Week 13 DVOA Ratings

Week 13 DVOA Ratings
Week 13 DVOA Ratings
Photo: USA Today Sports Images

by Aaron Schatz

Another week of NFL games, another week without many changes in the Football Outsiders DVOA ratings. Some people might consider this a bug, but we consider it a feature. A longer-term view generally works better than the way conventional wisdom on teams bounces up and down in the short term.

Of course, when you have a number of teams tightly packed together, one bad game can knock a team down a significant number of pegs. That's what happened this week to the Arizona Cardinals, who drop from sixth overall to 13th. Because teams are more spread out at the top and bottom of the league, the same game only moves Philadelphia up one spot, from fourth to third, even though right now their 48-20 win comes out as the second-best single game performance of the year according to DVOA.

Best Single-Game DVOA of 2008, as of Week 13
Team Week Opp. Score DVOA
IND 6 BAL 31-3 132.4%
PHI 13 ARI 48-20 126.2%
NYJ 10 STL 47-3 105.4%
TEN 13 DET 47-10 99.7%
MIA 3 NE 38-13 98.7%
NYG 5 SEA 44-6 92.0%
ATL 9 OAK 24-0 89.6%
NO 12 GB 51-29 88.7%
NYG 9 DAL 35-14 87.5%
TB 2 ATL 24-9 85.1%

Also dropping this week: the New York Jets, who are still the highest team in the AFC East at 14th. The four teams in that division have their stats this year remarkably skewed by the schedule, since they play six games against their equally mediocre division rivals plus four against the pathetic AFC West and four against the even more pathetic NFC West. If we rank schedules based on average DVOA of opponent -- for the whole year, not just the games played so far -- the four easiest schedules in the league belong to the four AFC East teams.

If every team in the NFL matches its current DVOA over the next four weeks, the 2008 AFC East will end up with three of the ten easiest schedules of the DVOA Era.

Easiest Schedule Based on
Average Opponent DVOA, 1995-2008
Team Year DVOA W-L
STL 1999 -14.3% 13-3
NYJ* 2008 -10.9% 8-4
SEA 2007 -10.6% 9-7
PHI 2000 -10.3% 11-5
SEA 2005 -10.1% 13-3
BUF* 2008 -10.0% 6-6
JAC 1999 -9.7% 14-2
ARI 1998 -9.6% 9-7
DAL 1998 -9.4% 10-6
MIA* 2008 -8.9% 7-5
*through Week 13

This is a big part of why there is such a colossal difference between VOA and DVOA for the AFC East teams. It also helps that the Dolphins and Jets have had great luck recovering fumbles on offense. Miami has recovered 8 of 11, while the Jets have recovered 11 of 17. Based on VOA -- no opponent adjustments, no penalty for fumbles recovered by the offense, and no weather adjustments on special teams -- the Dolphins and Jets are both top 10 teams. DVOA puts them 18th and 14th.

* * * * *

A quick note on Quick Reads: There was some sort of strange quirk at the NFL media website that kept the Broncos-Jets gamebook from being posted properly Sunday night, and so I missed it when compiling Quick Reads. Sorry about that. It actually would have made a difference, although perhaps not where people expected. Jay Cutler would have come out with the fifth best quarterback performance of the day (140 DYAR) and Brett Favre with the fifth worst (-42 DYAR). More interesting are the running backs and wide receivers. Peyton Hillis, the latest "who the hell is that" Denver starter, was the number-two running back of the week with 59 DYAR. On the other side of the coin, Laveranues Coles actually came out as the worst wide receiver of the week, not Anquan Boldin as reported. Coles caught just two passes out of seven and gained a whopping two yards. The Jets threw to him once before the fourth quarter -- and it was an interception.

We missed this somehow on Monday, but Brian Westbrook's 102 DYAR against Arizona actually ends up ranking as the best running back game of the season so far. Lest all of you believe we're only in this to say mean things about Michael Turner, I can tell you that Turner's Week 1 explosion against Detroit is the second best game -- even after we apply the very, very strong opponent adjustment for playing Detroit.

* * * * *

A couple of weeks ago, we posted something in Extra Points about the possibility of holding get-togethers for our readers to hang and watch football together -- I called them "beer feeds," as opposed to Baseball Prospectus' "pizza feeds." Well, we've planned three of them, so here are some basic details. We'll have more specifics soon, like directions to each place. Anyone can come, there's no fee except for buying stuff from the bar, but if you are coming please try to RSVP by e-mailing Bill Barnwell at mailbag- at-footballoutsiders.com. Also, if you are coming to the Boston or New York events, let us know which games you would like to watch on the closest, biggest televisions.

Sunday, December 14: Boston
Sports Depot in Allston, 1pm
featuring Aaron Schatz and Bill Barnwell

Thursday, December 18: Philadelphia
Chickie and Pete's near Lincoln Financial Field, 8pm
featuring Aaron Schatz and Mike Tanier

(Note: I'm combining this with our annual trip to NFL Films, which is why we're doing Thursday night rather than Sunday. The game is Colts-Jaguars, but hey, you're coming to see us, right? Right?)

Sunday, December 21: New York City
Coppersmith's (793 9th Ave. at 53rd St.), 1pm
featuring Bill Barnwell and Mike Tanier, and probably Sean McCormick and maybe Russell Levine or Bill Moore and a cast of thousands...

* * * * *

All stats pages are now updated, including playoff odds and the ALY/ASR pages I wasn't able to update after Week 12. One last housekeeping note: We've switched Dallas Clark back from wide receiver to tight end. He's playing tight end more this season, and that also helps us to keep the new "defense vs. receivers" table more accurate when it comes to helping fantasy football players. Good luck in your playoffs, everyone.

* * * * *

These are the Football Outsiders team efficiency ratings through 13 weeks of 2008, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average.(Explained further here.)

OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. These ratings also include opponent adjustments. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. WEIGHTED DVOA is adjusted so that earlier games in the season become gradually less important. It better reflects how the team is playing right now.

