Writers of Pro Football Prospectus 2008

Most Recent FO Features


» Futures: Nick Chubb & Sony Michel

The Georgia Bullddogs' dynamic duo should be on NFL rosters at some point in the next 72 hours. Which will be the better pro? That depends on what kind of running back you're looking for.

26 Dec 2006

Week 17 DVOA Ratings

by Aaron Schatz

Here are the latest DVOA ratings, with commentary now up on FOXSports.com. Prepare to be subjected to more long-winded Houston Texans analysis.

Pages are up now for offense, defense, and special teams. Individual stats also updated. Adjusted line yards stats will be updated later tonight. You can also check out the Mike Harris Playoff Odds Report. Mike has updated it with fixed tiebreaker rules, and you can even click a link that tells you how many times each tiebreaker comes into play in the 10,000 times it runs the remainder of the season.

* * * * *


Apparently, the Amazon folks decided they better get all the Pro Football Prospectus books out of the warehouse or something, because they've dramatically cut the price on our first two books to $2.99 each. There is a lot of stuff in these books that is evergreen research, and not at all dated just because the 2005 season is history and the 2006 season is close to done. The Patrick-Theismann-Maguire parody in the 2006 book is worth the three bucks alone. If you are new to Football Outsiders and want to know more about what we do, or you just never got around to buying one or both of the books, this is your chance. Please order through our site so that we get a small percentage.

It feels strange to be on remainder, but hey, our loss is your gain, and it isn't really our loss anyway, because maybe we'll get some more royalties or something.

The 2007 book is also on Amazon for pre-order, for anyone who likes to do things really, really, really early.

* * * * *

These are the Football Outsiders team efficiency ratings through Week 16 of 2006, 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 based on strength of opponent as well as to consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver/Mexico City) and week of season.

WEIGHTED DVOA is based on a formula which discounts games more than eight weeks ago in order to get a more accurate picture of how teams are playing now. This is the formula used for the rankings at FOXSports.com.

Remember that you can always use the keyword "DVOA" to access the latest DVOA commentary at FOXSports.com.

To save people some time, please use the zlionsfan template for all complaints:

<team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>

1 SD 33.0% 1 13-2 29.8% 2 23.5% 2 -5.5% 11 4.0% 4
2 BAL 31.0% 3 12-3 30.5% 1 2.8% 13 -24.7% 1 3.5% 5
3 CHI 29.3% 2 13-2 22.9% 5 0.5% 16 -20.5% 2 8.3% 1
4 PHI 24.5% 6 9-6 19.6% 6 17.8% 3 -8.7% 8 -2.0% 22
5 NE 23.0% 5 11-4 25.9% 3 9.8% 7 -8.9% 7 4.3% 3
6 JAC 22.3% 4 8-7 25.3% 4 3.2% 11 -18.9% 3 0.2% 17
7 IND 18.4% 7 11-4 19.1% 7 33.0% 1 11.3% 26 -3.3% 26
8 NO 14.1% 10 10-5 17.7% 8 14.9% 4 1.2% 18 0.4% 15
9 DAL 12.4% 8 9-6 8.3% 10 12.3% 6 -1.0% 15 -0.9% 20
10 NYG 11.1% 9 7-8 6.4% 11 7.5% 9 -3.4% 13 0.3% 16
11 CIN 10.5% 11 8-7 10.2% 9 13.3% 5 4.3% 22 1.5% 12
12 PIT 6.2% 12 7-8 5.9% 12 2.7% 14 -8.0% 10 -4.5% 30
13 KC 4.9% 13 8-7 1.6% 15 4.3% 10 1.0% 17 1.6% 11
14 CAR -0.2% 14 7-8 1.9% 14 -5.6% 20 -8.3% 9 -2.9% 24
15 BUF -1.1% 15 7-8 -1.6% 17 -6.9% 22 -0.1% 16 5.8% 2
16 DEN -3.3% 17 9-6 -3.5% 18 -5.2% 19 -1.7% 14 0.2% 18
17 MIA -4.1% 16 6-9 0.6% 16 -12.8% 27 -10.3% 5 -1.6% 21
18 MIN -5.7% 18 6-9 -5.2% 19 -14.4% 29 -13.2% 4 -4.5% 29
19 NYJ -7.4% 19 9-6 -8.4% 21 1.5% 15 12.1% 27 3.1% 6
20 GB -7.7% 22 7-8 -5.8% 20 -7.4% 23 -4.1% 12 -4.4% 28
21 ATL -9.2% 21 7-8 -14.7% 23 -4.1% 18 1.4% 19 -3.7% 27
22 WAS -10.4% 20 5-10 -16.1% 25 2.8% 12 14.6% 29 1.4% 13
23 TEN -11.0% 23 8-7 5.3% 13 -8.1% 24 4.7% 23 1.8% 10
24 STL -11.7% 24 7-8 -19.2% 28 9.0% 8 16.0% 31 -4.7% 31
25 SEA -15.8% 26 8-7 -16.5% 26 -14.2% 28 3.7% 21 2.1% 9
26 CLE -19.8% 25 4-11 -23.7% 31 -16.6% 30 5.8% 25 2.6% 8
27 TB -20.6% 29 4-11 -19.0% 27 -18.4% 31 2.8% 20 0.6% 14
28 HOU -20.8% 27 5-10 -15.5% 24 -0.9% 17 17.1% 32 -2.8% 23
29 ARI -21.7% 30 5-10 -11.0% 22 -12.0% 26 4.8% 24 -4.9% 32
30 DET -22.3% 31 2-13 -20.9% 30 -11.1% 25 14.1% 28 2.9% 7
31 SF -22.6% 28 6-9 -19.3% 29 -6.9% 21 15.5% 30 -0.2% 19
32 OAK -27.7% 32 2-13 -24.2% 32 -34.4% 32 -9.9% 6 -3.2% 25

  • NON-ADJ VOA shows what the rating looks like without adjustments for strength of schedule, luck recovering fumbles, or weather and altitude on special teams.
  • ESTIMATED WINS uses a statistic known as "Forest Index" that emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close.  It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles.  Teams that have had their bye week are projected as if they had played one game per week.
  • PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road. With just one game left, there's no need to list FUTURE SCHEDULE anymore.
  • 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).

1 SD 33.0% 13-2 38.7% 11.5 1 -5.8% 28 8.2% 30
2 BAL 31.0% 12-3 33.3% 11.0 3 -1.7% 22 17.4% 11
3 CHI 29.3% 13-2 38.4% 10.7 4 -9.0% 32 22.5% 2
4 PHI 24.5% 9-6 20.1% 11.0 2 -0.1% 18 13.8% 20
5 NE 23.0% 11-4 24.8% 10.2 5 -0.3% 19 19.9% 4
6 JAC 22.3% 8-7 20.1% 9.0 7 1.8% 12 31.7% 1
7 IND 18.4% 11-4 15.9% 9.8 6 2.7% 10 17.9% 9
8 NO 14.1% 10-5 15.0% 9.0 8 -1.6% 21 15.4% 16
9 DAL 12.4% 9-6 11.7% 8.5 9 1.5% 14 17.7% 10
10 NYG 11.1% 7-8 5.7% 8.4 10 4.7% 5 12.9% 22
11 CIN 10.5% 8-7 7.3% 8.3 11 4.1% 6 10.3% 26
12 PIT 6.2% 7-8 2.8% 7.8 14 2.8% 9 18.6% 8
13 KC 4.9% 8-7 5.5% 8.1 12 -2.9% 24 20.4% 3
14 CAR -0.2% 7-8 -1.7% 7.1 16 0.2% 17 11.6% 24
15 BUF -1.1% 7-8 -4.8% 7.5 15 3.9% 8 15.7% 15
16 DEN -3.3% 9-6 -3.2% 6.7 19 2.7% 11 17.2% 12
17 MIA -4.1% 6-9 0.0% 6.7 20 1.6% 13 14.1% 19
18 MIN -5.7% 6-9 -1.8% 7.8 13 -4.1% 27 6.4% 32
19 NYJ -7.4% 9-6 -6.2% 6.9 17 1.2% 15 14.4% 18
20 GB -7.7% 7-8 -8.9% 6.6 21 -3.3% 25 16.9% 14
21 ATL -9.2% 7-8 -2.4% 6.8 18 -1.1% 20 19.0% 7
22 WAS -10.4% 5-10 -13.7% 6.5 23 4.0% 7 9.9% 27
23 TEN -11.0% 8-7 -13.1% 6.6 22 8.6% 1 16.9% 13
24 STL -11.7% 7-8 -0.5% 6.4 24 -8.3% 30 11.8% 23
25 SEA -15.8% 8-7 -12.1% 5.5 27 -6.7% 29 10.6% 25
26 CLE -19.8% 4-11 -26.0% 4.8 31 5.3% 4 13.5% 21
27 TB -20.6% 4-11 -29.2% 4.9 30 6.1% 2 8.8% 29
28 HOU -20.8% 5-10 -22.0% 5.2 28 5.3% 3 14.6% 17
29 ARI -21.7% 5-10 -12.3% 5.5 26 -8.6% 31 19.2% 6
30 DET -22.3% 2-13 -22.6% 5.0 29 -2.6% 23 8.1% 31
31 SF -22.6% 6-9 -22.3% 5.5 25 -3.7% 26 19.3% 5
32 OAK -27.7% 2-13 -33.5% 3.9 32 0.3% 16 9.3% 28

