and 15 conversions? Do they gain 5 yards on every play? Red zone plays also count 20% more than other plays, so all things are not equal.

Most importantly, though, you're forgetting the D in DVOA. It's impossible to say which team DVOA would favor without knowing the relative strengths of the defenses.

168
by Scott (not verified) :: Thu, 12/13/2007 - 8:24pm

I'm glad DVOA is showing Willie Parker as having probably the worst season ever for a rushing yardage leader in the DVOA era.

Rushing Leader
2007 - Willie Parker, -15.0%, ranked 42nd
2006 - LaDainian Tomlinson, 24.1%, ranked 4th
2005 - Shaun Alexander, 22.3%, ranked 4th
2004 - Curtis Martin, 20.0%, ranked 8th
2003 - Jamal Lewis, 3.4%, ranked 19th
2002 - Ricky Williams, 6.5%, ranked 13th
2001 - Priest Holmes, 21.8%, ranked 3rd
2000 - Edgerrin James, 11.9%, ranked 5th

Yet every talking head has the nerve each week to say their offense starts with Willie Parker. Sadly the coaches seem to think the same thing.

Give more carries to Davenport and let Ben throw the ball on 1st and 2nd down.

169
by Pat (not verified) :: Thu, 12/13/2007 - 11:25pm

The problem is, as I understand it DVOA describes past performance, not current strength or likely future performance. So DVOA is saying that the Eagles have played better than the Giants this year. Which is preposterous.

No - Uncorrected VOA measures past performance. Corrected VOA measures past predictable performance. DVOA measures past predictable performance corrected for opponent.

Note PHI VOA vs NYG VOA: -3.2% vs 1.3%. Couple that with a few non-predictive events, and there's the entire difference.

170
by Rich Conley (not verified) :: Fri, 12/14/2007 - 1:53am

"I’m not so sure that’s true. DVOA doesn’t necessarily indicate what will lead to more points, but rather what will lead to a similar offensive performance"

RIght, but what DVOA choses to value is based on what correlates to points. The red zone plays are more valuable because red zone performance correlates to points. Same with 3rd down, etc.

171
by Kurt (not verified) :: Fri, 12/14/2007 - 2:09am

Thanks Pat. Dumb question, but is there some summary page that lays out how they calculate it, what's considered "predictable" performance, that sort of thing?

172
by Pat (not verified) :: Fri, 12/14/2007 - 5:35am

Well, the 'how they calculate it' thing is certainly in the "Our New Stats Explained." What's predictable (in the "it predicts" sense, not in the "it could be predicted" sense) performance is kindof scattered around a bit. The big thing is that all fumbles are treated equally, because fumble recovery is random, but there's also special teams bits as well. Namely, you can't control your opponent's field goal percentage (since field goal blocking is so ridiculously rare), nor punting distance, and you certainly can't control your opponent's kickoff distance. That's the big thing.

There's also a bit of a subtlety in that since DVOA is play-based, high-leverage plays (i.e. something like an 80 yard touchdown pass, or an interception returned for a TD) naturally won't have the same weight statistically as they do in the game.

But that's just ridiculously common in all sports: performance in high-leverage situations results in great previous performance, but is nonpredictive in future performance. Should be a basic theorem in quantitative sports analysis.

173
by BDC (not verified) :: Fri, 12/14/2007 - 10:17am

Re 150:

"149. Only this Miami team, which got blown out by two sub par teams, is not the same Miami team that started the year with Trent Green and Ronnie Brown and a defense that didn’t seem like it was on its last legs. The Dolphins have gotten much worse recently; that does not mean they were always this bad."

Except that they haven't, they have sucked for a while now and only now is DVOA finally realizing that.
Green was lost in week five, Brown was lost in week 7, and the defense tanked in week 5 (dropping to 31 in the league), before finally falling to dead last in week 7. So basically, they have sucked since week 5 or 7, depending on your point of view; which not so coincidentally is pretty much exactly the same time everyone else in the country outside of FO realized that they suck.

Look, I'm not trying to cut on FO or anything, but the person I was responding to claimed that DVOA had predictive powers vs. the mainstream media, and provided an instance where that proved to be true. I am simply providing a counter-example where the that is incorrect. In that same results-oriented sense, it doesn't really matter why the mainstream media was right and FO wrong, as it is still evidence of the point.

This is one of the fundamental problems with using trailing indicators in analysis. Sure sure, we all read the warnings that "past performance does not indicate future results", but we are all encouraged to ignore that. When in fact that warning is fairly accurate.

