Writers of Pro Football Prospectus 2008

Most Recent FO Features


» 2018 Free Agency Cost-Benefit Analysis

Is Kirk Cousins the best free-agent quarterback in recent memory? Should Trumaine Johnson or Malcolm Butler have gotten the larger contract? And what makes a free-agent contract good or bad, anyway?

31 Dec 2012

Final 2012 DVOA Ratings

by Aaron Schatz

A bit of a close call in their final game isn't enough to drop the Seahawks out of the top spot in the Football Outsiders DVOA ratings, making Seattle our top team for 2012. Denver and New England had convincing wins in the final week and move a bit closer to Seattle, but still rank two and three. San Francisco is fourth, and Green Bay rounds out the top five.

Each year, we go through and look at how the best and worst teams of the year fit in among all the teams which have DVOA ratings; right now, that goes back to 1991. Perhaps what's most remarkable about this season is that three different teams end up in the all-time DVOA top 12. When I announced this on Twitter, one follower responded that none of these teams really seems like an all-time top ten team. That seems true, and I wonder why that is. Some of it may be that Seattle stuffed so much of its value into three games, but that's not true of Denver or New England.

BEST TOTAL DVOA, 1991-2012
Year Team DVOA
1991 WAS 56.9%
2007 NE 52.9%
2010 NE 44.6%
1996 GB 42.0%
1995 SF 40.0%
2012 SEA 38.3%
2004 PIT 37.6%
2012 DEN 36.6%
2010 PIT 35.4%
1992 DAL 35.1%
2012 NE 34.9%
2004 NE 34.2%

Honestly, if you compare this year's best teams to some of the other great teams of the last 20 years, they look pretty good. The 1992 Cowboys, for example, went 13-3 and outscored opponents by 166 points. That's often considered the best of the three Cowboys teams that won the title. Well, the Broncos just went 13-3 and outscored opponents by 192 points. The Patriots went 12-4, the same as the 1993 Cowboys and 1995 Cowboys, and outscored opponents by 226 points, far more than either of those Cowboys teams.

The 1992 Cowboys lost to Philadelphia 31-7 in Week 5. The 1994 49ers lost to Philadelphia 40-8, coincidentally also in Week 5. The 1996 Packers lost to Dallas 21-6 in Week 12. That's three of the best teams of the 90's, and each one had at least one loss by two touchdowns. But Seattle, Denver, and New England combined for zero losses of more than two touchdowns. In fact, the only double-digit loss by any of those three teams was when New England beat Denver 31-21.

These are legitimately great teams, in part because they are so well-balanced. Seattle ended up ranking in the top four in offense, defense, and special teams. Denver (and San Francisco as well) ranks in the top five in both offense and defense, although the special teams are a little lower. The Patriots are known for being much better on offense than on defense, but even their defense ended up 15th in DVOA this year. Is that partly because they have so many turnovers? Sure, but FO metrics don't exactly overdose on the value of turnovers. We know that they are more variable than yardage totals.

The other remarkable thing about both Seattle and Denver is the improvement. Seattle was 19th in DVOA last year; Denver was 24th even though the Broncos snuck into the playoffs. That means that both Denver and Seattle make the list of the best year-to-year DVOA improvements we've ever measured. Denver is second behind only the 2010 Lions, while the Seahawks are fourth.

Best Year-to-Year DVOA Improvement, 1991-2012
Year Team DVOA Y-1 Rank DVOA Rank Change
2010 DET -51.6% 32 -1.1% 18 50.5%
2012 DEN -11.8% 24 36.6% 2 48.3%
1999 STL -9.9% 20 34.0% 1 43.8%
2012 SEA -1.5% 19 38.3% 1 39.8%
1999 OAK -18.3% 27 21.2% 3 39.5%
2000 NO -40.3% 31 -0.9% 19 39.4%
2004 PIT -1.6% 19 37.6% 1 39.1%
2004 BUF -7.3% 23 31.3% 3 38.7%
2008 CAR -20.6% 26 18.0% 6 38.6%
2007 TB -19.7% 28 17.8% 8 37.6%

Readers know we've been tracking how well the best 2012 teams rank in DVOA for a few weeks now, not only for total DVOA but also for each unit. In the end, the 2012 Patriots don't end up on the list of the top dozen offenses ever. The Bears defense and the Ravens special teams don't end up ranking as highly as we expected a few weeks ago, but they do make the all-time lists:

Year Team DVOA x Year Team DVOA
1991 PHI -42.4% x 2002 NO 12.2%
2002 TB -31.8% x 2007 CHI 11.2%
2008 PIT -29.0% x 1994 CLE1 10.1%
2004 BUF -28.5% x 1996 CAR 9.8%
2008 BAL -27.8% x 2009 CLE 9.7%
2012 CHI -26.8% x 1998 DAL 9.2%
2009 NYJ -25.5% x 2012 BAL 9.0%
2000 TEN -25.0% x 2001 PHI 8.9%
2003 BAL -25.0% x 1997 DAL 8.9%
1991 NO -24.5% x 2000 MIA 8.8%
2000 BAL -23.8% x 2005 BUF 8.8%
1995 SF -23.7% x 2004 BUF 8.7%

On the other hand, we've got the Indianapolis Colts, otherwise known as "the worst 11-5 team in NFL history." I don't mean to take anything away from the Colts' big win over Houston yesterday. The Colts get 54.3% DVOA for that game, by far their best single-game DVOA of the season. It was enough to raise them up almost five percentage points, and they went from 28th to 25th. But this is still the worst 11-5 team ever. The Colts have -16.0% DVOA, making them the first team to ever go 11-5 with a DVOA below -10%. The only other teams to go 11-5 with negative DVOA were the 2000 Vikings (-6.3%), the 2004 Falcons (-4.8%), and the 2005 Bears (-0.9%). In fact, even if they had lost to Houston on Sunday, the Colts would be the first team to ever go 10-6 with DVOA below -10%. 

Would you prefer simple points scored and allowed to complicated DVOA? OK, well, the Colts were the first 11-5 team ever to be outscored by their opponents, 387-357. The difference between the Colts' regular winning percentage and Pythagorean winning percentage, .238, is the second-highest since the AFL-NFL merger. The only team with a bigger gap was the 1992 Indianapolis Colts, who went 9-7 despite being outscored 302-216. That Colts team had a -27.2% DVOA. They were 1-15 the year before and 4-12 the year after. I have a feeling that the 2013 Colts won't be that bad, because they're going to get improvement from their young quarterback and they'll draft a lot of talent on defense, but their luck will regress even if their Luck does not. The Colts are also unlikely to play the league's easiest schedule again.

The Colts do not have the worst DVOA of any playoff team ever. Three teams were worse: the 2004 Rams (-27.2%), the 2010 Seahawks (-22.9%), and the 1998 Cardinals (-17.1%). All three of those teams won their first playoff game anyway.

The flip side of the Colts would be the Detroit Lions, who went 4-12 despite finishing 16th in DVOA. The Lions had 6.4 Pythagorean wins, and went 3-9 in games decided by a touchdown or less.