Posted by: Aaron Schatz on 26 Dec 2006

113 comments, Last at 31 Dec 2006, 7:47pm by The Mul Dawg


by aj (not verified) :: Tue, 12/26/2006 - 7:49pm

And this Bears fan weeps a bit.

by Go Chargers (not verified) :: Tue, 12/26/2006 - 7:51pm


by Travis (not verified) :: Tue, 12/26/2006 - 7:58pm

The tiebreakers on the Playoff Odds Report are still (somewhat?) broken - Jacksonville, Tennessee, and Kansas City have no chance at being the #5 seed - Jacksonville would lose all tiebreakers involving Denver, Tennessee would lose a tiebreaker involving either Denver or the Jets, and Kansas City would lose any tiebreaker involving the Jets.

by Eli (not verified) :: Tue, 12/26/2006 - 8:18pm

what does the jets dvoa look like if you remove the jacksonville game? there hasn't been a tremendous amount of movement up even as that debacle fades, so is it not a significant cause of the jets' low DVOA, especially as compared to our cake of a schedule?

by PackMan (not verified) :: Tue, 12/26/2006 - 8:23pm

What is the lowest team DVOA of any team to make the playoffs (obviously only since these stats have been recorded)?

I think Seattle (at 25) could very well set the new record.

by ChrisFromNJ (not verified) :: Tue, 12/26/2006 - 8:33pm


The 1998 Arizona Cardinals were #29 and -19.2% in DVOA, which you might think would be the worst.

However, the cake is taken by the 2004 St. Louis Rams, with a rank of 30 and DVOA of -23.9%.

Ladies and gentlemen, the NFC West.

by MLA (not verified) :: Tue, 12/26/2006 - 8:38pm

There is a blog post on the FO blog that goes into detail on the worst DVOA by a division winner and a playoff team. You can go to the blog and check it out, but the answer to your question is no. Saint Louis in 2004 had the lowest DVOA rank by a playoff team, ranking 30th.

by MLA (not verified) :: Tue, 12/26/2006 - 8:41pm

6 beat me to it. I really need to learn to type faster!

by MLA (not verified) :: Tue, 12/26/2006 - 8:48pm

Loathe as I am to make a triple post, I should note that in my post #7, I said "the answer to your question is no" when you didn't actually ask a yes or no question. I was trying to say that no, Seattle isn't in danger of becoming the worst team DVOA-wise to make the playoffs. St. Louis in 2004 holds that dubious honor. Sorry for the confusion.

by Travis (not verified) :: Tue, 12/26/2006 - 8:53pm

what does the jets dvoa look like if you remove the jacksonville game? there hasn’t been a tremendous amount of movement up even as that debacle fades, so is it not a significant cause of the jets’ low DVOA, especially as compared to our cake of a schedule?

The Jets rank 15th in past schedule (1.2% DVOA), so that shouldn't be the cause.

Is it possible to break down the Jets' DVOA by quarter? Just by observation, they allow a ton of points and yardage in the 4th, compared to the first 3.

by Larry (not verified) :: Tue, 12/26/2006 - 8:56pm

From the odds report you can discern from the expected wins (to 1 significant figure) the odds on each game.

Also, might as well give us the odds for each possible playoff matchup.

And Whoa - how about the difference between weighted and unweighted DVOA for Tennessee. That's huge.

by Duff Soviet Union (not verified) :: Tue, 12/26/2006 - 9:04pm

Do teams with a much higher weighted DVOA than full season DVOA usually enjoy success the next year? I remember the 1997 Falcons were a much better team at the end of the year and then went 14-2 in 1998, but are they the rule or an exception? A good poster would look this up themselves, but I'm lazy.

by kevinNYC (not verified) :: Tue, 12/26/2006 - 9:31pm

As I've said repeatedly, the one number on this site that directly correlates to the Giants' drop is the o-line play. The Giants are currently ranked seventh in adjusted line yards. That's not bad, but they were ranked first after week 9. They're ranked 30th in runs behind left tackle. Thank you Bob Whitfield... for absolutely nothing.

Should I take any solace in the fact only 2 teams with a top 10 past schedule has a winning record?

by Mike (not verified) :: Tue, 12/26/2006 - 9:50pm

DVOA has pinned the Eagles as a better-than-their record team all season. Then the Eagles went thru a swoon culminating in McNabb going on IR and everyone questioned the Eagles consistently being in the top 5-6 in DVOA. Looks like they are a better team than everyone thought though. Even after losing a guy having an MVP caliber season they managed to climb into position to win the NFC East, and they are doing it by knocking off division rivals on the road with JEFF GARCIA. That offensive line and Brian Westbrook deserve some serious love.

by Mike (not verified) :: Tue, 12/26/2006 - 9:54pm

#6: And that St. Louis team won a playoff game too! They went on the road and knocked off Seattle for the third time that season. That was the year that two 8-8 teams made it from the NFC and both of them advanced to the Divisional round. More proof that Jim Mora Jr. is/was overrated as a head coach. His seemingly great first season was mostly the result of Carolina's Super Bowl team falling apart due to injuries, the NFC sucking and the easiest draw in Divisional round history. It was a cakewalk to the NFC Championship game, and he got the Eagles without Terrell Owens to boot (though he got spanked anyway).

by Becephalus (not verified) :: Tue, 12/26/2006 - 10:17pm

DVOA knows all, DVOA sees all. I am curious about CAR ranking though. Their position seems out of place with their play. I suppose the DEF is just better than I realize.

by David (not verified) :: Tue, 12/26/2006 - 10:21pm

Am I the only one amazed at Oakland's four estimated wins? How bad does DVOA demand that a team be before it expects them to tank as badly as the Lions and Raiders have this year?

by throughthelookingglass (not verified) :: Tue, 12/26/2006 - 10:26pm

From the obscure stats department:
Indianapolis and Oakland each have a 44.3% gap between their offensive and defensive DVOA. Who will be the most unbalanced team in the NFL by years end?
FWIW, Oakland's ST DVOA is -3.2%, while Indy's is -3.3%. Clearly, something very strange is happening.

by Nathan (not verified) :: Tue, 12/26/2006 - 10:32pm

I know we have no run defense, but do you have to compare us to Oakland?


by Mike (not verified) :: Tue, 12/26/2006 - 10:36pm

Carolina's play has dropped significantly in my opinion due to the Jake Delhomme injury. I really don't understand the offense. One week Chris Weinke is throwing 62 passes and two weeks later he is throwing 7. I know that Mike Wahle was out leaving us with one legit starter against a great pass rush, but still... 7?!