174
by Starshatterer (not verified) :: Fri, 12/14/2007 - 10:50am

BDC (#172 )--

DVOA will tell you that a team's defense (for example) isn't playing well recently, but not why. The why is important in predicting what happens next

Example: Patriots' defense had two straight bad weeks (against the Eagles and Ravens). People suggested there were weaknesses, the linebackers were too old and slow (especially with the loss of Colvin), that the offense had covered for their deficiencies.

Then the defense rebounded against Pittsburgh. It turns out that there were other factors: though they're no younger or quicker than they were, a week of limited practice to rest up after three straight night games helped. Both Baltimore and Philadelphia played much better than their average day on offense (for example, not dropping many passes, which is mostly outside of New England's control), while Pittsburgh was a little off (due to Santonio Holmes's injury and some dubious play-calling, also largely outside New England's control), and so on. Conversely, the Eagles and Ravens did not repeat their powerful offensive performances in subsequent weeks, also for reasons (lost their emotional boost, not playing a tired defense)DVOA cannot predict.

What DVOA can show you is an expected baseline. You have to still figure out the cause of recent trend, up or down.

175
by Wanker79 (not verified) :: Fri, 12/14/2007 - 11:37am

Re: 172

DVOA never said that Miami was a good team. It just said that they were better than a handful of teams and with the way they've played they probably should have won a couple games.

If you look back at their weekly rankings (using DAVE in weeks 1-7, DVOA weeks 8-10, and WDVOA in weeks 11-14 because that's when those rankings were available), the highest Miami has been ranked was 22nd in week 5. The rest of the year they've bounced between 26th and 28th except for the past two weeks where they've dropped to 29th and 31th.

That seems to match up pretty well with my (not very well informed) opinion of the team. They've been a run-of-the-mill bad team all year that very easily could have won a couple games but have fallen apart the past two weeks.

176
by BDC (not verified) :: Fri, 12/14/2007 - 11:59am

Re: 173

I am not trying to point out flaws in DVOA (if I wanted to do that, I would simply reference their pre-season predictions) but rather, to address the point made in 137 that DVOA is predictive, even if the predictions don't appear to be accurate at the time. The point made in that post was to look at Detroit, who at the time was considered to be good by non-FO sources, but to be not so good by FO, and as he pointed out, look how that turned out. And so I simply pointed out that a team like Miami, which everyone outside of FO has considered to be at or around the worst team in the league almost all year has been classified by FO to be "just a garden-variety bad team with poor luck".

177
by BDC (not verified) :: Fri, 12/14/2007 - 12:08pm

Re 174:

I am well aware that DVOA never ranked Miami as being a good team, but as mentioned, it did indicate that they were (more then just slightly) better then their record would indicate. And as you mentioned, that they should have one a few games. This is an important distinction because, no other site seems to think that. And as I mentioned, this is the problem with using trailing indicators, they only tell you what happened in the distant past, and by the time they catch up with the present, it is already old news.

What I am trying to say, and I am not sure if I will be able to word this properly, is that if you are going to claim DVOA as a success when it looks at a team like Detroit (that everyone else thought was pretty good but DVOA didn't despite their record) and claim that as a success, you can't then go on to ignore a team like Miami which everyone else thought sucked, yet DVOA thought was "just a garden-variety bad team with poor luck", despite their record.

178
by Wanker79 (not verified) :: Fri, 12/14/2007 - 2:48pm

Re: 176

I don't think the point that we're differing on actually has anything to do with DVOA. You (and the general public) seem to think that Miami have truly been one of the worst teams in the league for most of the year and point to their winless record. I (and DVOA) seem to think that Miami has been just a bad team that lost a bunch of close games, but have been at least competitive for most of the year (at least until the past two weeks) and point to how they accumulated their winless record.

The only unexpected loss was the 18 point loss to Oakland. They lost by 17 to Dallas and 21 to NE, neither of which is an unexpected result for just your run-of-the-mill bad team. They lost by 10 to Cleveland and Philly. And they lost 6 games by only 3 points. Two of them were against other run-of-the-mill bad teams, 3 were against average/good teams, and then the Slop-Bowl game against a pretty good Pitt team (although you can probably ignore the result of any game where a punt looks like a game of lawn darts). That really doesn't sound like a team that deserves not to win a single game all season.

Just for comparison's sake, look at a truly awful team like San Fran who have 3 wins against other bad teams and they have been blown out 6 time by average/good teams and they've been blown out by a pretty bad Carolina team. Or look at St. Louis who have won 3 games against mediocre/bad teams and been blown out 4 times including a mediocre Baltimore team and a 14 point loss to a pretty bad Carolina team.

To me, any team that can at least stay competitive in all but three games (especially since 2 of them were form 2 of the best teams in the past decade) doesn't really strike me as being a team that's bad enough to deserve not to win a single game all season. And suggesting that they have been playing better than 5 or 6 teams for most of the year doesn't really sound wrong.