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

To save people some time, please use the following format for all complaints:

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

1 NYG 39.7% 1 39.0% 1 11-1 25.5% 1 -10.2% 7 3.9% 6
2 BAL 32.1% 2 30.6% 3 8-4 6.3% 17 -28.4% 1 -2.6% 25
3 PHI 30.7% 4 29.3% 4 6-5-1 11.2% 12 -17.8% 6 1.7% 10
4 TEN 30.2% 3 31.6% 2 11-1 9.5% 14 -19.3% 4 1.4% 13
5 PIT 22.2% 5 21.2% 5 9-3 0.3% 20 -23.1% 2 -1.2% 23
6 TB 19.9% 7 21.0% 6 9-3 -1.6% 22 -20.0% 3 1.5% 12
7 CAR 16.2% 8 16.1% 7 9-3 10.2% 13 -4.4% 8 1.6% 11
8 GB 13.7% 9 14.0% 8 5-7 12.0% 10 -0.7% 11 1.0% 15
9 ATL 12.9% 13 13.8% 9 8-4 18.4% 4 9.0% 21 3.5% 7
10 NO 11.6% 11 12.5% 11 6-6 20.8% 3 9.8% 22 0.6% 17
11 IND 10.0% 12 13.6% 10 8-4 15.8% 6 4.7% 14 -1.1% 22
12 WAS 9.2% 14 6.6% 13 7-5 12.5% 9 -1.2% 10 -4.5% 29
13 ARI 7.5% 6 7.2% 12 7-5 16.3% 5 5.7% 16 -3.0% 26
14 NYJ 6.1% 10 6.1% 14 8-4 4.6% 19 3.6% 12 5.1% 4
15 MIN 5.8% 19 5.6% 15 7-5 -4.0% 24 -19.1% 5 -9.4% 32
16 CHI 2.5% 15 -1.5% 17 6-6 -0.6% 21 -3.0% 9 0.2% 18
17 DAL 1.7% 20 -2.7% 19 8-4 11.3% 11 5.3% 15 -4.3% 28
18 MIA 1.3% 16 0.9% 16 7-5 15.6% 7 6.5% 17 -7.8% 31
19 SD 0.1% 17 -1.8% 18 4-8 13.5% 8 14.3% 26 0.9% 16
20 DEN -1.8% 21 -5.6% 22 7-5 21.6% 2 19.3% 29 -4.1% 27
21 NE -4.8% 18 -4.5% 21 7-5 7.4% 16 13.5% 24 1.3% 14
22 HOU -7.5% 24 -3.7% 20 5-7 8.2% 15 18.7% 28 3.0% 8
23 JAC -8.1% 22 -9.6% 24 4-8 6.1% 18 14.0% 25 -0.2% 19
24 CLE -10.4% 25 -6.9% 23 4-8 -7.7% 25 7.8% 18 5.1% 3
25 BUF -11.4% 23 -14.6% 25 6-6 -3.7% 23 13.2% 23 5.4% 2
26 SF -20.0% 26 -21.9% 27 4-8 -18.1% 29 7.9% 19 6.0% 1
27 OAK -20.5% 27 -18.5% 26 3-9 -20.8% 31 4.5% 13 4.9% 5
28 SEA -27.2% 28 -22.8% 28 2-10 -13.8% 27 16.1% 27 2.7% 9
29 CIN -28.5% 29 -28.0% 29 1-10-1 -17.6% 28 8.6% 20 -2.3% 24
30 KC -33.0% 30 -29.8% 30 2-10 -7.8% 26 19.7% 30 -5.4% 30
31 STL -51.2% 32 -48.3% 31 2-10 -25.4% 32 25.4% 31 -0.4% 21
32 DET -52.0% 31 -50.5% 32 0-12 -19.8% 30 31.9% 32 -0.3% 20

  • NON-ADJUSTED TOTAL DVOA gives performance without adjustments for schedule strength, fumble recovery luck, and weather/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.
  • PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • VARIANCE measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from most consistent (#1, lowest variance) to least consistent (#32, highest variance).

1 NYG 39.7% 11-1 41.3% 9.6 1 -1.9% 23 18.1% 5 17.3% 20
2 BAL 32.1% 8-4 31.4% 9.0 2 2.2% 10 8.3% 10 19.2% 23
3 PHI 30.7% 6-5-1 29.2% 8.4 4 0.1% 17 13.4% 8 19.3% 24
4 TEN 30.2% 11-1 33.6% 9.0 3 -5.2% 27 4.8% 11 8.9% 5
5 PIT 22.2% 9-3 23.0% 8.1 5 2.6% 9 17.9% 6 5.7% 1
6 TB 19.9% 9-3 22.4% 7.6 6 -2.9% 25 2.9% 14 16.4% 15
7 CAR 16.2% 9-3 16.2% 7.4 8 -1.4% 21 23.1% 2 16.8% 16
8 GB 13.7% 5-7 7.3% 7.1 10 2.9% 7 -21.7% 31 16.4% 14
9 ATL 12.9% 8-4 10.6% 6.8 12 0.3% 16 -4.6% 18 22.9% 28
10 NO 11.6% 6-6 10.7% 7.1 9 1.7% 11 -6.8% 19 12.5% 10
11 IND 10.0% 8-4 7.3% 7.4 7 5.3% 4 -19.5% 30 20.2% 25
12 WAS 9.2% 7-5 6.8% 6.8 13 1.1% 13 4.8% 12 6.7% 3
13 ARI 7.5% 7-5 8.5% 6.4 15 -1.9% 24 -25.8% 32 18.3% 22
14 NYJ 6.1% 8-4 13.5% 6.6 14 -9.0% 31 -19.1% 28 20.2% 26
15 MIN 5.8% 7-5 2.9% 6.3 17 4.1% 5 2.7% 15 11.1% 7
16 CHI 2.5% 6-6 1.4% 6.8 11 -0.8% 20 3.2% 13 10.0% 6
17 DAL 1.7% 8-4 6.0% 6.0 19 -0.6% 19 41.5% 1 27.9% 31
18 MIA 1.3% 7-5 14.9% 6.0 18 -6.4% 30 -19.4% 29 16.9% 18
19 SD 0.1% 4-8 -0.4% 6.3 16 0.7% 14 -11.8% 25 15.4% 12
20 DEN -1.8% 7-5 -1.0% 5.6 21 -3.5% 26 -9.4% 22 25.7% 29
21 NE -4.8% 7-5 -0.4% 5.7 20 -5.4% 28 -17.2% 27 21.2% 27
22 HOU -7.5% 5-7 -8.2% 4.6 25 0.3% 15 8.6% 9 12.4% 9
23 JAC -8.1% 4-8 -6.4% 5.5 22 -1.6% 22 19.4% 3 6.4% 2
24 CLE -10.4% 4-8 -17.7% 5.3 23 6.9% 2 18.2% 4 18.2% 21
25 BUF -11.4% 6-6 0.7% 4.8 24 -12.3% 32 0.2% 17 11.7% 8
26 SF -20.0% 4-8 -14.9% 4.3 26 -5.8% 29 -11.5% 24 7.0% 4
27 OAK -20.5% 3-9 -21.4% 3.9 27 -0.1% 18 2.6% 16 30.6% 32
28 SEA -27.2% 2-10 -26.6% 3.5 28 1.6% 12 -14.1% 26 17.2% 19
29 CIN -28.5% 1-10-1 -42.2% 3.1 30 14.7% 1 -8.0% 20 12.7% 11
30 KC -33.0% 2-10 -30.0% 3.1 29 2.9% 6 -9.6% 23 16.8% 17
31 STL -51.2% 2-10 -52.3% 1.5 32 2.7% 8 -8.9% 21 26.6% 30
32 DET -52.0% 0-12 -52.2% 1.8 31 6.0% 3 13.7% 7 16.3% 13