That said, before the 4th quarter of the Cowboys game the Panthers were playing like a team ranked in the top 10. They just couldn't close out games. They had leads going into the 4th quarter in every game except the Atlanta one. A stupid throwback on a punt return cost them the Minnesota game. A 32 yard diving catch by Chad Johnson on 4th down and a stupid interception in the endzone by Delhomme let the Bengals slip away with a win. Missed tackles let a TE Chris Cooley run 66 yards for the game winning touchdown in the 4th quarter against the Redskins. First and goal at the Philly 7 and Jake Delhomme's fade to the corner to Keyshawn is picked off by Lito Shepherd (who MAY or may not have pushed off). This was after Philly scored 10 straight to take a 3 point lead in the 4th.

Really, five bad plays may have kept Carolina from being 11-4 right now. Of course it's hard to say that, so let's say they split the Cincy and Philly games because they both go to OT. They would have won the Minnesota game without that stupid fumble. All the Minnesota fans I talked to all agree with that. The Redskins game could have gone either way without that long play. Still, those two extra wins would have put Carolina at 9-6 and that's about in line with their DVOA projection. They've just screwed up at very inoppurtune times this season.

by b-man (not verified) :: Tue, 12/26/2006 - 10:44pm

I updated the weekly game charts from the calculated data.
DVOA Game Charts

by Pete C (not verified) :: Tue, 12/26/2006 - 10:56pm

Only problem is that the NFC projections do not take into account the fact that the Bears have no incentive to play the game, and thus will likely rest most of their players (at least for the second half). I have to imagine that Green Bay's chance of making the post-season is greater than 2.7%. Otherwise, thanks for the data as its very fun and interesting.

by admin :: Tue, 12/26/2006 - 11:27pm

Commentary now up: http://msn.foxsports.com/nfl/story/6307744. Individual stats updated as well.

Someone asked how the Jets would do if you completely removed the Jacksonville game from the ratings. These are total ratings, not weighted ones:

Offense: 5.9% (12)
Defense: 10.3% (26)
S.T.: 4.4% (3)
Total: -0.1% (14)

by Chris (not verified) :: Tue, 12/26/2006 - 11:34pm

RE: 14

Did DVOA get it right? Because the Saints, with a better record and a win over the Eagles when they had McNabb, have been rated below the entire time. That is, knocking off the McNabb Eagles when they were unbeatable and near the top of these same DVOA ratings.

In fact, what DVOA "knew" prior to the Saints-Eagles game was that the Eagles at 45.5% would stomp the Saints at 11.1% (not including DAVE. DAVE makes it even more embarrassing). Oops

insert zlionsfan template hur hur, hur hur

by David (not verified) :: Tue, 12/26/2006 - 11:45pm

Yeah! A team on the road, with an outlier DVOA score at the time (a score which has since regressed to the mean), lost a close game to a good-but-lower-ranked team that has been brutal at home this year! Take THAT, Outsiders!

by Chris (not verified) :: Tue, 12/26/2006 - 11:47pm

Re: 25
Saints are actually 4-3 at home, 6-2 away

by milwaukee (not verified) :: Tue, 12/26/2006 - 11:50pm

Joe Thomas plays for Wisconsin, not Penn State (see Oakland blurb).

by admin :: Tue, 12/26/2006 - 11:53pm

(hits head against bricks regarding Joe Thomas)

by Jason H (not verified) :: Tue, 12/26/2006 - 11:56pm

So Chris, should the Titans be ranked ahead of Indy & Jax?

Or maybe you think Houston is better than Indy?

Or should Washington be ahead of the Saints?

One game does not make a season.

by Yaguar (not verified) :: Wed, 12/27/2006 - 12:06am

17: Try the 2005 49ers.

They were actually extremely lucky to win 4 games according to FO metrics. The Lions and Raiders, according to FO metrics, have been a little bit unlucky. Most 2-14 teams are actually just 4-12 teams that got a little bit unlucky. (And some 4-12 teams, like the 2005 Niners, are 2-14 teams who got a little bit lucky.)

The Raiders might be an extreme black hole of suck on offense, but to get really low estimated win numbers, they'd have to be terrible on defense like the 2005 Niners were. They were last in offense, second to last in defense.

by Moltar (not verified) :: Wed, 12/27/2006 - 12:09am

re 24: "Message Board Curse" jokes notwithstanding, no statistical ranking method is going to be a foolproof predictor of which team will win any particular matchup; that's not what statistics are for. Get over it already.

More to the point, the Eagles' recent 4-1 run vindicates their season-long placement in the upper regions of the DVOA rankings quite nicely.

by young curmudgeon (not verified) :: Wed, 12/27/2006 - 12:19am

Aaron, a number of times in the commentary on Fox, you mentioned luck or serendipity. Too many analysts and fans forget the role that luck/fate/whatever you want to call it plays in deciding the outcome of games. I'm very glad to see you pointing it out. Branch Rickey maintained that "luck is the residue of design," and he was correct in that a team still has to be good enough to take advantage of the opportunity that fate places before it, but (as unsatisfying as it might be for analysts or fans), sometimes a win or loss just occurs because of a random event. (I'm a Steelers fan, I gratefully remember the Immaculate Reception, when fortune smiles on you, you should smile back.)

by kleph (not verified) :: Wed, 12/27/2006 - 12:25am

most raider fans i know would not term the immaculate reception "luck."

by Jerry (not verified) :: Wed, 12/27/2006 - 12:37am

most raider fans i know would not term the immaculate reception “luck.�

Too many syllables?

(My apologies. I couldn't resist.)

by Mr Shush (not verified) :: Wed, 12/27/2006 - 12:53am

I agree that even Millen probably won't pass on Quinn, but I can definitely see Al Davis ignoring Thomas in favour of Calvin Johnson. If only had had a bit more of an attitude . . .

Seriously, I pity the first two players drafted this year. Going to a bad team is one thing. Going to a team that looks set to spend the entire forseeable future mired in its ongoing organisational ineptitude, with no real prospect of relief nor of any attempt at change, just doesn't bear thinking about.

by emcee fleshy (atl/sd) (not verified) :: Wed, 12/27/2006 - 12:57am

Re - the Playoff odds chart. I am terrified to see that there is a 0.8% chance of Jim Mora not getting fired.

Oh, and #34 - well done.

by Sophandros (not verified) :: Wed, 12/27/2006 - 2:35am

Chris, here's an idea: Calm down and read the explanation of the stats. Additionally, check out the glossary, particularly the FOMBC.

by NF (not verified) :: Wed, 12/27/2006 - 2:40am

If Kurt Warner throw 5 TDs in Week 17, Matt Millen is going to trade the #1 pick to Arizona for Kurt Warner and their pick, and will draft Calvin Johnson. Okay, they won't do it because fans will set fire to Ford Field, but you know he will want to.

by kibbles (not verified) :: Wed, 12/27/2006 - 2:46am

From the Chiefs section at Foxsports:
Larry Johnson: 383 carries and counting. I've written about the "370 barrier" many times: Running backs with more than 370 carries in a season either lose effectiveness or suffer a major injury within two years unless they are named "Eric Dickerson." Kansas City fans are invited to e-mail mailbag-at-footballoutsiders.com with reasons why LJ will follow in the footsteps of ED instead of guys like Jamal Anderson, Terrell Davis, Jamal Lewis, Eddie George and Shaun Alexander.

Ladanian Tomlinson down?

I'm not saying that the wheels aren't going to suddenly and violently fly off of the LJ train at some point, I'm just saying that Tomlinson has sort of been worked pretty hard in the past. If you count rushes and receptions, in his 5 seasons in the league he's had 398, 451, 413, 427, and 390 touches. So far this season, he's had another 387- and by all accounts, he's having a an alright year.

LJ's had 419 touches this year. He'd need another 32 next week just to tie Tomlinson's career high... and there's still a very strong possibility that even if Johnson bettered Tomlinson's career mark next week, that Tomlinson would just go and break Johnson's new mark in the playoffs (13 touches next week, plus 20 a game in 3 playoff games would get Tomlinson to 460- and that's a very low projection, all things considered).

True, I'm combining rushes and receptions here, but unless someone knows any good reason why rushes would be any more punishing than receptions, I don't see what's wrong with that. Both end in a tackle, right?