Despite the yardage totals of Adrian Peterson and Calvin Johnson, this year there really weren't any players who came close to any DVOA or DYAR records. We'll look at the best and worst players in a Quick Reads article tomorrow. 

All stats pages should now be updated (or, at least, will be in the next few minutes) including snap counts and the FO Premium database. Loser League results will be updated with the winner announced on Wednesday afternoon.

* * * * *

These are the Football Outsiders team efficiency ratings through the entire 2012 regular season, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)

OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. WEIGHTED DVOA represents an attempt to figure out how a team is playing right now, as opposed to over the season as a whole, by making recent games more important than earlier games. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE. LAST WEEK represents rank after Week 16, while LAST YEAR represents rank in 2011.

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 SEA 38.3% 1 19 11-5 46.6% 1 18.5% 4 -14.1% 4 5.7% 3
2 DEN 36.6% 2 24 13-3 41.4% 2 22.1% 2 -13.8% 5 0.7% 13
3 NE 34.9% 3 3 12-4 37.5% 3 30.8% 1 1.3% 15 5.5% 4
4 SF 29.9% 4 6 11-4-1 24.1% 5 17.0% 5 -14.3% 2 -1.5% 20
5 GB 26.6% 5 1 11-5 24.2% 4 19.5% 3 -7.3% 8 -0.2% 18
6 CHI 20.5% 6 15 10-6 15.7% 8 -10.9% 26 -26.8% 1 4.7% 6
7 NYG 13.5% 9 12 9-7 9.5% 10 12.9% 7 1.4% 16 2.0% 10
8 BAL 9.8% 8 7 10-6 8.3% 11 3.0% 13 2.2% 19 9.0% 1
9 WAS 9.6% 11 21 10-6 18.1% 6 15.3% 6 1.8% 17 -4.0% 27
10 ATL 9.1% 7 8 13-3 2.4% 13 6.1% 12 -2.9% 12 0.1% 16
11 HOU 6.6% 10 5 12-4 -3.0% 19 0.1% 16 -14.2% 3 -7.7% 32
12 CIN 6.1% 13 17 10-6 18.0% 7 -1.8% 17 -3.8% 10 4.1% 7
13 CAR 5.5% 12 20 7-9 10.9% 9 7.1% 10 -3.1% 11 -4.8% 29
14 MIN 2.0% 16 29 10-6 -1.5% 17 0.3% 15 3.1% 21 4.7% 5
15 STL 1.1% 17 32 7-8-1 4.3% 12 -4.7% 21 -9.2% 7 -3.4% 26
16 DET 0.2% 15 11 4-12 1.3% 14 12.3% 8 7.0% 24 -5.1% 30
17 DAL -0.4% 14 14 8-8 -0.4% 16 6.1% 11 6.7% 23 0.2% 15
18 PIT -1.0% 18 4 8-8 0.6% 15 -3.8% 19 -2.9% 13 -0.1% 17
19 NO -5.2% 20 2 7-9 -1.8% 18 11.9% 9 14.8% 32 -2.3% 23
20 TB -6.6% 21 30 7-9 -9.1% 23 0.6% 14 3.0% 20 -4.3% 28
21 MIA -7.3% 19 18 7-9 -12.1% 24 -8.4% 22 -0.6% 14 0.4% 14
22 SD -9.1% 22 16 7-9 -7.0% 22 -10.1% 24 2.0% 18 3.0% 8
23 BUF -12.2% 24 23 6-10 -6.9% 21 -4.3% 20 10.6% 27 2.7% 9
24 CLE -13.5% 23 25 5-11 -6.6% 20 -15.1% 27 4.5% 22 6.1% 2
25 IND -16.0% 28 31 11-5 -14.0% 25 -2.9% 18 14.0% 31 0.9% 12
26 ARI -16.3% 26 28 5-11 -23.4% 27 -31.0% 32 -13.5% 6 1.1% 11
27 NYJ -17.9% 25 10 6-10 -18.2% 26 -20.6% 30 -4.2% 9 -1.5% 21
28 PHI -22.7% 27 9 4-12 -27.3% 30 -10.8% 25 9.4% 26 -2.6% 24
29 OAK -27.8% 29 22 4-12 -26.2% 29 -9.6% 23 12.5% 29 -5.8% 31
30 TEN -29.3% 31 13 6-10 -23.6% 28 -20.5% 29 7.4% 25 -1.4% 19
31 JAC -33.0% 30 27 2-14 -29.6% 31 -18.3% 28 11.7% 28 -3.0% 25
32 KC -40.4% 32 26 2-14 -39.8% 32 -25.1% 31 13.2% 30 -2.1% 22
  • NON-ADJUSTED TOTAL DVOA does not include the adjustments for opponent strength or the adjustments for weather and altitude in special teams, and only penalizes offenses for lost fumbles rather than all fumbles.
  • 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.
  • 2012 SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative).
  • PYTHAGOREAN WINS represent a projection of the team's expected wins based solely on points scored and allowed. Please note that this is based on the new formula introduced last year which changes the exponent of the Pythagorean formula based on the offensive environment of each team's games.
  • 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).
RANK 2012
1 SEA 38.3% 11-5 30.6% 13.0 3 5.1% 4 12.5 3 14.3% 21
2 DEN 36.6% 13-3 38.8% 14.7 1 -6.8% 31 12.5 2 7.7% 4
3 NE 34.9% 12-4 37.6% 13.4 2 -2.9% 21 12.7 1 11.8% 11
4 SF 29.9% 11-4-1 24.0% 12.5 4 6.3% 3 11.4 4 21.7% 31
5 GB 26.6% 11-5 18.8% 11.8 5 2.2% 9 10.5 7 16.3% 27
6 CHI 20.5% 10-6 18.2% 11.0 6 2.8% 8 10.8 6 12.6% 12
7 NYG 13.5% 9-7 12.3% 9.5 8 2.1% 10 10.2 9 33.5% 32
8 BAL 9.8% 10-6 11.6% 9.2 9 -1.0% 16 9.4 11 15.6% 24
9 WAS 9.6% 10-6 13.7% 9.9 7 -0.7% 15 9.2 12 8.5% 5
10 ATL 9.1% 13-3 14.7% 9.1 10 -4.3% 27 11.2 5 13.7% 17
11 HOU 6.6% 12-4 14.7% 8.3 14 -4.0% 26 10.2 8 14.0% 18
12 CIN 6.1% 10-6 12.1% 8.7 13 -5.6% 29 9.9 10 14.0% 19
13 CAR 5.5% 7-9 5.0% 8.8 11 0.8% 13 7.8 18 13.1% 13
14 MIN 2.0% 10-6 -0.3% 8.8 12 4.9% 5 8.8 13 7.6% 3
15 STL 1.1% 7-8-1 -6.3% 8.1 15 9.6% 2 6.6 22 9.9% 8
16 DET 0.2% 4-12 -5.7% 7.6 18 4.1% 6 6.4 23 5.3% 1
RANK 2012
17 DAL -0.4% 8-8 -4.6% 7.9 16 4.0% 7 7.4 19 6.9% 2
18 PIT -1.0% 8-8 5.5% 7.4 20 -5.2% 28 8.7 14 14.5% 23
19 NO -5.2% 7-9 -3.3% 6.4 23 2.0% 12 8.2 15 14.1% 20
20 TB -6.6% 7-9 -3.6% 7.8 17 -1.2% 17 7.9 17 13.6% 16
21 MIA -7.3% 7-9 -4.9% 7.6 19 -1.9% 19 7.1 21 16.1% 25
22 SD -9.1% 7-9 -4.5% 6.6 21 -6.6% 30 8.0 16 9.3% 7
23 BUF -12.2% 6-10 -13.7% 6.5 22 -3.3% 24 5.7 25 16.5% 28
24 CLE -13.5% 5-11 -8.3% 6.2 25 -2.4% 20 6.1 24 11.5% 10
25 IND -16.0% 11-5 -9.7% 6.2 24 -7.4% 32 7.2 20 11.4% 9
26 ARI -16.3% 5-11 -27.7% 4.8 27 10.7% 1 4.8 27 16.1% 26
27 NYJ -17.9% 6-10 -15.9% 5.6 26 0.1% 14 5.3 26 17.6% 29
28 PHI -22.7% 4-12 -26.6% 4.5 28 2.1% 11 3.9 30 13.4% 14
29 OAK -27.8% 4-12 -23.5% 3.7 29 -3.9% 25 4.1 29 14.4% 22
30 TEN -29.3% 6-10 -28.3% 3.3 30 -3.0% 22 4.6 28 20.2% 30
31 JAC -33.0% 2-14 -29.7% 2.7 31 -3.3% 23 3.3 31 8.8% 6
32 KC -40.4% 2-14 -44.0% 2.3 32 -1.5% 18 2.5 32 13.4% 15