Worst DVOA Ever Watch

The Giants actually fall out of the "historical top ten" despite their win over Washington, but things just get worse and worse for Detroit. At this point, even if you adjust for the offensive environment of the league, Detroit has the worst defense of the DVOA Era and possibly the worst defense in NFL history.

2005 SF -58.3%   2008 DET 32.0%
2008 DET -52.0%   2000 ARI 27.0%
1999 CLE -51.5%   2004 NO 26.9%
2008 STL -51.2%   1996 ATL 25.8%
2004 SF -47.9%   2005 HOU 25.3%
2000 CIN -47.4%   2008 STL 25.4%
2007 SF -44.7%   2004 STL 24.9%
1998 PHI -43.8%   2000 SF 24.1%
2000 ARI -43.6%   2005 STL 22.5%
2002 ARI -42.9%   2002 ARI 22.2%


97 comments, Last at 08 Dec 2008, 1:55am

1 Re: Week 13 DVOA Ratings

Denver: 29th in variance...could be the reason why they suck it up against the Raiders and then pound the Jets...plus, 29th in D...please get D.J. and Champ back!

2 Re: Week 13 DVOA Ratings

So...the Packers lost and they move up (13.3%->13.7%). They really need some fiery leadership. They underperform so much.

At least Frost is gone. Too bad the punter market is thin right now.

3 Re: Week 13 DVOA Ratings

Seattle played good football against that soft schedule for years, and it seemed to count against them. It hits the AFC East, and it seems to be more of an asterisk to their DVOA adjustment, suggesting they're better than that. Eye of the beholder, I guess.

66 Re: Week 13 DVOA Ratings

Well the four AFC East teams are ranked 14, 18, 21, and 25, and none of them have a losing record; I'd say that the weak schedules are hitting them pretty hard. For instance, a lot of "power ranking" type lists have/had the Jets or NE in the top ten, and Miami fairly high. DVOA doesn't see them that way.

75 Re: Week 13 DVOA Ratings

I'm still of the opinion that DVOA has some issues when the schedules are skewed too far one way. It under ranks good teams with really weak schedules and over ranks poor teams with really strong schedules. Not drastically, but by enough points to make a difference. It makes the assumption that the opponent adjustment should be linear when I don't think that how things actually work.

As far as I'm concerned, for NYG, theres no difference between playing KC and Detroit, yet DVOA sees it as the same difference between playing Tenessee and Minnesota, or Pittsburgh and Denver.

91 Re: Week 13 DVOA Ratings

In reply to by Rich Conley (not verified)

That's exactly my opinion as well. Aaron's pointed out that blowouts against bad teams are more indicative of superbowl contender strength than close wins against good teams, so that alone suggests the opponent adjustment should be somewhat algorithmic.

95 Re: Week 13 DVOA Ratings

i dunno man, i think it's pretty clear that there's a difference between playing detroit and kc. kc is a run of the mill terrible team, the kind that shows up every season in the nfl. detroit is historically bad.

92 Re: Week 13 DVOA Ratings

You seem to be under the impression that I don't think the AFC East has had substantial negative opponent adjustments. All I need to do is see the difference between VOA and DVOA to see that's obviously the case.

What I'm saying is, Aaron's comment sounded like an excuse for their lower ratings, as though they're good teams punished for having a weak schedule. My subjective view of general consensuses (FO staff, FO readers) about Seattle during their good years, was they were dismissed as feasting on cupcakes even after being routinely afflicted with those kind of opponent adjustments. It's not DVOA, it's the interpretation, that I took issue with.

4 Re: Week 13 DVOA Ratings

All right, the time has finally come. Dallas versus the DVOA juggernauts. #5, #1, #2, and #3 to finish the season. Let's see what they've got!

I'm really curious as to what Dallas' DVOA would be if the 3 Brad Bollinger [sic] games were removed.

Man, ARI fell a long way.

I'd really be interested to see a TEN/PHI game.

IND is #11 but 4th best in the AFC. Man this year is so much different from last!

Looks like TB/CAR and ATL/NO should be great games.

Oh, and I almost forgot, PHI/NYG! Should be very interesting.