Re #35: Seriously, I pity the first two players drafted this year. Going to a bad team is one thing. Going to a team that looks set to spend the entire forseeable future mired in its ongoing organisational ineptitude, with no real prospect of relief nor of any attempt at change, just doesn’t bear thinking about.

I agree. Like, remember the Eli Manning draft? The Chargers were so bad that season that their columnists were joking that they were the worst team on their own schedule, and there was *NO* reason to think they'd improve in the slightest, ever. Poor Eli came thisclose to being stuck in woeful San Diego, mired in absolute suck for years on end. I betcha he feels like he dodged a bullet, there, avoiding getting stuck in San Freaking Diego for the rest of his career.

by NF (not verified) :: Wed, 12/27/2006 - 2:54am

I'm not a Titans fan, but I have to think Aaron is over-antihyping th Titans for 2007. Unlike the Miami Dolphins, trhe Titans have been in full rebuilding mode for 3 years now, and are playing respectably with Travis Henry as their #1 RB and Drew Bennet as their #1 WR. They have shed everyone who was or is good and was not on their rookie contract nor a bargain FA pick-up. The team will make some serious noise in the offseason FA market after Steve McNair's cap hit is gone. The defense is a mix of savvy veterans and promising young players. The team has been getting better as the season goes on and should continue to do so during the offseason.

by hector (not verified) :: Wed, 12/27/2006 - 3:00am

The collective worth of the teams the Titans just beat seems to be a lot more than the collective worth of Miami's run last year (at San Diego, impressive, but the Patriots mailed in the finale in 2005 and there's a lot of filler there). See Miami schedule on link.

Tennessee has won six straight games as an underdog, which is pretty much ridiculous.

by Scott de B. (not verified) :: Wed, 12/27/2006 - 3:01am

True, I’m combining rushes and receptions here, but unless someone knows any good reason why rushes would be any more punishing than receptions, I don’t see what’s wrong with that.

The stats show it is carries, not touches, that determine running back wear and tear.

by Yaguar (not verified) :: Wed, 12/27/2006 - 3:11am

40: As a Colts fan, I can tell you that sometimes all of that gets thrown out the window when an elite quarterback enters his second year. Sure, the Colts were a perennial rebuilding team in 1998. In 1999, that didn't matter in the least, because they had Peyton Manning.

I think that for the first time in a while, there will be serious competition in the AFC South again. I'm kind of looking forward to it.

by Yaguar (not verified) :: Wed, 12/27/2006 - 3:16am

35: I'm almost certain Millen will take Quinn if he's there. He seems to gravitate towards the Fantasy Football positions with his first rounders. Was he there to take Jeff Backus? Since then, it's been QB, WR, WR, WR + RB, LB.

by NF (not verified) :: Wed, 12/27/2006 - 3:17am


PFP2006 did a fine statistical study on the 370 carries phenomenon and found that high touches are not negatively correlated with a decline in performance the next year. They also found that the mark including the playoffs is 390 carries. The 390 number explained the decline of a number of backs who had a high number of carries but did not pass 370 carries in the regular season.

I don't think the LT comparison works for LJ, and I'm not sure about LT himself if he passes 390 in the playuoffs. LT was in just his second season when he exceeded 370, and today LT and LJ are both closer to Shaun Alexander's age last year than to their age in their rookie year. Also the running style of LT and LJ are very different, with LT having some power but relying on finding the hole or cutback lane along with agility and break-away speed, while LJ's strength is his ability to break tackles and explode through the hole.

by NF (not verified) :: Wed, 12/27/2006 - 3:21am

Aaron, did you ever check whether excluding games where starters are rested improved playoff predictivity? In PFP2006 you said there was no obvious way to improve predictive accuracy while excluding them when predicting the next season, but said you hadn't checked if excluding them improved accuracy for predicting playoff games.

by Andrew (not verified) :: Wed, 12/27/2006 - 4:00am

Chris #24:

In fact, what DVOA “knew� prior to the Saints-Eagles game was that the Eagles at 45.5% would stomp the Saints at 11.1% (not including DAVE. DAVE makes it even more embarrassing). Oops

DVOA is not predictive with regard to special teamers running into their own punt returner and causing a muffed punt in scoring range for the opposition or rookies deciding to be the real 12th man on critical 3rd downs throwing a team out of field goal range with a sack. 2 plays equalling 10 points in that game.

by Malene, cph (not verified) :: Wed, 12/27/2006 - 4:44am

regarding LT and the 370 mark:

click link in my name for the Fox blog post on how a lot of LTs carries end without an actual tackle (ie out of bounds or TD).

Also, on why no. of catches doesn't mean a lot to performance decline: I'm guessing all pass plays are pretty much crap if you're the RB. I suspect picking up the blitz from DE's brings the hurt at least as much as being tackled by a CB after making a catch.

by Richard (not verified) :: Wed, 12/27/2006 - 6:12am

48: I forgot about that post. Neither of those commenters I addressed responded to my questions.

by Crushinator (not verified) :: Wed, 12/27/2006 - 6:22am

One comment, if it hasn't been mentioned.

Intangibles like what, the power to will Matt Bryant to miss a 62-yard field goal?

I've heard this a lot lately. While McNabb is clearly the better QB, the only reason the Bucs were able to win a game on a 62 yard field goal is because McNabb threw two Pick-6s in that game. Might not be the best example of a game to show how McNabb is clearly better than Garcia.

by paytonrules (not verified) :: Wed, 12/27/2006 - 8:43am

Is the ST ranking on FoxSports.com weighted? Because the bears are listed as 2nd not 1st anymore there.

by Teximu (not verified) :: Wed, 12/27/2006 - 9:43am


Yeah, but the only reason that the Bucs had to kick a 62-yard field goal to win the game is that McNabb led the team back from their deficit (that he was mainly responsible for creating, yes) in the second half.

by Charles the Philly Homer (not verified) :: Wed, 12/27/2006 - 11:14am


Chris's rantings are welcome. They correlate powerfully (1.0) to the Saints losing. I certainly hope he gets back to his consistently unenlightening commentary in the playoffs just before the Saints play the Eagles. You'd think a Saints fan would respect a gris-gris when he sees one.

by Rich Conley (not verified) :: Wed, 12/27/2006 - 11:48am



While we all know that McNabb is the better QB, isnt it possible that the system Reid is running under Garcia is a better one? Theyre certainly running more, and maybe the team is better that way. McNabb is great, but he does make a lot of mistakes, and maybe its not the best idea to have your QB throw 40 times every game, when he makes mistakes.

Maybe the eagles are better because Reid doesnt feel Garcia can throw 40+ times a game, and thats a good thing....

by David (not verified) :: Wed, 12/27/2006 - 12:11pm

That's exactly what I've been thinking. McNabb's the better QB, but the play-calling has been dramatically better with Garcia in. So, if McNabb returns hale and hearty next year and the run/pass balance goes back to 66/33, something's got to change.

by Charles the Philly Homer (not verified) :: Wed, 12/27/2006 - 12:15pm


The playcalling has changed significantly since McNabb went down. I'll characterize it as a "nothing to lose" style in which they (subjectively seem to) blitz more often late with leads and "overuse" Brian Westbrook by running him more often and having him return punts.

Also, I hate to use the word "intensity" as a determinant characteristic of a team, but they really do seem to be playing with more of "it." Fewer dropped passes, better tackling, etc. are the subtle things that have improved with more "intensity."

by Shannon (not verified) :: Wed, 12/27/2006 - 12:18pm

One of the big numbers that always gets thrown around is the number of player-days lost to injury (or alternatively, the number of starters a team plays in a year.

The Patriots are consistently very high in these numnbers, especially among 'good' or playoff teams.

I wonder if this isn't the result of a conscious strategy of building depth and experience, and if the Pats don't hold starters out of games when other teams might play them.