(Note: Although this post is titled "Final DVOA Ratings," unofficial postseason ratings will continue each Monday through the playoffs. Also, play-by-play changes made over the next few weeks could result in some small changes to these final ratings.)

Posted by: Aaron Schatz on 31 Dec 2012

190 comments, Last at 04 Jan 2013, 8:59pm by jonnyblazin


by nuk :: Mon, 12/31/2012 - 5:32pm

In addition to being the worst DVOA team ever with at least 10 wins ('12), the Colts are the worst-rated team with at least 9 wins ('92), 13 wins ('99), and 14 wins ('09). FO hates the Colts :)

by Bobman :: Wed, 01/02/2013 - 3:35am

That's kind of neat. I used to gripe to Aaron that they always under-projected the Colts and DVOA just couldn't really grasp Manning's contribution. Turns out it's the damn horseshoe that's the problem. Given the turnover at all levels over the years, that's a pretty odd quirk for one team out of 32 to have.

Though regarding "luck" I'd have to say this year was actually pretty crappy. Injuries, turnovers, Cecil Shorts--half of the Jags' win total! Crosby missing a late kick for GB is the only "they won because of a lucky bounce" event I can remember offhand. I DID read that Luck had a high total of dropped INTs, which is both fortuitous and correctable as a QB matures.

I cannot see their luck being measurably worse next year.

by JoeyHarringtonsPiano :: Wed, 01/02/2013 - 6:59am

You may be not be remembering other "lucky bounces". For instance, on their game-winning drive against Detroit, Drayton Florence dropped a Luck INT that hit him in the hands (no, I'm not still angry/bitter). I didn't watch every Colts game, but with has many close wins as they had, there are bound to be others (and not just on the final drive).

by whitexhysteria (not verified) :: Thu, 01/03/2013 - 12:55am

You have obviously forget about the absolute garbage blowing of the whistle towards the end of regulation against Tennessee, in which a fumble that was recovered by Tennessee in winning field goal range was called due to "forward progress." If you watched that game, then you can probably agree the receiver/tight end/back that had the ball was not even close to the point that the whistle is blown the other 99 times. Thus, being another lucky bounce that they received.

So, with the current comments we have three lucky bounces they got. They outplayed themselves a lot this year, though with Luck progressing over the offseason I don't think they will fall off much, if at all.

by TonyinShanghai (not verified) :: Thu, 01/03/2013 - 1:00am

Sometimes it IS about winning ;)

by Nick Bradley (not verified) :: Mon, 12/31/2012 - 5:36pm

your model is broken.

by Perfundle :: Mon, 12/31/2012 - 5:47pm

Best teams in DVOA history does not equal best teams in NFL history. To begin with, DVOA only goes back to 1991; no 1985 Bears or 1972 Dolphins, for instance. Second, DVOA has been adjusted so that the league average is 0. All you can say is that Seattle, Denver and NE are among the best teams at beating the league average.

by AT (not verified) :: Tue, 01/01/2013 - 12:49pm

Other more comprehensive metrics have considered the '91 Redskins and '85 Bears to be essentially tied for best team of all time.

by Cuenca Guy :: Tue, 01/01/2013 - 6:12pm

Excellent point. DVOA isn't about total quality, but about quality compared to "average".

by Gus (not verified) :: Mon, 12/31/2012 - 5:55pm

Great analysis.

by Dan in Philly (not verified) :: Mon, 12/31/2012 - 5:40pm

Peyton Manning is overrated...

by anon24 (not verified) :: Mon, 12/31/2012 - 9:30pm

God, yes. It's so obvious that the wide receivers and the offensive line have been carrying the Broncs all season. Peyton was just along for a ride.

by Bobman :: Wed, 01/02/2013 - 3:37am

Absolutely. I even heard that Archie and Olivia filed papers to disinherit Peyton and adopt Decker and Thomas, so it must be true.

by In_Belichick_We... :: Mon, 12/31/2012 - 5:50pm

Defining accomplishment of the 2012 Jacksonville Jaguars:
"In the end, the 2012 Patriots don't end up on the list of the top dozen offenses ever."
The Jags got to spoil something anyway.

by Myk (not verified) :: Mon, 12/31/2012 - 6:05pm

The point about not having a big loss is a big deal for teams like Seattle. Last night I went back and looked to if d that for the entire season the team was behind by more than 7 points for a TOTAL of 25 minutes and 44 seconds. Most of that comes from the New England game where Seattle fell behind by 13 points for most of a quarter. Of course, they ended up coming back to win that game. Take out the NE game and Seattle was playing while behind by more than one possession for roughly 10 minutes. Someone might be able to prove me wrong...but that seems to be a pretty good sign of a great team.

by kramer (not verified) :: Mon, 12/31/2012 - 7:20pm

Similar to this, the 2010 Packers never trailed by more than 7 points all season.

by zdneal (not verified) :: Thu, 01/03/2013 - 1:40pm

And similar to those Packers, Seattle benefited from an improperly called touchdown catch.

by Fkdc (not verified) :: Mon, 12/31/2012 - 7:29pm

I'm not sure that really means anything. The Redskins, who are a merely a good team, had only one loss by two possessions (steelers game). They trailed for most of the Panthers game too, but I think they led in the 4th quarter in their other 4 losses.

I haven't looked back at the game logs, but I don't remember them trailing by double digits for extended period of time other than the Steelers game.