Man, there are really a LOT of good games this week!

Too bad DET/STL is not on the schedule.

39 Re: Week 13 DVOA Ratings

At this point, I would actually tune in to watch a Detroit St. Louis matchup. What a historic display of ineptitude that would be.

Dallas has one hell of a murderer's row coming up to close the season. And with the NFC South so hot, they need to win at least 2-3 of their final 4. That's a tall order against Pittsburgh, NYG, Ball'mer, and Philly.

I'm stuck with hoping that my Williams Wall-less Vikes can beat Detroit, then pick up at least 2 more wind at ARI, or hosting Atlanta or the Giants.

I think the Giants are a likely win, since New York will probably have already locked up it's #1 seeding, and will be resting its starters.

-formerly known as 'Tundrapat' (until my membership login gets sorted out)

49 Re: Week 13 DVOA Ratings

The Giants can lock up the #2 by the time they play Dallas, but not the #1. Even if the Giants win this week, the winner of Carolina-TB will be 2 games behind them with three to play. That said, Dallas will have a lot more to play for than the Giants for sure.

70 Re: Week 13 DVOA Ratings

I think he was referring to the Vikes-Giants week 17 game. After Coughlin made such a big deal about not resting the starters last year, I'm real curious to see how the Giants handle any meaningless games they might have this year.

5 Re: Week 13 DVOA Ratings

Dallas's future schedule DVOA (41.5%) is higher than any individual team's DVOA (39.7%). Is this right?

32 Re: Week 13 DVOA Ratings

yeah, that does look ... odd. i was going to ask if there was a home/road adjustment for strength of schedule, but they have 2 remaining home games and 2 away, so that doesn't explain it.

37 Re: Week 13 DVOA Ratings

yeh, Dallas doesn't make any sense (with average being higher than the highest single DVOA out there), nor does Pittsburgh
{remaing games against TEN,BAL,CLE,DAL(32.10%,30.20%, 1.70%,-10.40%)} .. which by my math "averages" to 13.40% Future Schedule

Didn't bother checking any other teams because I didn't know anyone else's future schedule off the top of my head. :)

Either the formula is more complex than the simple average it claims, or somebody can't do simple math.

6 Special Teams

Can any Buffalo fans explain to me the horror that was apparently the 2000 season?

Bottom 5 Special Teams of the DVOA Era:

1997 CHI: -8.0%
2002 CIN: -8.1%
1997 SEA: -8.6%
2008 MIN: -9.4% (12 games)
2000 BUF: -13.1%

9 Re: Special Teams

In reply to by Independent George

Music City Miracle Hangover?

11 Re: Special Teams

In reply to by Independent George

I'd rather not relive the Rob Johnson Years.

40 Re: Special Teams

In reply to by Independent George

Without looking through the old game logs, I'd venture to say that the Bills probably 1) provided opposing teams with consistently good starting field position (not due to turnover), and/or 2) consistently started their drives with poor field position.

At a guess.

52 Re: Special Teams

In reply to by Independent George

Didn't the Bills fire their special teams' coach after the Miracle, even though they'd had a great year and he is caught on tape before the kickoff saying "watch the trick play"? Maybe that had something to do with it.

57 Re: Special Teams

In reply to by Cliff Claven (not verified)

Yes, that exactly what happened. The fired ST coach was pretty damn good. After a dreadful year on special teams in 2000, Wade Phillips was told to fire his new special teams coach. (ST that year were so bad that Phillips referred to his punt returner as a "punt catcher," since the concern was simply preventing fumbles on fair catches. I think they gave up a crapload of long kick returns, too.) Phillips refused, and was canned. In came Gregg Williams as HC, and I think that's also when Bobby April, Pagan God of Special Teams, arrived.

8 Re: Week 13 DVOA Ratings

It's interesting to me that the Ravens defense has retained its lead on the Steelers despite the fact that the Ravens were pummeled by the Colts and the Giants, while the Steelers (ultimately lost but) held both of those top-flight offenses in check. I guess the Ravens D has had more dominant games to make up for it, whereas the Steelers have been more quietly consistent but not as spectacularly efficient. At least that's what the variance rankings suggest to me.

23 Re: Week 13 DVOA Ratings

Baltimore made the leap to #1 based on the strength of their dismantling of the Kolb-led DVOA darling Eagles.

Before that they were neck and neck with Tennessee for 2/3 behind the Steelers.

33 Re: Week 13 DVOA Ratings

If I had to make an educated guess, it's probably because the Ravens lead the league in 3rd down efficiency by a large margin. For example, the Ravens have given up only 13.8 first downs per game (Pittsburgh allows 15.8, still very good); the Ravens have only allowed 46 of 152 3rd down conversions compared to 58 of 171 for Pittsburgh; the Ravens lead in time of possession; etc.

Another important factor--although the Ravens have given up 20 TD's compared to 16 for Pittsburgh (and 19 for Tampa Bay, 20 for Tennessee), out of all those other teams the Ravens have given up 5 TD's that are not the fault of the defense, whereas those numbers for Pittsburgh, Tampa Bay and Tennessee, respectively, are 1, 2, and 0. (When I say "not the fault of the defense" I mean in the case of the Ravens, 2 kickoff return TD's, 2 fumble return TD's and 1 interception return TD.) In other words, the Ravens and Steelers defenses are tied for the league lead in allowing the fewest TD's, 15 each, so the Steelers don't really have a better scoring defense than the Ravens even though a superficial look at the stats suggests otherwise.

10 Re: Week 13 DVOA Ratings

Regarding the all-time easiest sked list, I could have sworn Indy's 2005 year would have been there--all season until they lost to SD in week 13 or so, people were saying they were paper tigers because their sked was so soft, blah blah.

I guess the critics were only half right. No, no, I take that back. If Nick Harper's wife hadn't stabbed him in the leg the night before, he'd probably have returned Bettis's fumble all the way and they would have beaten the eventual SB champs. Of course that includes Polamalu's non-football-move non-INT.