See link for a Boston Globe article on same.

by SJM (not verified) :: Wed, 12/27/2006 - 12:34pm


The high number of injuries may be due to the fact that the Pats seem to employ a lot of injury-prone players, but only on defense. Their offense is usually fine. (And they definitely put an emphasis on depth, as you say.)

by Bob Cook (not verified) :: Wed, 12/27/2006 - 12:54pm

I've already asked Aaron this, but maybe someone else can answer this question -- is it possible that Indianapolis is going to set a record for fewest possessions in a year?

Against Houston, Indianapolis had seven possessions -- one of which was a one-play kneeldown at the end of the first half.

In their game, Chicago and Detroit each had eight possessions -- in the first half.

It would be heartening if the Colts had limited possessions because its own offense was grinding up the clock, but stuff like a nine-minute Houston drive tells me otherwise. Not counting the kneeldown, the Colts scored on four of six possessions. One of those possessions ended with a fumble. So they only ground to a halt once.

With that kind of pressure to score, I would say Manning is having the best year of his career. Figure the last time the Colts defense was so bad, and Manning didn't have James, the Colts were 6-10 (in 2001), which is what they would be if Manning wasn't able to pull rabbits out of his helmet.

by Tally (not verified) :: Wed, 12/27/2006 - 12:57pm

LT did have a down season two years ago, after having gone for 370+ carries. He wasn't exempt. I think that the correlation holds better for rushes than for receptions. Rushers are tackled by multiple 300 lb. linemen. Receivers are tackled by 200-250 lb. DBs.

I wonder if age plays a role in it, as well. When ED went 370+ carries, he was pretty early into his career. It makes sense that younger backs can recover more easily from such a workload.

by James (not verified) :: Wed, 12/27/2006 - 1:09pm

Re: LT and carries/touches.

Also remember that LT gets exactly zero touches at all in the preseason, and now has been very lightly used in weekly practices. This would greatly reduce the amount of wear and tear a running back receives.

by richabbs (not verified) :: Wed, 12/27/2006 - 1:14pm

#42 & 39 True, I’m combining rushes and receptions here, but unless someone knows any good reason why rushes would be any more punishing than receptions, I don’t see what’s wrong with that.

One "good reason" why carrying the ball is more punishing than receiving it may be because when a back carries the ball, he is likely to be hit by linemen and inside linebackers weighing around 250-300 lbs. If he catches the ball out of the backfield, he is likely to be hit by outside linebackers or defensive backs weighing around 195-240 lbs.

by richabbs (not verified) :: Wed, 12/27/2006 - 1:18pm

#59 Tally, I didn't see your post when I put mine up. I guess great minds do think alike.

by Teximu (not verified) :: Wed, 12/27/2006 - 1:32pm


Oh, there's no doubt about it; yes, the system is being run very differently with Garcia as opposed to McNabb, and yes, it's much more reliable and consistently productive now, with a much bigger emphasis (read: any emphasis at all) on the running game and shorter, safer passes.

During the losing streak in the middle of the season, one of the really aggravating things was that the coaches (i.e. Andy Reid) kept the playcalling so pass-heavy, even early in games and against teams whose most obvious weaknesses were run defense (Saints) and when receivers were not getting open and/or dropping balls (Saints, Bucs, Jaguars, Titans, and on and on) and when the weather conditions discouraged passing (Jaguars), and when the defense was obviously struggling (take your pick)... well, you get the point.

I think, though, that if this current play-calling style were the norm with McNabb at quarterback, the Eagles could be even more productive. They could keep the committment to running (because that offensive line is just killing people), plus McNabb's arm strength would allow for some real downfield throws off play-action, which naturally becomes more effective with an actual running game.

by Sid (not verified) :: Wed, 12/27/2006 - 2:08pm

too bad I bought the books when they came out, heh. Maybe I should've waited until now to buy the 2006 copy, because I hardly got the chance to read it, what with the craziness of this past semester.

by Rich Conley (not verified) :: Wed, 12/27/2006 - 2:09pm

"I think, though, that if this current play-calling style were the norm with McNabb at quarterback, the Eagles could be even more productive. They could keep the committment to running (because that offensive line is just killing people), plus McNabb’s arm strength would allow for some real downfield throws off play-action, which naturally becomes more effective with an actual running game."

As do I, but the question is, when McNabb comes back, will Reid stay with this style? Personally, I dont think so.... and I think that makes Garcia the better quarterback for this system, because Reid knows hes limited.

by Bobman (not verified) :: Wed, 12/27/2006 - 2:18pm

Is it just me, or are the above tables not in synch with the Fox site and commentary?

Must be the eggnog talking.

Regarding the commentary, Aaron, are you kidding me about Indy/NE in the AFCCG? Of course that will happen, and this week determines if it's in NE again or Indy. Indy's achilles heel is, um, something to do with the football, guys running, tackling; it's all very nebulous and too hard to explain to laymen. But their worst enemies in the playoffs would be in approximate order Jac (if they make it), SD, NE (if Maroney is 100%), maybe Den, and Balt. Jamal Lewis doesn't really scare anybody these days and with Steve McNair, the game plan will be just stupid enough to pass a few times instead of run 50 times. McNair will see a look he likes, and audible into a clock-killing, drive-killing incomplete as he breaks his entire rib cage. (But returns for the next series in a full body cast, making the Indy DBs step six inches closer to the LOS).

But I get ahead of myself. Indy's first game will be Den or NYJ. The law of RPS gives them a win over Den and as of now the opponent will be NYJ: Leon Washington is probably not gonna ring up 150 yards in Indy. A special teams freak event may kill Indy, but they'll likely beat the Jets. They had poor luck in the Meadowlands a couple months ago and pulled it out--twice. NE plays the other team and may have to break RPS laws to top Denver, but they're likely to do so, especially at home. Same if they play the Manginis. Putting NE in SD and Indy in Bal for the divisional round.

NE plays the run well and Philip Rivers has been stumbling of late. NE win. Close and low-scoring. Ravens' D is monstrous, but it was two years ago when Manning was very patient on a Monday night. It was about 7-7 until he broke it open late and they won 20-10. Jamal Lewis is somewhat broken down by now. I would not go so far as to call Indy a favorite in Balt (especially if you are talking to the Unitas family), but I don't think a win would surprise anybody. Probably in the 24-21 range.

Leaving us with the most unlikely playoff game in a while: NE in Indy for the AFC.

Let the hype machinery begin churning. Dueling credit card commercials, in which the NE O-line mangles the supermarket crowd of extras from Manning's Visa spots. Addai and Maroney meet at midfield to kick a Reggie Bush doll in the nads. Sauerbrun bench presses an F-150 with a syringe sticking out of his left buttock while Adam Vinatieri chants to himself, "It's only a choke if you miss in the playoffs. It's only a choke if you miss in the playoffs...." Belichick puts the entire 53-man roster on IR, chess-master Polian does the same (including himself and OC Tom Moore), and Dungy reminds the home crowd, as well as the assembled media from 167 nations and 43 planets that, "It's just another game."

Other notable events: The home crowd cheering confuses the Pats for the first half, until they realize they are not playing Indy at home, and the "Thread That Must Not be Named" makes the Wordpress servers explode before the game is two minutes old. (Which I would consider a blessing, but that's another issue....)

by Pat (not verified) :: Wed, 12/27/2006 - 2:20pm

As do I, but the question is, when McNabb comes back, will Reid stay with this style?


Was Reid cautious in the beginning of the year? Sure. But he also had a fleet of new and young players, and a bunch of players coming back from injury. By now, he knows. I don't think it's surprising that they're leaning on Westbrook more. Westbrook missed a game earlier in the year, remember? Coming off of a season-ending injury last year, and injured this year, I'm not surprised they didn't use him heavily.

The exact same could be said of every skill player in Philly.

by David (not verified) :: Wed, 12/27/2006 - 2:29pm

Bobman: Houston just rode Ron Freakin' Dayne to a win over Indy. You don't think Denver can do the same with the Bell twins?

by Led (not verified) :: Wed, 12/27/2006 - 2:31pm

"They had poor luck in the Meadowlands a couple months ago and pulled it out–twice."