179
by Alex (not verified) :: Fri, 12/14/2007 - 6:09pm

Of all things, DVOA values consistency

Common misconception. Of all things, DVOA values success. Being consistent won't help your DVOA at all if you consistently fail to gain the needed yardage. And being inconsistent won't necessarily kill your DVOA, as long as your long gains outweigh your failed plays. Just ask Adrian Peterson, who's 4th in Rushing DPAR, despite being highly inconsistent, because he has so many big runs that it more than makes up for his stuffs.

There’s also a bit of a subtlety in that since DVOA is play-based, high-leverage plays (i.e. something like an 80 yard touchdown pass, or an interception returned for a TD) naturally won’t have the same weight statistically as they do in the game.

I agree completely, but it bears mentioning that an 80 yard touchdown pass would get a higher DVOA than most any other play. DVOA does like big plays.

180
by Rich Conley (not verified) :: Fri, 12/14/2007 - 6:30pm

"I am well aware that DVOA never ranked Miami as being a good team, but as mentioned, it did indicate that they were (more then just slightly) better then their record would indicate. And as you mentioned, that they should have one a few games."

BDC, do you remember these games?

WAS: 16 MIA: 13
HOU: 22 MIA: 19
NYG: 13 MIA: 10
BUF: 13 MIA: 10
PIT: 3 MIA: 0

Thats 5 playoff teams they lost to by 3 points. They lost to another playoff team (CLE) 41-31, Dallas 37-20, and hung 30 points on the best team the NFL has ever seen (in the dvoa error).

The casual, uninformed fan would "know" miami was the worst team in the league. Informed fans would know better.

181
by Rich Conley (not verified) :: Fri, 12/14/2007 - 6:32pm

sorry, dvoa era

182
by BDC (not verified) :: Fri, 12/14/2007 - 11:04pm

179, 180. I am probably not articulating my thoughts properly. My point in all of this goes all the way back to post 137 and is specifically regarding the *predictability* of DVOA. Now, we can argue all day long about whether or not the mainstream media had a *good* reason to claim Miami was terrible, or whether or not DVOA had a *good* reason to claim Miami was bad but unlucky, not terrible, but the fact of the matter is, it is now clear to all of us that Miami is in fact terrible (31 in WDVOA I believe). No matter how you want to mince words, in this specific instance, DVOA was simply wrong, and mainstream thought was in fact correct.

Look at it another way. For most of the season, mainstream thought has been that Miami is at or near the bottom of the league. DVOA has said that no, they aren't. They're bad, but not *that* bad. While DVOA may very well have been more accurate in describing their past performance, from a predictive standpoint, i.e., where will they finish the season, it certainly appears at this point that DVOA was wrong and mainstream thought was correct.

I am not arguing that DVOA was wrong in its assessment that until recently Miami was a little better then their record indicated (personally, I thought they were going to win the WAS game, and felt pretty good about the NYG game as well). DVOA is great at telling us how teams have performed in the past (I don't consider that to be particularly useful, as I can simply look at the box score, but whatever). However, the poster in 137 was specifically speaking about future results, and in this instance, it is clear that the mainstream opinion was more accurate then DVOA.

183
by Yaguar (not verified) :: Sat, 12/15/2007 - 2:12pm

182: I'd argue that's because of some specific aspects of DVOA that require subjective adjustment. DVOA doesn't account for personnel changes because it doesn't watch games and doesn't have a means to evaluate that sort of thing. (So in that way, yes, it describes past performance more than future performance.)

The Cleo Lemon/Ronnie Brown Dolphins were far from the worst team in the league. The John Beck/Jesse Chatman Dolphins probably are. DVOA can tell you that the Dolphins offense has gotten much worse, but it can't tell you why.

It's definitely not the best predictive system out there. Suppose Peyton Manning got injured after week 8 next year. A set of predictions that downgraded the Colts post-week 8, but otherwise relied on DVOA, would most certainly be better.

It is probably a little bit better than conventional wisdom even by itself, but it's much better than conventional wisdom if it's interpreted intelligently.

184
by Pat (not verified) :: Sat, 12/15/2007 - 6:32pm

I agree completely, but it bears mentioning that an 80 yard touchdown pass would get a higher DVOA than most any other play. DVOA does like big plays.

I think it's fairly obvious that most people would realize that an 80 yard touchdown is "good," although keep in mind that an 80 yard TD pass shouldn't be significantly better than a 60 yard TD pass (if at all). Once you get behind the secondary, the number of yards you gain has nothing to do with the play, it's just where you were on the field.