Okay, cardboard tigers in 2005.

12 Re: Week 13 DVOA Ratings

Even as a Seahawk fan I agree with you, that they'd likely have won it all, were it not for Nick Harper's wife. Which I've routinely joked about...and now someone else gets my twisted humor.

20 Re: Week 13 DVOA Ratings

In reply to by Jacob Stevens (not verified)

I've said that as well. I also noticed looking at the replay of that play multiple Pittsburgh coaches/players on the sideline are on the field and Harper cuts back away from them and gets tackled. Coincidence?

76 Re: Week 13 DVOA Ratings

In reply to by turbohappy (not verified)

I never really understood what happened on that play. It looked to me like he had a clear route to the endzone, and then turned back inside and ran directly at Roethlisburger.

13 Re: Week 13 DVOA Ratings

All the NFC South in the top10 doesn't deserve a mention, or did I somehow miss it? Pretty great football in that division this year, every game is interesting.

59 Re: NFC South

No, you're seeing it correctly--and yes, it DOES deserve a mention. It also bears mention that New Orleans, at the bottom of the division at 6-6, would be leading the NFCW, NFCN, & AFCW if they were playing in those divisions (with their schedules). It's slightly encouraging that at least we will get the last place teams in the NFCN & NFCW (Lions & Rams?) since, as I have heard, we get the NFCE & AFCE in the other non-divisional games. At worst, we will finish 7-9 by beating the Lions if we lose the other 3 games left on the sched.
Just think--what would the records (and DVOA) of the NFCS teams be if we got the AFCE schedules??!!!!

81 Re: NFC South

In reply to by Joseph

Looks like they would be leading the AFCE too :)

93 Re: NFC South

In reply to by Joseph

"Just think--what would the records (and DVOA) of the NFCS teams be if we got the AFCE schedules??!!!!"

If DVOA works as its supposed to, their DVOA would be exactly the same.

14 Re: Week 13 DVOA Ratings

Hmmm, not having watched all that many non-Broncos games this season, I can't speak with much authority on this, but I find it unimaginable that there are five teams with more variance on offense than the Broncos.

15 Re: Week 13 DVOA Ratings

The Titans play the 4th best game of the year and still fall in the rankings. Awesome. I'm also amused that they lose their #1 ranking in variance. I kind of wonder what their non-Lions DVOA is. I also believe this is the first time in a while the Titans have had a past or future schedule that ranked as above-average.

As for the Eagles, I wonder if Aaron made them angry with his "2003 Bucs-good DVOA, mediocre record team that collapsed" comments in last week's Audibles.

16 Re: Week 13 DVOA Ratings

I want to ask FO a favor:
What is San Francisco's VOA and DVOA before and after Nolan was fired? It definitely seems that they have been playing a lot better.

21 Re: Week 13 DVOA Ratings

That analysis would allow FO staff to create a new metric to measure teams by--pants-optional coaches would automatically be assumed to improve a team's DVOA by about 5.0% I'm guessing.

Which gives me some thoughts for coaching in my kid's spring flag football league. Just call me Coach Skivvies.

17 Re: Week 13 DVOA Ratings

Lions DVOA: 52.0%.
Lions Weighted DVOA: 50.5%.

Getting better every week! Look out ... umm ... look out Gators!

30 Re: Week 13 DVOA Ratings

Actually, Weighted DVOA gives more weight to recent games than does DVOA.

They're getting worse.

Somewhere, MDS is weeping into his beer.

36 Re: Week 13 DVOA Ratings

Yeah, I omitted the negative signs because I forgot.

Although maybe I was trying to make a subconscious statement about the Lions being the equivalent of last season's Patriots. Think about it ... it might just be time for an irrational Culpepper/Brady thread. I'll be there in a minute ... promise ...

18 Re: Week 13 DVOA Ratings

Wow, the generally not-too-impressive Colts suddenly have a #6 O, #14 D, and #22 ST (believe me--that's a HUGE improvement over the past), plus only 6 teams ahead of them on the Playoff Odds page likely to have more post-season success. I assume they can beat CAR and TB, plus they HAVE beaten Pitt and Balt, leaving the two one-loss monsters. I'm getting a distinct 1995/2006 vibe here.

29 Re: Week 13 DVOA Ratings

I've been getting a 2006 in reverse vibe. The schedule this year was a flip-flop of that year so it makes sense. 2006 was their lowest regular season DVOA by far of the surrounding years ('04, '05, '07 are all >30%, '06 is <20%). You just never know, it's all about matchups, just like it was that year (I'm still sure Indy would have lost to SD in SD if NE hadn't pulled out the miracle).

19 Re: Week 13 DVOA Ratings

Wow - the one place on the planet that the NFCS gets some respect. Bottom team = 10.

22 Re: Week 13 DVOA Ratings

I wonder what Dallas' numbers look like if you take out the disaster that was the Rams game.

24 Re: Week 13 DVOA Ratings

Those of a gambling persuasion may want to note that the Packers are available at 16/1 for the NFC North. The DVOA playoff odds report makes them a 3/1 shot!

78 Packers and NFC North

I noticed this seeming anomoly with the Packers too, but think this is either a statistical error or (or likely) just shows the (expected) shortcomings of any purely mathematical system. The Packers are looking very good on the tiebreaker front given their (both actual and expected) divisional record, but even ignoring the Bears, to catch the Vikings, the Packers have to make up two games (5-7 vs. 7-5) to get to the tiebreaker. This practically means that the Packers have to run the table in the last four -- by no means impossible given their schedule as well as history -- but at the same time, the Vikings need to lose 2 of 4. The Vikes play winless Detroit next week; what are the odds they lose that game (minimal - MIN's a -360 there, FWIW)? So really you're looking at the Vikings having to lose 2 of their last 3 for the Packers to win the North EVEN IF they run the table. After the hapless Lions the Vikings play (1) Arizona, who's already locked up their (pathetic) division, (2) Atlanta (figure 50/50 either way?), and (3) New York, which surely will rest their starters on the last game before the playoffs and with nothing to gain.