I'm sure some other good or even average teams have seasons where they don't lose any big ones.

by Perfundle :: Mon, 12/31/2012 - 8:58pm

Yeah, weird stats like that don't really mean that much. Stats like Pythogorean wins are much more indicative of ability, the 2011 Giants notwithstanding, and here, Seattle is 3rd and Washington is 12th.

by Myk (not verified) :: Mon, 12/31/2012 - 10:41pm

Pretty sure that never being down by a large amount is important to judging a teams skill. They listed above that one reason the three team performed so well in DVOA was cause they never got blown out. Staying close consistently. Through the ups and downs of your team is a good thing

by Anger...rising :: Mon, 12/31/2012 - 10:50pm

Not getting blown out matters. The order in which the points are distributed within the game doesn't. It's like crediting the 1990 Cincinnati Reds as a team of historical significance, despite winning 91 games, based on leading the NL West every day of the season. That's trivia.

by Perfundle :: Tue, 01/01/2013 - 12:49am

Right. New England's DVOA didn't get dropped more than normal when they went down 31-3 against San Francisco (and thus were behind by more than 7 points for a long duration), because they ended up losing 41-34. I mean, even the 1985 Bears lost their only game by double digits. One or two bad games doesn't disqualify a team from greatness.

by The Ninjalectual :: Mon, 12/31/2012 - 9:24pm

That Panthers game was a fluke. The Panthers touchdown came 35 yards after the play had been whistled dead. The Redskins defenders heard the whistle and stopped trying to tackle DeAngelo Williams, because that's a good way to get a 15-yard penalty and a hefty fine. Even though Williams never stepped out of bounds and continued running to the end zone, the ref thought he had stepped out, and blew his whistle. Therefore, it sucks that it was a bad call, but the play is over.

But no! The refs retroactively decided the whistle didn't count (and in fact denied that there even was a whistle) and awarded a touchdown to the Panthers. It was one of the most infuriating calls of the season, even moreso than the more famous Seattle-Green Bay miscall. On that one the replacement refs were close, if incorrect. On this one the regular refs were absolutely wrong, and wrong in a way that destroys the authority of the referee and the integrity of the game. Just a terrible, terrible call.

by Zach (not verified) :: Mon, 12/31/2012 - 11:09pm

Agreed, that call was just insanely bad.

by Aaron Brooks Go... :: Wed, 01/02/2013 - 10:54am

I think the most infuriating call of the season was when the NFL changed their rule on coaches challenges after a TD/TO after the Detroit-Houston game, and didn't tell anyone.

Because they sure didn't enforce it in the MIN-GB game.

by Eddo :: Wed, 01/02/2013 - 12:24pm

They didn't "change[] their rule". Per the officials, McCarthy's challenge came after the replay booth official signaled to the crew chief that the Jones fumble should be reviewed.

Schwartz, on the other hand, threw his challenge flag before the replay booth official signaled for a review. (I believe he even threw it while the play was still going on.)

Now, you can quibble with the timing (did McCarthy actually throw his flag prior to the signal, and the officials didn't see it - or worse, willfully ignored it?), but the rule absolutely did not change.

by George (not verified) :: Wed, 01/02/2013 - 12:34pm

If every TD or turnover is to be reviewed by default, then why would it matter when the challenge flag came out? Both plays were going to be reviewed, it shouldn't matter if the booth signaled for review before the flag or not. The officials changed how the rule was enforced.

by Eddo :: Wed, 01/02/2013 - 1:24pm

Every touchdown or turnover is most emphatically NOT reviewed "by default". Scoring plays and turnovers are treated as any other play that happens inside the two-minute warning: the booth official is the one to trigger a review, but it's still at his discretion.

by Aaron Brooks Go... :: Wed, 01/02/2013 - 2:14pm

The actual statute is interesting in the context of the DET-HOU and GB-MIN rulings. Regardless of interpretation, they aren't a consistent application of the rules.

The rule (Rule 15, Section 9) is that the replay official has authority over scoring plays and turnovers, as well as two-minute warning situations. The penalty application is specifically that a challenging team cannot benefit if they trigger a delay-of-game penalty. (Which is how the NFL justifies not reviewing the erroneous Forsett TD).

Thing is, GB was penalized 15 yards for an illegitimately challenge. Thus, they were ruled to have delayed the game. Under 15:9, that means that the challenge had to stop -- GB was now forbidden by rule from benefiting from the ruling. The rules make no allowance for the timing of a coach's challenge.

Either the referees completely botched one of the two situations, or the NFL changed the interpretation midstream.

by Eddo :: Wed, 01/02/2013 - 2:57pm

I disagree.

The Schwartz foul was considered a delay, since it came before the booth official requested a review.

The McCarthy foul was not considered a delay, since it came after the booth official requested a review.

The fouls themselves are considered Unsportsmalike Conduct.

I'm curious as to where you found that text, as well. Is that from the full rulebook or the digest?

by Aaron Brooks Go... :: Wed, 01/02/2013 - 3:43pm

Full rulebook. (page 97 of 120 of the pdf)

Thing is, if the booth is going to review it anyway, then throwing the flag *isn't a delay*.

More and more, I'm of the opinion that the officials completely botched the application of the rule in the DET-HOU game, and the NFL won't come out and admit it because it determined the outcome of the game. They are damned lucky that also bungling the GB-MIN call didn't.

Because god knows, complete rule misapplications didn't totally change the seeding of Houston, Minnesota, GB, or Seattle this year.

by Eddo :: Wed, 01/02/2013 - 4:09pm

"Thing is, if the booth is going to review it anyway, then throwing the flag *isn't a delay*."

But when Schwartz(*) threw the flag, no one knew that the play would be reviewed. The booth official had yet to make a decision. In theory - as in, why the rule is in place - Schwartz could have been throwing the flag to stall while the booth official has more time to look at the call.

(*) Possibly McCarthy, too, but the officials' position is that McCarthy's flag came out after the booth official called for a review.

by zdneal (not verified) :: Thu, 01/03/2013 - 1:44pm

So are you saying that if the official in the booth had decided not to review the Lions play then Schwartz would have been unable to challenge it? And if the booth decided not to review it Schwartz would still have gotten the penalty?

by Eddo :: Thu, 01/03/2013 - 2:03pm

"So are you saying that if the official in the booth had decided not to review the Lions play then Schwartz would have been unable to challenge it?"

Correct. Coaches cannot challenge turnovers and scoring plays. Only the booth official can call for a review.

"And if the booth decided not to review it Schwartz would still have gotten the penalty?"

Correct. Throwing a challenge flag on an unchallengeable play is a fifteen-yard unsportsmanlike conduct foul.

by Cuenca Guy :: Tue, 01/01/2013 - 6:18pm

The Seahawks, however, had the 4th toughest schedule compared to the Redskins' 15th toughest schedule. That makes a difference when looking at the "close games" issue. Also looking at total point differential I think also helps if you're going to look at close games...the Seahawks ended up 3rd in point differential compared to 12th for the Redskins.

by JoeyHarringtonsPiano :: Mon, 12/31/2012 - 6:13pm

If I were Jim Schwartz, and I were about to meet with the owner about my job, I would pull up this article on my iPad and bring it with me to the meeting (for the all-time DVOA improvement in 2010, and for all the advanced metrics saying his team had rotten luck in 2012, and will improve next year from regression to the mean alone).