I seem to recall a dodgy PI call on 2nd and forever that gave Indy the ball inside the 5 and set up their second to last touchdown. Huge play. So Indy had its share of good luck at the Meadowlands as well.

by Charles the Philly Homer (not verified) :: Wed, 12/27/2006 - 3:21pm


That eggnog is doing an awful lot of talking. Are you sure it wasn't kool-aid?

by karl (not verified) :: Wed, 12/27/2006 - 3:40pm

About the Titans comment - they're run is against better teams than the Dolphins played last year, no? Jax, Philly, NYG, Indy. Even the late season win at Buffalo is somewhat impressive. That's not to say they're a great team, just as you pointed out: these were all close wins.

But the point I'm trying to make is that they aren't hanging out with the NFC West and Oakland any more, which they definitely were at the beginning of the year under Kerry Collins' direction. The jump from the level of play they were at then to what they're doing now is nothing short of miraculous, especially considering that VY is a rookie QB - he should be losing them games, not pulling out miracles in the clutch.

(I understand they've won games where he's been a non-factor for 58 minutes, but it's about the wins and he puts them over the top for those 2 minutes. I love watching this guy step up to the pressure). Once VY gets solid enough to produce quality football for the first 80 yards, combined with his nose for closing in the red zone and tendency to step up when his team truly needs it, the guy will be unstoppable. Isn't it great to see what good coaching can do with great talent?

Now if only Arthur Blank would bring back Dan Reeves...

by jonnyblazin (not verified) :: Wed, 12/27/2006 - 4:03pm

while it might be a bit silly to be looking ahead to 2nd round playoff action, I must admit I've being doing it too. And I think Baltimore matches up very well with Indy. Remember how the Colts struggled against the Steeler pass rush? The Ravens entire front 7 is freakishly athletic and they run blitzes out of many different formations, I can easily picture the Colts O-lineman being confused and overwhelmed. On the other side of the ball, I think the Ravens O-line should be able to push around the Colts front 4 with ease, so Jamal Lewis (who's been getting better as the year goes on) should be able to get enough to win.

by Bobman (not verified) :: Wed, 12/27/2006 - 4:10pm


I think it comes down to running style--the Bell (Tatum) who is the quick, end-around, cutback guy was marginal against Indy (and I think he hurt his toe vs them) whereas the battering-ram type (Mike) was the Bell who did the real damage late int he game. Denver's OL is like Indy's DL--smaller, nimble, etc. A matchup I like better than a wall of 350 lbers against Indy's 270 LB DL average. Same reason Indy did okay against Portis--if you try to outrun them, you play to their speed (you could almost call it a strength. Almost), but if you fire up the middle, you hit them where they are weakest.

And in the Indy/Denver game earlier this year, it was the passing game (with a couple huge, rather unusual but beautiful plays) that really hurt Indy early--Plummer, who had a horrid rating coming in, just torched Indy early. Some crazy rollout to the right, the entire D converging on him, then a cross-field pass to an uncovered WR.

69/Led, The "Luck" I referred to included penalties--Indy had twice as many as NY, which is very un-Indy-like, but mainly I was focused on the KO return for a TD late. (Of course one could say it was good luck for indy that NY scored so fast....) The Jets had all 3 turnovers, which undermines my luck argument more than the PI penalty--I think it's generally considered, at least here, that fumble recvoveries are close to pure luck whereas PI could easily be the result of covering a pro-bowl WR, as well as a bad call (un-luck), a slip on a muddy field (un-luck), etc.

But look at the rushing games back then: Indy had 159 and NY had 84. Indy's ground game has gotten better, if anything, and while their rush D has gotten worse statistically, this was pre-McFarland's arrival, so it's probably similar. And as an Indy fan, it's all about avoiding potentially dominating run games, as long as you can.

My real fear is that Mangini and crew have learned from the past game, whereas the Colts' approach has generally been, "We do what we do, so the key is execution." Um, sometimes it's adaptation, guys.... Ah well, good luck.

by Moses Taylor (not verified) :: Wed, 12/27/2006 - 4:17pm

From the commentary on FOX on the Lions:
Every week, another ex-Lion has a nice game against his old team. This week, it was Chicago defensive tackle Alfonso Boone. If you don't know who Alfonso Boone is, that's because he's never done anything worth mentioning before because he's never played against the Lions before.

Huh? Boone has never played on the Lions. He's been on the Bears his whole career, and thus has been on the team for their last 12 matchups. As a matter of fact, I don't think there's any ex-Lions on the Bears defense (I'm not as sure about offense).

by Bobman (not verified) :: Wed, 12/27/2006 - 4:28pm

I have to look forward because if I look back I get nauseous.

Jamal Lewis has one 100+ yard game in the past 8--vs Cleveland, and only two all year. His YPCs since their week 7 bye look like this: 3.5, 3.0, 2.8, 4.1, 3.9, 3.6, 3.4, 5.0, 3.2. I have not gone over his DVOA weekly, but I think if he is improving, he's benefitting from the more stable McNair passing game, rather than what was going on early in the season. And the more the opponent passes, the more I like it! If Boller plays, I bet Balt runs more and Indy has a worse chance.

Indy, for whatever reason, seems to have no major problems with blitzes from a 4-3 D. Isn't that Balt's base D? Now freakishly athletic secondaries covering their WRs 1 on 1 may make that blitz pickup less successful..... And freakishly athletic anybodies are problems for anyone else. Clearly Balt's performances in their last 8 games would make them a major favorite over Indy's performances in their final 8. DVOA and your average blind/deaf/mute would be able to confirm that.

But matching strength against strength, I like Indy's chances. This year Manning is has a 129.6 rating vs the blitz, vs 100.9 overall. And we all know (cough, gag) how important QB rating is to winning games. Like the last Houston one. (Gasp....)

And if I knew why Cin's competent OL could not push Indy around, well, I'd bottle it and make lots of money selling it in Indy. Maybe it's prime-time games... they're 10 for their last 10. Ergo, so long as playoffs are at night--and we all know the SB is--Indy's a shoe-in.

Yes folks, I am kidding.

by Bobman (not verified) :: Wed, 12/27/2006 - 4:29pm

jonnyblazin, sorry for the extra "h".

by Gordon (not verified) :: Wed, 12/27/2006 - 4:51pm

#67 and related posters above:

My reaction to the play-calling change with Garcia in was similar: ie, new play-calling good, McNabb still better quarterback than Garcia, McNabb + new play-calling = theoretically awesome.

I hope we actually get to see this concept in action next year. Pat, you are scarily optimistic for an Eagles fan. I am terrified that when McNabb returns, the Eagles will start throwing more than ever before. But I also think the Eagles win mostly because I bet against them every week, so what do I know?

After this week's games, I'd be interested in seeing a "what the hell happened" analysis on FO. That is, most teams in the league didn't move around too much in the rankings from last year to this year, both in terms of wins and DVOA. A few, however, made large moves, either up or down. Most of these I understand: Baltimore, for instance, moved up because they found an offense and got even better on defense. Philly moved up because they avoided injuries, mostly. But there are a few things I don't fully understand: Tampa Bay's huge decline overall, Oakland's apparent offensive decline (I say "apparent" because I didn't know they had an offense last year) offset by a huge defensive improvement (the Raiders have a top 10 defense, apparently), Denver's overall decline, and San Francisco's continued ownership of the bottom, despite halfway decent play for at least some of the season. I'd like to see the Outsiders' take on some of these things.