So even if we totally ignore Chicago (which is a game ahead of the Packers, and even if the math is correct, I think this may be one where the market has a more accurate assessment than a purely mathematical system -- regardless of its (huge) benefits -- ever could.

25 Re: Week 13 DVOA Ratings

Dallas is probably better than their DVOA suggests because of how steep the fall-off was when Romo was injured. (Not a knock on DVOA--I understand how it works, just adding a common sense observation.) It looks like we're going to find out how good they are the next four weeks. What a brutal schedule!

27 Re: Week 13 DVOA Ratings

Quick Playoff Odds question: How does the "Mean Wins" stat account for the PHI-CIN tie? Is the tie ignored, or does it count for half of a win? (i.e., is the "9.0" Mean Wins projection for Philly indicate 9-6-1, or somewhere between 8-7-1 and 9-6-1?)

28 Re: Week 13 DVOA Ratings

How can the Raiders be #6 against #1 receivers with Nnamdi Asomugha having the year he is? It can't be the league-worst coverage of #2 receivers that is doing it.

31 Re: Week 13 DVOA Ratings

that's just DVOA vs #1 receivers. look at yards per game or even attempts per game and the raiders are far and away the best. i suspect the DVOA difference is because asomugha doesn't have as many interceptions, or perhaps because nobody throws his way on 3rd or 4th down (which are more important and hence lead to bigger DVOA swings).

41 Re: Week 13 DVOA Ratings

From what I've seen of the Raiders this year (admittedly only like 3 games) they tend to just leave asomugha on one side of the field (Offensive left defensive right) and have him cover the receiver that is there. I've seen some offenses just like their best receiver up on the other side and just abuse that cornerback (See Eddie Royal week 1 who would have been their number 1 receiver with Marshall suspended.)

96 Re: Week 13 DVOA Ratings

most teams do that. people seem to think that the # 1 corner always covers the #1 wr, but it doesn't happen. i can't remember if it was here or if it was a kc joyner article, but i remember reading that the #1 cb is on the # 1 wr for around 50% of nfl plays.

it's also worth pointing out that offensive coordinators get paid hundreds of thousands of dollars to create mismatches. motion, pick/rub plays, etc. they're trying to get their best receiver matched up on a weaker db.

45 Re: Week 13 DVOA Ratings

In reply to by ChicagoRaider (not verified)

Alsop, keep in mind that a single cornerback doesn't usually spend every snap of every game facing the opposition's #1 wideout. Motion, shifting, 'picks' (Hi, Indy!) will all lead to the top CB covering a #2 or #3 wideout.

It's not a knock on Asomugha.

97 Re: Week 13 DVOA Ratings

i could have saved myself a little bit of time if i had scrolled down and read this post instead of clicking reply.

34 Re: Week 13 DVOA Ratings

A lot of people are talking about Dallas, so I thought I'd put the season splits for them on here (numbers are not going to be exactly correct, I don't have access to DVOA premium, so these come directly from the DVOA rankings articles, which obviously aren't retroactively corrected to include up-to-date opponent adjustments)

After 6 Games: 24.7% DVOA, ranked 3rd (Before Romo gets hurt)
After 9 Games: 4.1% DVOA, ranked 17th (the Brad Bollinger experience)
After 13 Games: 11.3% DVOA, ranked 11th (the Savior returns)

After 6 Games: 8.1% DVOA, ranked 21st (Before Romo gets hurt)
After 7 Games: 13.1% DVOA, ranked 24th (the Massacre at St. Louis)
After 13 Games: 5.3% DVOA, ranked 15th (Phillips realizes his job is on the line, takes over control of underachieving defense)

60 Re: Week 13 DVOA Ratings

Well, for one thing, it may be Week 13, but the Cowboys have only played 12 games. For another, why didn't you do an estimate based on the numbers?

If three of nine games lower the offensive DVOA by 20.6%, then I'll assume that 3 of 12 games will lower the offensive DVOA by 15.45%. I'm also assuming Tony Romo's disappearance for 3 weeks didn't significantly affect the special teams or the defense. The revised stat line:

Offense: 26.8% DVOA
Defense: 5.3% DVOA
Special Teams: -4.3% DVOA
Overall: 17.2% DVOA
Rank: #7

This still leaves them below all of their December opponents in overall DVOA.

63 Re: Week 13 DVOA Ratings

My bad on the week 13/game 13 thing.

The rest wasn't really serious, there's little validity to the numbers I put up and going even further with them is fairly pointless.

35 Re: Week 13 DVOA Ratings

"Anonymous" brings up an interesting point that someone can explain for me hopefully. Does the Ravens' defense get penalized more for a game like the Giants had against them (outside of variance) or does it not hurt them as badly because the Giants' run offense is so highly rated (or was at the time)? Or are the numbers constantly changing week to week? Or am I just clueless?

The Giants run offense DVOA dominance has lowered significantly, but their pass offense keeps climbing by the week.

38 Re: Week 13 DVOA Ratings

Fascinating stuff this week:
Almost the entire difference between the Giants and Ravens is in special teams play. Detroit's defense is even more inept than the Ravens's defense is apt. The Giants, Ravens, and Eagles all have about the same variance (17.3%, 19.2%, 19.3%), whereas the #4 & #5 teams, Titans and Steelers, are both models of consistency (8.9% & 5.7% = #1). Is the roughly 1 1/2 point gap between the Eagles's weighted & regular DVOA due to their performance against the Bengals and Ravens? And does Baltimore's similar DVOA gap stem mostly from the Giants game?

It's my impression that the Eagles weren't as dominated by the Ravens as the final score suggests -- the Eagles's defense played well throughout, with the huge difference in the final score resulting from a handful of plays (such as the TD pass to Mark Clayton). Ditto with the Giants/Ravens game. Though the Giants totally outclassed the Ravens in the first half, the Ravens played well enough in the second half for the final score to have been close -- again except for a couple plays (the Flacco interception returned for a TD and one long run). DVOA suggests that on a neutral field the Giants are about a field goal better than the Ravens, Eagles, and Titans; but the high variance of NY, Baltimore, and Philly suggests that the outcome of any of the three playing one of the other two would be more uncertain than usual, with the result quite possibly determined by turnover differential and the presence or absence of game-turning plays.