Also, I would hack into Youtube and erase all the videos showing the Justin Forsett touchdown and challenge.

I still would like for him to be fired, though.

by Aaron Brooks Go... :: Wed, 01/02/2013 - 10:57am

The NFL has already changed the enforcement of the Forsett rule, based on week 17.

WC Ford has never fired a coach who showed even the faintest glimmer of promise, and has even retained some who stunk like old cheese.

There are a lot of coaches worse than Jim Schwartz, and the man deserves at least one mulligan year. He did inherit an 0-16 team and got them to the playoffs fast. Besides, since Bobby Layne left, when the Lions haven't boomeranged from season to season, their other option has been putrescence. I'll take the Fontes boomerang.

by JoeyHarringtonsPiano :: Wed, 01/02/2013 - 12:23pm

I would prefer, rather than settling for the Fontes boomerang/treadmill of first-round playoff exits, the ownership try to reach the next level. Of course, when the Fords try to "think outside the box" they do silly things like hire Matt Millen, so maybe I should be careful what I wish for.

by Aaron Brooks Go... :: Wed, 01/02/2013 - 2:17pm

Thing is, last year's team was basically a 7-9 team that was lucky. This year's team was basically a 7-9 team that was unlucky. Otherwise, they had almost the same performance.

Schwartz may not turn out to be what they need, but I don't think he's done anything to get himself thrown out onto the street. Not without some damn good replacement idea in place.

by JoeyHarringtonsPiano :: Thu, 01/03/2013 - 12:23pm

"Thing is, last year's team was basically a 7-9 team that was lucky."

I don't disagree that the 2011 team had a lot of luck, but saying they were 7-9 quality is overstating it just a bit. The had a solid DVOA, had a few blowout wins, only 1 blowout loss, and pretty much matched their pythagorean win total exactly, while only overperforming their estimated wins by 0.6. A few of the unsustainable things that we should have expected to regress this year included fumble luck, third down defense, and a few other things. They all regressed far more than I expected.

But your overall point about Schwartz may still be valid. He did a good job in 2010 and 2011, but subjectively he did not seem to be on top of things in 2012...game management and preparation/motivation of the team seeming to be chief amongst them. Those were two things he used to be good at, so maybe he'll recapture the mojo in 2013, but I'm somewhat skeptical.

by zdneal (not verified) :: Thu, 01/03/2013 - 1:49pm

I can almost guarantee that any replacement coach will be no better than Schwartz and may be worse.

by Rafi89 :: Mon, 12/31/2012 - 6:15pm

3 teams in the top 10 for highest DVOA would seem to indicate a lack of parity. How do the bottom teams, by DVOA, compare to the lowest DVOA values of all time?

by Ranccor :: Mon, 12/31/2012 - 8:48pm

In my opinion, parity in the NFL doesn't mean all the teams are equal. It means that any team has a shot at greatness. When you see the top/bottom teams in the league rising and falling consistently across the league, parity is working as intended.

by RickD :: Mon, 12/31/2012 - 10:47pm

That's not really the definition of parity.

The dictionary definition would be that all teams are roughly equal, with low variance in their quality. It has nothing to do with having "an equal shot at greatness."

by Kurt :: Mon, 12/31/2012 - 11:26pm

But Ranccor is right about the functional definition of parity as applied to the NFL. When people talk about parity it's always about how many teams have won the Super Bowl or been in the playoffs in the last X years.

by The Ninjalectual :: Tue, 01/01/2013 - 12:15pm

Wouldn't it be "the Ranccor is right..."?

by Anonymousse (not verified) :: Mon, 12/31/2012 - 11:35pm

They all are roughly equal, just not in the period of one season.

by Aaron Schatz :: Tue, 01/01/2013 - 8:08pm

This year's worst teams were nowhere near as extreme as this year's best teams, either in total DVOA or in any of the three splits. The Chiefs end up as the 10th worst team ever in total DVOA. The Jaguars aren't even in the bottom 25.

The Cardinals are the 11th worst offense since 1991.

The Saints don't even make the 40 worst defenses since 1991. This year's defenses were very closely packed except for Chicago.

The Texans are the 14th worst special teams since 1991.

by Jeff Huter (not verified) :: Mon, 12/31/2012 - 6:21pm

While DVOA and conventional stats hate the Colts, they actually faired well (win/loss) against other playoff teams. They played 5 games against teams going to the playoffs and went 3/2 (Wins against MIN, GB, HOU, Losses to NE and HOU). So to say, this team can't compete in the playoffs, I think is short-sighted. In fact, I like their chances in BAL despite DVOA greatly favoring BAL.

The other teams in the AFC haven't exactly done tons better against other playoff teams. In fact, the Colts are the only team in the AFC with a winning record against playoff teams.

NE => 3/3
DEV => 1/3
HOU => 3/3
BAL => 2/4
CIN => 2/2

Both of BAL wins against playoff teams were in weeks 1 and 3. They definitely have faired well against playoff teams since.

** Forgive any errors in the records. I compiled these quickly and may have overlooked something.

by Hummingbird Cyborg :: Mon, 12/31/2012 - 7:01pm

Denver went 2-3 but that aside, here's an article that analyzes how predictive wins against quality opponents are compared to large wins against weaker opponents.


I mean, I agree that it's be foolish to say that the Colts can't win in the playoffs. I mean it was pointed out above that other teams that made the playoffs rated even worse won games in the playoffs nonetheless, but it does mean that DVOA doesn't favor them in any game that they play and DVOA correlates fairly well with future success, so it's not a good sign.

But just saying that you didn't choose a particularly predictive stat for what you're saying.

by CaptFamous (not verified) :: Mon, 12/31/2012 - 7:02pm

The concept of things like "record against playoff teams" has a lot of circular-esque logic to it. Besides the obvious (best teams = more parity, smaller sample size of games), there's the fact that your record against playoff teams is based on how they play in every other game of that season.

Case 1: If Green Bay had it's worst game of the year with the worst turnover luck (not saying they did) against Indy, then beat a bunch of other teams in normal performances, they're status as a playoff team doesn't necessarily reflect on the quality of Indy's performance in the win.

Case 2: If Minnesota loses to Green Bay, Indy's record against playoff teams drops to 2-3, as Minnesota is no longer a playoff team, and Chicago is (and they beat Indy in the 1st week), despite the fact that the Colts didn't even play in this game.

Similar weird example: Last season, lots of people were harping in the Patriots for their poor record against teams with winning records (0-2 I think) and their lack of games against said teams. However, they played 7 (7!) games against teams who finished 8-8. In addition, one of their losses was to the Giants, who finished 9-7. So in reality, there were 7 games they could have lost that would have made their record against winning teams worse, but only one game they could have won that would have made it better.

by CaptFamous (not verified) :: Mon, 12/31/2012 - 7:32pm

Interesting (and mostly unrelated) side note (and likely just as inconsequential) - If you look at point differential instead of record against playoff teams, the AFC seeds as follows (assuming I can still add in my head) (Sequence is seed, number is ranking by point differential):


And only the first two are positive (NE is something like +65, DEN is +3). Indianapolis is around -40.

by Hummingbird Cyborg :: Mon, 12/31/2012 - 8:01pm

Only the Colts have a negative point differential.