185
by Alex (not verified) :: Sat, 12/15/2007 - 8:22pm

I think it’s fairly obvious that most people would realize that an 80 yard touchdown is “good,”

Yeah, you'd think, but everyone keeps talking about how "DVOA values consistency" as if a few 5 yard runs would be better than a long TD.

186
by BDC (not verified) :: Sun, 12/16/2007 - 4:34am

183: The problem with that line of thinking is that is not much more then an excuse for failure. Say for instance Matt Light, Wes Welker, Tom Brady, and Randy Moss are all driving together in a car and they get in an accident which will cause them to miss the rest of the season. Subjective analysis would immediately downgrade the Pats, while DVOA wouldn't. That is one of the biggest advantages of subjective analysis. Supporters such as yourself would point out that DVOA can't possibly know this event happened (as you are doing here) and can't account for it. That's fine, I understand that. My problem is that supporters will continue to say that DVOA is the best there is, and play up its strengths, while ignoring its weaknesses. While at the same time denigrating subjective analysis, by playing up its weaknesses and ignoring its strengths.

You can't just say DVOA is the best because it predicted xyz, which turned out to be true, while making excuses as to why it also predicted abc which turned out to be wrong. I get the feeling that the average FOer knows these flaws exist, but allows themselves to ignore them because there is a good reason (in their mind) why FO missed that particular call.

Taken right from the main page of FO, "With these stats, we will attempt to bring objective analysis to football...",

except that it isn't really objective analysis if you play up all the correct predictions while making excuses for the wrong ones.

This post is getting long so I am going to start another one to address your last sentence....

187
by BDC (not verified) :: Sun, 12/16/2007 - 5:03am

"It is probably a little bit better than conventional wisdom even by itself, but it’s much better than conventional wisdom if it’s interpreted intelligently."

Except that it isn't. I don't know how to quantify conventional wisdom, so I will just use ESPN, as everyone is familiar with them and can see their predictions. I'll say they are close enough to "conventional wisdom". Let's compare predictions.

Everyone had NE at 1, so let's skip this one and move on.
ESPN 0, FO 0.

FO had Philly at 2. ESPN had them down around number 8. They are currently ranked 13. Conventional wisdom wasn't correct on this, but it was a lot more correct then FO. (incidentally, FO had the entire NFC east exactly backwards).
Conventional wisdom 1, FO 0.

Who is actually at number 2? Dallas, who ESPN had at 7, but FO had at 24.
CW 2, FO 0.

What about number 3? That would be Indy, who ESPN had at 2. FO had them down at 11, behind such powerhouses as NYJ (currently 25) and CAR (currently 29).
CW 3, FO 0.

What about 4? Both ESPN and FO had the Ravens, all were wrong. So it's a wash.

Who is actually at number 4? GB. This is often considered an FO success, as CW had them much much lower. Let's be clear here though, FO had them predicted to be slightly above average; this isn't exactly a dead-on hit. Still, we'll give them the point.
CW 3, FO 1

5? That would be Washington, according to FO's prediction. They are currently at 18, right where ESPN predicted them to be.
CW 4, FO 1.

6? The Bucs. Ah, FO's first clear success.
CW4, FO 2.

7? CAR. Um, no, they are currently 29, much closer to where ESPN placed them.
CW5, FO 2.

8th was PIT, both had the same ranking.

9th was SD; both were pretty accurate but FO was a little more accurate.
CW 5, FO 3.

10 11 12 and 13 we already covered.

14 Denver, both were pretty accurate on this but again FO was a little more accurate.

CW 5, FO 4.

15 ATL, while everyone else was thinking ATL would suck after losing Vick, FO insisted they would be average. Nope, they do in fact suck.
CW6, FO 4.

16 SEA, but ESPN said they would be 9. They are actually 9.
CW 7, FO 4.

I'm not going to list every single team; I think this is enough. I have actually gone through every team though, and can provide the data if you would like. FO actually catches up slightly in the end, such that they end up fairly similar in accuracy. But of course, FO supporters don't claim it is AS good as any other system, they claim it is better. That is however simply false; anyone who cares to can go back and see where they ranked teams and when, and compare that to where mainstream sites such as ESPN (or whomever) ranked them and see that it is no worse, but also no better, then any other ranking system out there.

188
by Rich Conley (not verified) :: Mon, 12/17/2007 - 2:28am

187

What the hell are you getting those numbers from?

189
by Kurt (not verified) :: Mon, 12/17/2007 - 2:49am

187 - It's pretty ironic that DVOA seems to be eking out close wins, and surrendering horrendous blowout losses. DVOA would hate DVOA ;)

190
by BDC (not verified) :: Wed, 12/19/2007 - 6:40pm

188: FO's preseason projections vs espn's preseason projections.