42 Re: Week 13 DVOA Ratings

I don't know about the other two teams but isn't the Giants' variance so high because of their one comparatively bad game in Cleveland in relevance to the rest of their ridiculous performances? I know Aaron said something like that a few weeks back... That the Giants' variance was high cause their best is so good and their worst is only average.

68 Re: Week 13 DVOA Ratings

The Giants game against the Bengals was pretty bad too (the Bengals took them to overtime), and the Rams were down by just 7 points with 7 minutes 45 seconds left in regulation, and the Giants stuck in 2nd and 10. Two long passes later, an interception return TD, and a long punt return and long run later (in other words, five broken plays happening closely together), and the Rams had suddenly metled down and lost by 28.

Playing 25% of your games like total crap against really bad teams will affect your variance. The game against the 49ers wasn't anything special either.

The Original Andrew

74 Re: Week 13 DVOA Ratings

I'm pretty sure DVOA includes all 60 minutes of the game, and that long passes and long runs are good things. They're only "broken plays" if you're desperately hoping against hope that the other team wins.

I mean, honestly. "Total crap"? The Giants gained 441 yards in that game to the Rams' 200, 45 of which came on the TD which wasn't exactly Bradshaw to Swann. At least argue that the Rams were gameplanning to stop Kevin Boss, or something.

80 Re: Week 13 DVOA Ratings

The point is that is 11 fairly poor quarters (lets call the last rams quarter fantastic). So you have 30 or so great quarters, 4 mediocre ones (SF) and 11 bad ones, = high variance. That is all that they are pointing out, you dont need to be so defensive about the NYG all the time Kurt. It is just a football team.

84 Re: Week 13 DVOA Ratings

I swear, someone could assert the Giants wear pink and gold uniforms, a Giant fan could say "no, they'e blue" and some pipsqueak will come along to accuse the Giant fan of being whiny and defensive.

Can we focus on the issue now? The Rams game didn't involve 3 bad quarters. The Giants destroyed the Rams for three quarters, outgaining the Rams 308 yards to 91. Eventually, the score caught up to the play on the field. DVOA doesn't care what the score was. Even if you thought they did play bad for three quarters, and great in the fourth, that wouldn't explain the high variance. Here, lets look at the top of the page:

"VARIANCE measures the statistical variance of the team's weekly DVOA performance."

You see that word? Weekly? Not quarterly, or minute-by-minute?

I don't have the premium package, so I don't know what the DVOA's are for each game. I do know that in the week 2 writeup, the Rams were described as having a VOA below -100% for the game. Which leads me to believe that the Giants' variance isn't explained by their playing like crap during the Ram game. That's all.

87 Re: Week 13 DVOA Ratings

I don't think it's fair to call him defensive for objecting to inaccurate statements. As he and Dales have pointed out, the Rams game was a pretty solid thumping, and the Bengals game wasn't bad, merely mediocre(And that was with Palmer starting, not Fitzpatrick). I don't know about the 49ers game, but at most you're talking about 1 bad game and 2 mediocre ones. And I agree trying to break it down by quarter becomes problematic as there were probably bad quarters in good games and good quarters in bad games.

82 Re: Week 13 DVOA Ratings

DVOA does not see the Rams game as "total crap". Even after opponent adjustments, it is considered the Giants' fourth best game of the year, behind the efforts against Seattle, Dallas, and Pittsburgh.

Also, the Cincy game grades out as being just ever-so-slightly below league average.

The high variance is due to the absolute clunker against Cleveland, the nearly average game against Cincy, and the rest being very high.

ETA- why does it not see the StL game as crap? Consider that the points that had the Rams within 7 at halftime came from two 54 yard field goals, and DVOA understands that those are going to fail somewhat often. In the first half, when the score was close, the Rams had a total of 68 yards of offense compared to 193 for the Giants. A mid-game snapshot of DVOA would have indicated that while the score was close, the Rams were being very badly outplayed and were fortunate to be that close. And, as it turns out, things got worse for the Rams from there.

46 Total DVOA and ST effect

Maybe I'm missing something here but total DVOA rating seems to be the simple addition of the three components (off, def, st). But special teams is only 1/7 of the game according to this site. Are they given a equal 1/3 rating in Total DVOA?

48 Re: Total DVOA and ST effect

If you look at the deviation of a team's overall ST DVOA, you'll see that it tends to be much, much, much more clustered around 0 than Off DVOA or Def DVOA. My assumption is that Aaron fiddles the numbers somehow-either making ST plays 1/3 as valuable as they would otherwise be or divides ST numbers by 3 to properly reflect their "accurate" value.

EDIT: If you look at the ST page, you find "The total [points a team receives from the components of special teams] is then converted into a DVOA percentage so that it can be added in to offense and defense to create total team DVOA." I'm sure that conversion takes into account the research that ST is 1/3 as valuable as Offense or Defense.

47 Vikings Prediction

When the Vikings were 3-4 I predicted they would go on a 5-1 run but then lose their last three to miss the playoffs. Now with the Williams tackles gone for the season I'm starting to think my prediction is looking pretty good right now.

It will be intriguing to see how much having those two out of the lineup effects the defence. The entire middle of their defence will now be gone with EJ Henderson out as well.

50 Re: Week 13 DVOA Ratings

News to Tom - makes sense. My gut tells me that special teams performance has a far larger luck factor to it than does offence and defence.

51 Re: Week 13 DVOA Ratings

Playoff odds has the Lions with only a 24% chance of going 0-16. I humbly submit that this is way too conservative. It means that Playoff Odds thinks the Lions have a 30% chance of winning each game they play, roughly. They aren't nearly that good.