New England is +226 and Denver is +192, but Baltimore is +54, Cincy is +71, Houston is +85 and Indy is -30.

by Hummingbird Cyborg :: Mon, 12/31/2012 - 8:04pm

Oh, I get what you're saying now. Against playoff teams Denver and New England are positive in point differential.

by Revenge of the NURBS (not verified) :: Wed, 01/02/2013 - 9:27am

This is all true, but at least in regards to your case #1, any measure that factors in opponent strength suffers from the same problem. One of the big reasons the Colts rank so low in DVOA is their strength of schedule. SOS is mostly driven by how those teams did in all their other games.

by JonFrum :: Mon, 12/31/2012 - 8:19pm

If the Patriots had lost one of their two games against the Jets last year, they would have had one more win against a team with a winning record - the Jets ended up 8-8.

Cherry picking a metric will tell you anything you want it to. The season is too short to pick apart wins and losses to tell a story.

by The Ninjalectual :: Mon, 12/31/2012 - 9:33pm

Ha, this point is brilliant! Well done.

by CaptFamous (not verified) :: Mon, 12/31/2012 - 9:45pm

True, though it also would have given them one more loss, so they would have been 1-3 instead of 0-2. Technically better, but yeah, the whole point is that once you start getting into all these relative statistics, things get weird quick.

by Hummingbird Cyborg :: Mon, 12/31/2012 - 7:01pm

I'm curious how the Giant's variance rates.

by Jerry :: Mon, 12/31/2012 - 7:09pm

I wonder if these would be all-time top teams using the old multi-year baselines instead of the single-year one.

by The Hypno-Toad :: Mon, 12/31/2012 - 7:22pm

That's what I was trying to ask in the post below, but instead of my rambling idiocy, expressed concisely and intelligently.

by The Hypno-Toad :: Mon, 12/31/2012 - 7:13pm

Wasn't there a change in methodology during the off season? Something about normalizing to where average dvoa will always be zero for a given season? I am doing a very bad job of remembering/explaining what i'm talking about, but if there was a significant change in how dvoa is distributed/calculated, could that lead to teams popping up on "all-time" lists that would not have shown up on those lists under the old system?

by intel_chris (not verified) :: Tue, 01/01/2013 - 5:14pm

I think the change in metrics (zeroing each seasons averages) should be considered when comparing teams from different years and that change renders such comparisons unreliable (incorrect). I think a professional statistician could even explain why this feature (assuming the mean of two disjoint samples is the same) distorts the underlying model.

For example, I would take (bet on) the 1998 Broncos (not even on the list) against the 2012 Broncos, not that this effects my feeling as a fan that this year's team is one of the better ones and a hopeful sign for the coming years. Of course, since we can't actually pit the two teams from different years against each other, other than in perhaps Madden or some other simulation, we certainly cannot directly validate the argument.

Still, in my humble opinion, there are differences in years that go beyond the evolution of the game. Something about this year caused the best teams to get separated substantially from the pack. There seems to be similar effect at the bottom also. I will leave it to the rest of you to explain what the cause of this effect is. However, I don't think that the reason is that the best teams of this year are simply among the historical best, unless you can back that up by either the old measurement that didn't introduce this statistical quirk into the numbers, or by some other independent confirmation.

by Aaron Schatz :: Tue, 01/01/2013 - 8:02pm

Actually, the change from DVOA v6.0 to DVOA v7.0 did very little to change teams' total DVOA ratings. What the normalization did was allow us to correct for the changes in the offensive environment over time, which makes it easier to compare offenses or defenses or specific positions without saying that Joe Flacco is better than John Elway simply because offensive levels are higher now than they were a dozen years ago. The average quarterback right now completes 60.9% of passes, with TDs on 4.3% of passes and INTs on 2.6% of passes. In 2000, those numbers were 58.2%, 3.7%, and 3.3%. Maybe some of that has to do with the quality of quarterbacks in today's game, but a lot of it has to do with rules changes and the development of certain offensive schemes. I don't think it is ridiculous to say that the 1999 Rams had a better offense than the 2012 Lions, but by the old version of DVOA, the 2012 Lions would have a higher offensive DVOA than the 1999 Rams. With the current version,they do not.

For each year, however, remember that offensive DVOA goes up the same amount that defensive DVOA goes up. There are slight differences for each team but in general, total DVOA ratings now are the same as they were before. For an example, check out the article where I first announced the new version of DVOA:


The exact same 10 teams that were the top 10 by DVOA v6.0 are also the top 10 by DVOA v7.0, not counting 2012 teams of course. There are just some slight changes in the order.

Remember also that the DVOA rating for the 1998 Broncos includes four games with Bubby Brister as the starting quarterback.

by The Hypno-Toad :: Tue, 01/01/2013 - 8:59pm

Thanks for the clarification, Aaron.

by intel_chris (not verified) :: Tue, 01/01/2013 - 10:43pm

Yes, thank you that's actually an interesting point. So, you are saying without the renormalization the current teams would actually fare even better in comparative DVOAs. The fact that VOAs can be used to substantiate that comparison. I'll take that as independent confirmation.

I had missed the earlier article on DVOA 7.0 and it's discussion. While it still seems statistically suspect to me that normalizing means should improve the comparability, I think I am making an error by assuming that they are drawn from a uniform pool and neglecting the fact that the changes in the rules and thus how the game is played has a major effect. I should not be surprised that you have put significant effort into evaluating the change before implementing it.

by Aaron Schatz :: Wed, 01/02/2013 - 11:06am

I'm not saying that without re-normalization, the current teams would actually fare even better in TOTAL DVOA.

What I'm saying is that without re-normalization, the current teams would fare better in OFFENSIVE DVOA but would be worse in DEFENSIVE DVOA.

by Karl Cuba :: Tue, 01/01/2013 - 10:09pm

My take on this is that the teams at the top this year look so good because the league is deluged by dross. There are an unusual number of crappy teams that make the better teams look good.

by Eggwasp (not verified) :: Wed, 01/02/2013 - 4:26am

You could have said that in the 80s. Those Bears had 6 sure wins in the NFC Central each year.

The main reason why we consider modern teams to be not as dominant as the great teams of the past is selective memory. Those blowout losses by great teams referenced above aren't part of the DVD collection. Our memories remember their great performances, capped usually by the SB wins, they don't remember the scrappy wins, the lucky wins, the disappointing losses. In ten years time, we'll have a similar selective memory about whoever takes the title this year.

by Karl Cuba :: Wed, 01/02/2013 - 4:45am

I'm not trying to make the deluge of dross case for the whole decade, just this year. Maybe it's a hangover from the increased player movement following the labour dispute (I think players tend to produce better when allowed to become acclimatised to one scheme) but it seemed like half the league was just awful. Take the harrowed AFC, there were three poor teams in the west, two in the South, two in the East and the miserable rabble that is Cleveland.

Aaron says that this years' poor teams weren't historically bad but that could be as a result of a plethora of dreck producing a higher baseline.

by theslothook :: Wed, 01/02/2013 - 5:55am

Im curious Karl, is there a particular part of the last decade you felt was much better than this year? These kinds of questions are impossibly difficult because we simply cannot separate quality from poverty.