Standard disclaimer: All of this is based on hasty comparison of last year's final DVOA rankings (see link below) and of today's new DVOA rankings, and is in no way intended to serve as accurate analysis of anything.

by jonnyblazin (not verified) :: Wed, 12/27/2006 - 6:07pm

I think Baltimore is technically a 4-3, simply because Ngata and Gregg are both considered DTs and they both start. But sometimes Suggs plays in a 3 point stance, sometimes he stands. He got elected to the probowl as a LB even though I think the majority of games he was a DE. The bottom line is that the Ravens have 3 edge players who can play on both the rush the passer and cover very well (Suggs, Thomas, and Scott) and throughout the game generally play anything from a 2-5 to a 4-3, so I think the whole 'multiple fronts' aspect of what they do is more crucial in the analysis than the alledged base 4-3.
That being said, why Indy might have something on Baltimore is because they have 2 legitimate deep threats, and at least for part of the season the Ravens struggled when their blitzes were picked up and QBs heaved the ball deep down the field with success because the CBs and safeties were playing ultra-aggressive on the line. I think their secondary (especially Reed) have been playing better recently, though.

by Waverly (not verified) :: Wed, 12/27/2006 - 6:13pm

What's the highest DVOA for a team that didn't make the playoffs? (and variations: best offensive/defensive/special-teams DVOA)

by Oh, Mathematics (not verified) :: Wed, 12/27/2006 - 6:17pm

I have to say that the Redskins commentary really made my day. That Daniel Snyder really either has Bad Brains or no brain at all, huh?
On a somewhat related note, all of this talk of McNabb being Blacklisted in Philly is making me sick.

...and that is why /I/ am not a sports columnist.

by Mr Shush (not verified) :: Wed, 12/27/2006 - 6:41pm

Kibbles (#39) - I hear what you're saying, and of course the Lions and Raiders will turn it around eventually. But right now, I think it's fair to say that both teams are still being run by the person primarily responsible for their ongoing suckitude, in Millen and Davis. The 2004 Chargers, on the other hand, were only one year into the tenure as general manager of one A.J. Smith, who I think most would agree has since proven a fairly respectable evaluator of talent. Eli had no way of knowing that Smith would be good, or that the Chargers would turn things around and become a contender, but nor was he in a position, as the first two players taken this year will be, to know that the man responsible for selecting him was an incomparable cretin hell-bent on destroying his team. Detroit is a better situation than Oakland, because it seems possible that when they suck again next year Millen will be fired, whereas Al Davis shows no signs of dying, and even fewer of selling up or relinquishing any degree of control.

by chris clark (not verified) :: Wed, 12/27/2006 - 6:44pm

:32 I'd really like to see a breakdown of "luck" and how it affects scores, because I agree with Branch Richey that some luck is the residue of good design. Some things we ascribe to luck are really "intangibles" or something else that aren't present in the play-by-play that DVOA is based on, but may be something that could be factored in. However, I'm not sure how to measure it. Yet, I still think there was a reason that PHI underplayed their DVOA for a good bit of the year and why DEN is 1 win from the playoffs despite being a mediocre team. And, no, I'm not satisfied with a "luck" explanation. That's a cop-out for not getting the facts. And, one of the good (best) things about FO is that they don't stop at pat answers, but instead dig deeper.

For example, I like the idea of comparing JAC home/away DVOAs and seeing how much of their variance that causes. I love the charts which plot DVOA over time. As I recall, Aaron used re-running the DVOA stats to smooth out some of the swings and better approximate the real values. I also loved the guest article about long-term team metrics, despite everyone else seemingly getting caught up in the "cyclic" part of the prediction. I intrigued with the idea of calculating DVOAs with and with certain key players or comparing DVOAs to DPARs, e.g. how much does Brady's DPAR (going up or down) affect the PAT's DVOA. Somewhere, in that melange, some of the "luck" factors become measurable and tangible, even if they aren't "predictable".

I first fell in love with football and football stats in the late 1960s early 70s. It was something that gave me hope as a young Broncos fan that Otis Armstrong's running and Rick Upchurch's punt returns might someday help them beat the Raiders and get to the playoffs and Superbowl.

I'd like more of the same from stats now. Is there something about how Denver played against Cincinatti that helped cause Cincinatti to leave points on the playing field? Is that some of the same "stuff" that has caused previous DEN/NE games to be close?

And, at some level, I'd like to know if it is measurable? Can we say to a certain degree that some team has 5 points worth of luck or +10% luck? Does that luck fluctuate randomly or does it ebb and flow? Is there different home v. away luck? All questions I'd like answers to.

by are-tee (not verified) :: Wed, 12/27/2006 - 7:36pm

"There's no question Mangini has done an awesome coaching job with a team that we thought would be completely out of it by November. But the fact is that they just aren't any better than all those 8-7 AFC teams that are now a game below them. They've just had better luck in close contests."

I've watched every play of every Jet game this year. Can someone please tell me which close contests they've been lucky to win, and what the element of luck was?

by pcs (not verified) :: Wed, 12/27/2006 - 7:50pm

#74: Boone was drafted by the Lions with the second-to-last pick in 2000. Don't know if that technically makes him an "ex-Lion," but it does indicate they had a look at him and passed.

by Jin (not verified) :: Wed, 12/27/2006 - 8:07pm

Can estimated wins be used as a barometer for coaching? Because as a Viking fan I haven't been impressed by Childress and I see they are under-performing while the Saints and Jets (esp. the Jets) have more wins than estimated, and both of those teams got first year head coaches.

by Jason Mulgrew aka The Mul Dawg aka J Rocka (not verified) :: Wed, 12/27/2006 - 10:11pm

The Eags are 4th in DVOA. I knew it! I knew they were good! I told you they would win the NFC East even after they lost Donovan McNabb. I wrote that somewhere on this website.

I also told you in July that Alex Smith would turn it around and have a fine season.

2 for 2.

by Jason Mulgrew aka The Mul Dawg aka J Rocka (not verified) :: Wed, 12/27/2006 - 10:13pm

read in Borat voice

I like the Eags. They nice. Very nice. Play the better football than most of teams. 4th from top in DVOA. High five!

by Bill (not verified) :: Wed, 12/27/2006 - 11:13pm

"The Steelers played the hardest defensive schedule in the league this year."

"NEXT: at CIN"


by Led (not verified) :: Wed, 12/27/2006 - 11:27pm

I haven't done the calculation, but it seems like the average number of actual wins exceeds the average expected wins by a significant amount. What does that mean (if anything)?

by The Other Tom G. (not verified) :: Wed, 12/27/2006 - 11:41pm

Re: 79

Not sure if it's the best as I didn't look any further back but KC missed the playoffs last year with a DVOA of about 27%.

by Rick (not verified) :: Thu, 12/28/2006 - 12:59am

McNabb makes alot of mistakes? I'll give you that, but consider this, when you're asked to throw the ball 66% of the time, you have more opportunity to make those mistakes.

Were this current run-happy offense of the Eagles in existence 2 years ago, they'd have had a better shot at winning the Super Bowl, McNabb may not have gotten injured in Atlanta last year, and the Owens haters (of which I am one) would possibly have less to say.

McNabb is a better quarterback because he has alot more weapons at his disposal than Garcia. But the system won't change when he returns for one reason - his mobility will be hampered by the ACL surgery. This is one of his key weapons, one which allows broken plays to go for either terrible errors or big plays. It also creates a situation where running the ball isn't as important, because it can get run anyway if there is no one to pass to.

Nope, the injury says this system stays.
McNabb returns to be a more effective starter and Garcia stays because he's proven himself again and wants to teach.

by Ricky (not verified) :: Thu, 12/28/2006 - 1:23am

FINALLY! Ur giving my Ravens some respect. WERE GOIN ALL THE WAY!!!!!!!!!!!!

by Jason Mulgrew aka The Mul Dawg aka J Rocka (not verified) :: Thu, 12/28/2006 - 3:02am

re: Ricky

The Eagles will get to the Super Bowl and clip the Ravens.

by Peter Libero (not verified) :: Thu, 12/28/2006 - 6:01am

Whether McNabb makes a lot of mistakes depends on your definition of mistake. He throws a very small number of interceptions, though in my opinion that's because he tends to throw inaccurately in ways that are uncatchable by anyone (i.e., in the ground). He tends to have a low completion percentage, particularly as he plays in an ostensibly West Coast offense. McNabb is a very effective QB, and I personally am of the opinion that there is not a single football skill at which Garcia is superior. I mean that, I really don't know if there is ONE. It's clear to me that Reid has not usually run the ball as much as he should, but injuries to his runningbacks and a lack of quality at that position in general probably contributed. I don't think this conversation matters much... the Eagles are in no way a Superbowl team, and McNabb will start next year, when the quest begins anew.