65 Re: Week 13 DVOA Ratings

Yea, the expectation that the Lions should win between one and two of their remaining four games is wildly optimistic, given that they lost their first twelve.

I think it's probably about a 55% chance they lose all their remaining games. While the offense really isn't so horrid, the defense is abominable. It's going to allow around 2900 rushing yards at 5+ yards per carry, and about 30 rushing touchdowns before the season is over. They will likely shatter the NFL record for points allowed in a season.

79 Re: Week 13 DVOA Ratings

I think systems like DVOA might undervalue great teams and overvalue terrible teams because it cannot see that in the blowouts more substitutions are made and the play calling gets very conservative. There are some adjustments for this, but simply going by the play by play I think it is hard to account for how being up say 35-10 in the third quarter effects the game. I get the feeling a lot of teams could have put 60 or 70 up on the Lions if they went for broke this year, but there is almost no incentive to do so.

89 Re: Week 13 DVOA Ratings

IIRC, Aaron looked at removing some garbage time (specifically meaningless Week 17 games) and found that it didn't improve the numbers.

55 Re: Week 13 DVOA Ratings

Your math is wrong Yaguar, the odds are giving them somewhere around a 6.5% chance to win each game (if the odds were equal for all 4).

83 Re: Week 13 DVOA Ratings

The edit button is available if and only if you are a registered user who has logged in.

It is free- you don't need to buy premium to register.

67 Re: Week 13 DVOA Ratings

No, Yaguar is correct. If they have a 70% chance to lose each game, the chances of losing all four is (.70*.70*.70*.70), or .2401 (24.01 %) If they really had a 6.5% chance of winning each game (93.5% chance of losing) they'd have a 76.4% chance of going 0-16.

54 Re: Week 13 DVOA Ratings


Sorry. Just had to say it. Excuse me, I feel a relapse coming on...


58 Re: Week 13 DVOA Ratings

Jimm, Tundra, don't give up hope, at least not until tomorrow or Thursday. There's a reasonable chance that the Williams, et. al., will be granted an injunction, thereby delaying any possible suspensions until next year, especially if the request is heard by Judge David Doty in Minneapolis. Then again, since I've been half-hopin' that the Vikings miss the playoffs, thus guaranteeing that the brilliant talent evaluator that is Brad Childress would get the boot, and thus perhaps preventing Leslie Frazier from being hired as a head coach by another team, maybe I should be rooting for the NFL!

77 Re: Week 13 DVOA Ratings

In reply to by Will Allen (not verified)

Wow, you really draw out the pessimistically positive in the situation. So either we make the playoffs (booyah!), and Childress stays (oy vey!), or we miss out on the playoffs (oy vey!) and Childress gets canned (booyah!).

Except that Childress getting canned isn't a given (he should, obviously, but still...not a given), and that still doesn't solve the problem of Nine Lives Spielman, the Fredo of NFL front offices, occupying the GM role.

- the poster formerly known as 'Tundrapat'

PS - I see the adverts for beery get-togethers in NY, Boston, and Philly. Why none in Seattle? Or better yet, Anchorage? We've got sooo much to recommend (five hours of daylight and 10-degree weather!).

61 Re: Week 13 DVOA Ratings

OK, so I was just looking at the "Defense vs. Receivers" page, and noticed that Kansas City had an insanely low yards/game for no.2 receivers: 7.1 yards. Is that right? I need to replace Plax (man, after his ONLY good game this year, someone offered me a pretty good trade [can't remember it off the top of my head] and I turned it down. Son of a...) and was looking at Bess from Miami, but if that number is true, he might not really help me that much.

62 Re: Week 13 DVOA Ratings

OK, so I was just looking at the "Defense vs. Receivers" page, and noticed that Kansas City had an insanely low yards/game for no.2 receivers: 7.1 yards. Is that right? I need to replace Plax (man, after his ONLY good game this year, someone offered me a pretty good trade [can't remember it off the top of my head] and I turned it down. Son of a...) and was looking at Bess from Miami, but if that number is true, he might not really help me that much.

69 Raiderjoe still has hope

The Raiders still have a .1% chance of winning their division. Wild.

71 Re: Week 13 DVOA Ratings

Evolutionary psychologists contend that there is something innately attractive about perfect symmetry: petals, cathedral domes, human faces, the Rams' offensive and defensive DVOA...

Maybe not.

73 Re: Week 13 DVOA Ratings

Has there ever been a team ranked as a high as #4 by DVOA not make the playoffs? The Eagles seem like a they have a decent chance this year. In fact they could easily reach the #2 rank in DVOA and still miss the playoffs. Wow.

85 Re: Week 13 DVOA Ratings

The Ravens are rated even higher and hardly have an easy task ahead of them. They've got the Redskins, Steelers, Cowboys, and Jaguars on tap, and have a 1-game lead for the last wildcard spot. (I'm penciling Indy in as the 5 seed.) If the Ravens go 2-2, somebody from the AFC East will pass them. (Note: if the Dolphins win their last four games, they win the division based on what would be a 3-1 record against the Pats and Jets.)

88 Re: Week 13 DVOA Ratings

Denver is probably rated where they need to be, but they have to be the most difficult team to evaluate I have seen. They are fully capable of beating anyone (even on the road) and losing to anyone (even at home).

90 Re: Week 13 DVOA Ratings

Funny, the Jets have played two games against teams ahead of them and won both by 21 points. against teams 19-30 they are 4-4, with two of their wins being scant victories at home vs cincy and kc. tough team to read

94 Re: Week 13 DVOA Ratings

would of gone to dce 18th one at Chichie and Petes but already have plans to go to work Christmas party. Party is at 8 on same night.
going to get drunk.

Raiders bad but still ahead of Chjiefs so that is good. At beginning of season nobody would expect Bills and Raiders to be in bottom 8 teams in Decemver . Raiders were 1-1 and Bills 2-0.
Raiders goign to get real head coach in offeseqson and going to turn around team.
Lane Kiffin going to take Tennessee down. Just a crappy coach and liar.