There were no 7-9 or 8-8 teams. There were also very few 2-14. However one chooses to define parity, this year featured a number of good and balanced teams I think. Maybe the afc was watered down, but it features two extremely powerful teams while the nfc is full of treacherous clubs and plenty of depth to go around.

by LionInAZ :: Wed, 01/02/2013 - 5:51pm

I dunno, Karl. Is this year really worse than 2010, when *every* team in the NFC West had a losing record, there were two poor teams in the NFCE and NFCN, plus the miserable rabble that was Carolina? Some would say that year was much more shameful for producing a 7-9 division winner.

by Karl Cuba :: Wed, 01/02/2013 - 6:25pm

I'm not sure but 2010 is one of the years that also produced two of the best ever teams by DVOA.

by LionInAZ :: Wed, 01/02/2013 - 6:47pm

And the 2nd-worst playoff team ever by DVOA. Compare that to 2012, with 3 of the best ever teams by DVOA and the worst ever 11-5 team. I don't see much difference, except that in 2010 eight of the worst ten teams by DVOA were in the NFC, including all four NFC West teams.

by Karl Cuba :: Wed, 01/02/2013 - 8:45pm

Not sure what you're driving at, unless you're saying that 2010 was full of crappy teams too.

by Karl Cuba :: Thu, 01/03/2013 - 12:43pm

Just noticed from listening to the BS podcast that 2004 also features two top teams and Aaron quotes that year as having a pile of awfullness too, in the NFC again. So those three years account for seven of the top 12, seems like quite a lot. Is top historical DVOA driven by having a plethora of bad teams that drive up decent teams in comparison?

If you look at the other five teams; 07 Pats, 91 Redskins, 96 Packers, 95 49ers and 92 Cowboys, then you have a pretty good list of the teams that most people would regard as some of the best of the last twenty years.

by intel_chris (not verified) :: Wed, 01/02/2013 - 3:15pm

My feeling is something like this. However, I suspect there is some kind of analysis that could make the meaning of this crisper and clearer. Is it that was have a few worse bottom teams that are dragging everyone down? Is it that the general level of play on defense is worse than it used to be, as in our offenses simply generating more successful plays on average. Are certain kinds of plays more successful than they have been in the past? Are there fewer good teams, so the ones which are good seem to be so much better? Has the way that penalties are called changed the effectiveness of certain plays?

My reading of the DVOA 7.0 announcement posting is that the FO folks have noticed a long term trend for DVOA scores for the passing game to improve over time. That isn't surprising because the NFL has clearly found that the passing game is more exciting and have tried to encourage teams to use it by altering the rules to make it more effective. The NFL has also encouraged "parity" so that more teams should be competitive, i.e. the Giants and Cowboys (and other similar teams) cannot simply buy their way to titles--at least not as easily, since there clearly are still haves and have-nots. Those changes could have affected the questions I asked at the beginning in different ways. It would seem possible to determine what effects there have been. One of the things I like about this site, is that they do seem to ask and address these questions.

Finally, over time sports competitors simply improve, standing on the shoulders of giants. Once upon a time a 4 minute mile was considered an impossible achievement. Swimmers continually break records. Skaters regularly do tricks that would have once earned them an Olympic medal. Similarly, Joe Montana running the West Coast Offense was once a unique quarterback. Perhaps these days, he would simply be replacement level.

by ChrisS (not verified) :: Wed, 01/02/2013 - 4:44pm

I think our perception of teams have changed since the eighties because free agency has destroyed the possibilty of dynasty teams. For exampleyou e great Steelers teams were remembered because they had continuity and we almost knew Swan and Green and Ham and Franco and ..... and considered them great players so the Steelers were a great team because they had great players.

by ChrisS (not verified) :: Wed, 01/02/2013 - 4:48pm

No more responding from my phone. Tpyo Free Version. I think our perception of teams have changed since the eighties because free agency has destroyed the possibilty of dynasty teams. For example the great Steelers teams were remembered because they had continuity and we all knew Swan and Green and Ham and Franco and ... and they were all thought of as great players so the Steelers were a great team because they had great (well known) players.

by MatMan :: Mon, 12/31/2012 - 7:24pm

"...and went 3-9 in games decided by a touchdown or less."

Does that mean "6 points or less," or "7 points or less?"

by JoeyHarringtonsPiano :: Mon, 12/31/2012 - 7:51pm

In this case, they mean 8 points or less (since the Lions lost by 8 to San Fransisco, and they seem to be counting that). Maybe a more accurate term would be "one possession games".

I would argue that the San Fransisco game should be taken out of this discussion, since at no point was Detroit even close to winning or tying.

by Aaron Brooks Go... :: Wed, 01/02/2013 - 11:01am

I seem to recall that game started out at 0-0.

by JoeyHarringtonsPiano :: Wed, 01/02/2013 - 12:20pm

Of course all games start 0-0, but I meant towards the end of the game, which is what we're talking about when defining "close games".

by Kal :: Mon, 12/31/2012 - 7:27pm

One issue that comes up when talking about DVOA (and any opponent-weighed measure) is how 'telling' it is when you crush weak foes. This year we should be able to see that somewhat in spades; Seattle had one of the hardest schedules in the league and is highly ranked; Denver had the second-easiest schedule and is also highly ranked.

My gut tells me that Denver's a bit overrated, but also that Seattle is largely feasting on a few late-season big wins that mean more to DVOA than they should. But it'll be interesting to see.

by G (not verified) :: Mon, 12/31/2012 - 7:49pm

I think the parity this year has been so strong; which is what makes these top 3 teams stand out on the all-time list of best DVOA ever recorded.

We're talking about a season where some of the all-time great records were threatened. AP goes for 2,000 yards, Rice's record falls, and some of the most dynamic rookie quaterbacks ever come into the NFL and light it up.

Not to crazy to me. Model affirmed ;)

by G (not verified) :: Mon, 12/31/2012 - 7:52pm

Seattle didn't just come out of nowhere; they were routinely ranked in the top... 6? Before they went seahulk. The other thing about Seattle's late season dominance is it is clearly tied to a new offensive gameplan, so it's not as if there were fluky reasons for those games.

by Kal :: Mon, 12/31/2012 - 8:01pm

The Seahawks weren't rated that high until the midseason. Really, until the Bears game they were simply a good but not great team.

by Perfundle :: Mon, 12/31/2012 - 8:50pm

Well, the Bears game was after midseason; they were already a great DVOA team when they got to the Bears game. Here is their weekly DVOA ranking:

-4.1%, 16th, at Arizona
12.4%, 12th, Dallas
11.5%, 12th, Green Bay
15.0%, 11th, at St. Louis (?)
21.7%, 10th, at Carolina
23.5%, 7th, New England
21.7%, 8th, at San Francisco
20.3%, 9th, at Detroit
29.3%, 6th, Minnesota
33.1%, 3rd, NY Jets
34.4%, 4th, bye
30.7%, 4th, at Miami
31.9%, 4th, at Chicago
38.8%, 2nd, Arizona
38.5%, 2nd, at Buffalo
41.1%, 1st, San Francisco
38.3%, 1st, St. Louis

The large whack of DVOA increase was not because of the Chicago, Arizona, or SF games; they were top 5 and above 30% long before then.

by Kal :: Tue, 01/01/2013 - 3:03am

Actually we are both right. They went from being 20% to almost 30% after the Minnesota game (I have no idea why) and then another 10 point jump after the Chicago game. Wierd.

by Insancipitory :: Tue, 01/01/2013 - 3:13am

As I recall, in the second half of the Vikings game the Vikings hardly even touched the football, and the Seahawks just sort of effortlessly strolled away with it.