As a side note, in an argument on some thread over the summer I noted that there's about a 50% chance that the Superbowl champ will be tops in DVOA, and all but the fluky 2001 Pats were in the top 3. In other words, it's looking very good for San Diego.

by Wanker79 (not verified) :: Thu, 12/28/2006 - 11:02am

Oh, great! Like it's not already hard enough to overcome the stigma of being a Philly fan. Thanks for making my life a little harder, Mulgrew.

by Rick (not verified) :: Thu, 12/28/2006 - 11:24am

94: McNabb throws a few uncatchable balls, but only about as many as a top QB would (aside from say, Peyton). I have watched a number of games with my Philly fan, McNabb-hating cousin. I pointed out that McNabb's receivers drop more passes than those that are uncatchable (throwing out the passes McNabb HAS to get rid of). I have consistently been proven right.

His mistakes are, typically, ones that are a result of poor, last second, decision making. The result of losing calm as a situation breaks down. But, as I said, when you throw 66% of the time, you're going to get alot more of those situations than you would if you only threw 50% of the time. In fact, statistically, your chance of having them rises exponentially once you pass a certain threshold as your O line tires out. And the Philly O line is on the record as saying they prefer run blocking to pass blocking as it is less tiring (I'm willing to bet that's a consistent claim across the NFL).

I still would like to see FO do a study of a concept my brother and I have discussed many times, and I mentioned in a thread last year - does a good QB make a receiver better or does a good receiver make a QB better?
I'm a believer that it can work both ways. After all...Garcia is a great example of how a QB with little to work with (Cleveland and Detroit) does poorly, while given the right tools (Philly), can raise his game. On the other hand, who can't play with a Montana or Peyton and NOT look good (well, maybe a few can)?

Point is, McNabb's skill set, purely based on skill, would put him up in the highest of echelons in passing QBs. It's his decision making in difficult situations that prevents him from walking the stage with a Peyton or Montana.

by Starshatterer (not verified) :: Thu, 12/28/2006 - 12:00pm

The nice thing about the Dul Dawg is, you can take
read in Borat voice
...as a flag for "skip to the next post."

The long dribble of AKAs is also a good tipoff.

by Wolffe (not verified) :: Thu, 12/28/2006 - 3:49pm

PFP2006 did a fine statistical study on the 370 carries phenomenon

One man's fine statistical study is another man's data manipulation.

Lower the threshold to 369, and it turns out that Terrell Davis, Edgerrin James, Curtis Martin and Walter Payton responded to "heavy workloads" by averaging 1660 yards the following season. Lower the threshold to 368, and Earl Campbell and Emmitt Smith respond with the best seasons of their respective careers.

The magic number was chosen because that's the lowest point at which the data fit the theory. If a lower threshold would've yielded a player who blew out a knee making a tackle after an interception or lost effectiveness after a stint in federal prison instead of expanding the pool by over 25% while adding only one (arguably) negative data point (Martin '96), you can rest assured the magic number would've become 365 or something.

by Independent George (not verified) :: Thu, 12/28/2006 - 5:37pm

Is Mulgrew the new Yang?

by White Rose Duelist (not verified) :: Thu, 12/28/2006 - 5:40pm

#98 - I see your point, but wasn't the objective of the study to figure out where that threshold was?

by Rich Conley (not verified) :: Thu, 12/28/2006 - 6:02pm


But isnt it common sense that running backs with a ridiculous amount of carries will decline the next year? Just simple regression to the mean?

by Wanker79 (not verified) :: Thu, 12/28/2006 - 6:07pm

Re: 98

I always looked at the 370 carries as more of a zone where performance is more likely to take a hit; rather than a hard threshold where you're guaranteed a drop in performance.

by shannon (not verified) :: Thu, 12/28/2006 - 6:12pm

not regression to the mean... that would be, running backs with a ridiculous number of carries have an average number of carries the next year.

This theory is, backs with a ridiculous number of carries get broken down, and have generally bad seasons the next year... not average ones.

by Led (not verified) :: Fri, 12/29/2006 - 2:40pm

"This theory is, backs with a ridiculous number of carries get broken down, and have generally bad seasons the next year… not average ones."

I agree that this was the argument. Like with pitch counts in baseball, FO wants to show that overuse of a RB is harmful to the RB's career. RB's get injured so frequently, however (isn't the average career length for an NFL back t3 years?), that FO would need to show that backs with a ridiculous number of carries are MORE likely to get broken down the next season than backs with average number of carries for a feature back. Otherwise, the problem isn't OVERuse of a feature back but just USE as a feature back that is seriously hazardous to a RB's health. Merely showing precipitous decline in a small sample of RB's with 370 or more carries doesn't get you there, in my opinion, particularly given the number of RB's with 368 and 369 carries identified by Wolffe that did not show a precipitous decline the next season. I'm no statistician, but couldn't you calculate the correlation between number of carries in season 1 (given a minimum of, say, 250 carries) with total rushing yards in season 2? If there's a significant negative correlation that would suggest an incremental risk associated with higher numbers of carries.

by dbt (not verified) :: Fri, 12/29/2006 - 8:28pm

The Bears had an above-average defensive DVOA in every single game until three weeks ago. Not only have they had below-average defense for three straight weeks, but each game is worse than the previous one. They need to figure out what's wrong — hint, it isn't just Tank Johnson — and fix it before the playoffs.

The problem has been that Ian Scott and Alfonso Boone are run-stoppers, and when you rotate after that (Lovie's scheme is rotation heavy because these guys are small) you get Antonio Garay. That doesn't scare you.

Tank will be back in the playoffs. Tommie won't be.

Also, add in that Hester has been starting at corner for the last two weeks. That boy is a lot of things, but a starting NFL corner he ain't (yet), and the fact that these guys know they're playing scrubs with nothing on the line. I hope they show up Sunday night just to remember what it feels like before the playoffs start.

Also, I updated the dvoa graphs a couple days ago, click my name for updates, etc.

by The Other Tom G. (not verified) :: Fri, 12/29/2006 - 10:58pm

Re: 79 (again)

I finally got a chance to look at the complete DVOA era.

The highest DVOA (by percent) for a non-playoff team is 29.8% by Buffalo in 2004. They were ranked 4th overall and finished 9-7. This was the team that started the season 1-5, then won 8 of 9 before losing to Pittsburgh's second string in the final game when (I believe) they would have clinched a playoff spot with a win.

The highest rank for a non-playoff team is 3 by Oakland in 1999. Their overall DVOA was only 19.4% and they finished third in their division behind Seattle and KC (not a playoff team either, but ranked 4 in DVOA that year). Anyone have the story behind this one?

One other notable thing I found: In 2001, San Diego missed the playoffs with a DVOA of 6.8%, good enough for 11th in the league and that year's highest DVOA by a non-playoff team. Their final record? 5-11. I wonder what the boards here would have looked like that season...

by The Mul Dawg (not verified) :: Sat, 12/30/2006 - 2:30pm

Do you guys think Jacksonville will win on Sunday? Or is Kansas City going to win?

by The Mul Dawg (not verified) :: Sat, 12/30/2006 - 3:29pm

I get the idea there are many students posting here. They must not have computer access at home. In previous weeks the DVOA threads had plenty more posts. I figured my question would have been answered by now.
Maybe if I wrote in like Borat it would have garnered a response.

by The Mul Dawg (not verified) :: Sat, 12/30/2006 - 5:52pm

Can anybody help me with the Jacksonville-KC game?

I wish this was like one of those V-bulletin message boards. Then people would notice these posts. As it is, I guess people are not checking this thread anymore.

by The Mul Dawg (not verified) :: Sat, 12/30/2006 - 6:41pm

read in Borat voice

I like if you people tell me who win Jaguars-Chiefs game very much. It would be nice, very nice.

by The Mul Dawg (not verified) :: Sun, 12/31/2006 - 1:43am

If anyone wants to offer an opinion on that game, you don't have to now. I already submitted my picks.

by The Mul Dawg (not verified) :: Sun, 12/31/2006 - 7:47pm

I got it right!!!!!!

I took the Chiefs!!