The Chicago game gives a big opponant adjustment to an offense that had iirc a 94, a 97 and 80 yard TD drives.

Then Arizona gets creamed, historically, in every phase of the game, with an adjustment from what DVOA considers to be the then #2 defense, without variables to account for confusion or quiting.

Just chains of big chunk plays against opponants DVOA respected

by EricL :: Wed, 01/02/2013 - 4:00pm

Correct regarding the Min/Sea game: With 10:24 left in the 3rd, Seattle got the ball on their own 28, up 20-17.

Of the 25:24 remaining in the game, Seattle had the ball for 20:12, spread over four drives, including the final 5:27 ending in a kneeldown on the Min 9.

by formido :: Mon, 12/31/2012 - 8:52pm

More recent results are more predictive so suggesting it's unfair that Seattle is being rated so highly for recent results is specious. If we were talking about a random results generator weighted smoothly throughout the season that'd be one thing, but we're talking about dynamic systems made of people and context matters. Anyone who follows the Seahawks knows that their rookie QB has played (has been allowed to play) much, MUCH better since the first month of season. Wilson leads the NFL in passer rating and QBR since week 6, I believe. Seattle implemented the read option at just the Chicago game and no one has stopped it, including St. Louis when Seattle started finally running those sets in the second half. In sports, some teams get better and some teams get better faster than everyone. This year that's Seattle.

by Rowdy Roddy Piper :: Mon, 12/31/2012 - 8:54pm

Any model that tells me last year's Falcons team was better than this year's could use some adjustment. Offense is better. Defense is definitely better.

by The Ninjalectual :: Mon, 12/31/2012 - 9:28pm

You forgot to mention which subjective ranking system is way better than this, and you left out your unrelated comment with chat-acceptable spelling entirely.

by bill (not verified) :: Mon, 12/31/2012 - 9:37pm

Hmm..the three worst teams to make playoffs (acc. to DVOA) all win.?
Still nothing wrong with your ranking system. You have Bal. rated way to high.
Indy WILL upset them this week. Making it 4 /4 for your worst rated playoff teams...#fadeFO

by jonnyblazin :: Mon, 12/31/2012 - 11:39pm

"You have Bal. rated way to high.
Indy WILL upset them this week. Making it 4 /4 for your worst rated playoff teams"

If its a certainty, then go bet every cent you own on it. You'll more than double your worth.

by Cuenca Guy :: Tue, 01/01/2013 - 6:30pm

Yes, because the 7-9 Seahawks (2010), 8-8 Rams (2004), and 9-7 Cardinals (1998) were such great teams. Any team can win on any given week. Indy might win this week, but the fact remains that this isn't a team that under normal circumstances would be 11-5.

Also, if you're going to bash DVOA, please use the format listed in the article.

by anderson721 (not verified) :: Mon, 12/31/2012 - 9:39pm

Re the Colts: Without looking, I'm going to guess that no other playoff team got mauled by a losing team, as the Colts were by the Jets. And considering the Pats beat the Colts by 30 something points, it's very hard to argue that the Colts had better success against playoff teams.

by Carlos12 (not verified) :: Mon, 12/31/2012 - 9:53pm

Re the Colts: Without looking, I'm going to guess that no other playoff team got mauled by a losing team, as the Colts were by the Jets. And considering the Pats beat the Colts by 30 something points, it's very hard to argue that the Colts had better success against playoff teams.

The Ravens lost to the Eagles. I'm guessing no other playoff team lost to the Eagles.

The Patriots lost to Arizona at home. I'm guessing no other playoff team did that.

And considering you're talking about a 1 game sample size, it's impossible to even take your post seriously.

by RickD :: Mon, 12/31/2012 - 10:54pm

The argument was "no other playoff team got mauled by a losing team."

It wasn't "no other playoff team lost to team X".

The Patriots lost by 2 points to Arizona when a last-second FG attempt was missed. The Ravens lost by 1 point to the Eagles. Neither loss qualifies as "getting mauled."

If your point is "some playoff teams lost to different teams," that's neither terribly interesting nor relevant to the point you're replying to.

by Andrew Potter :: Tue, 01/01/2013 - 2:53am

I'm going to guess that no other playoff team got mauled by a losing team, as the Colts were by the Jets.

The 7-9 Buccaneers plundered the playoff Vikings in Minnesota, 36-17.

by Bobman :: Wed, 01/02/2013 - 3:56am

Great use of plunder.

I might as well pile on and point out that the Colts lost to Jax, though it was a last minute 80-yard reception that gave the Jags half of their season win total. The Colts had some stinkers, especially early, that's for sure. But the loss to Chicago, while not a playoff team, was to a team that was all but unbeatable in weeks 1-8. That team looked incredible.

I don't know where the 95 Colts fit in DVOA-wise, but the 2012 version feels a lot like them. The comeback wins are not just a product of the QB, but somehow the D that had been pushed around all game has tended to get stops at the end of all those close games, allowing Luck and Co to do their thing. Will they go to the AFCCG and lose because the ref missed Kordell Stewart stepping OB bfore catching a TD pass? (not that I'm bitter...) Not likely. But a first round win looks like a toss-up to me, if not to DVOA.

by Perfundle :: Wed, 01/02/2013 - 1:36pm

"But the loss to Chicago, while not a playoff team, was to a team that was all but unbeatable in weeks 1-8. That team looked incredible."

That team went 6-0 against teams that didn't have a winning record, including barely beating Detroit and Carolina at home, and crashed and burned against the only great team they faced, though crushing Indianapolis, even if it was Luck's first game, was pretty impressive. Their losses in the second half of the season? All to playoff teams. Looked at that way, they were incredibly consistent in their performance.

by Spielman :: Fri, 01/04/2013 - 12:46pm

"The comeback wins are not just a product of the QB, but somehow the D that had been pushed around all game has tended to get stops at the end of all those close games, allowing Luck and Co to do their thing."

This seems like something that happens a *lot*. Team gets reasonably soundly beaten for most of the game. Then, late, the team that is winning turtles up on offense, and goes ultra-conservative, leading to some short possessions. Then they go into a prevent style defense as well, and all of a sudden the team that has been unable to move the ball the whole game is driving with seeming ease. And then, boom, it's suddenly a close game.

by JoeyHarringtonsPiano :: Tue, 01/01/2013 - 7:08am

I'm going to guess that no other playoff team got mauled by a losing team.

Not true. Doesn't happen often, but it happens. Off the top of my head, I can remember the 1990 Bears finished 11-5, but lost to the 6-10 Lions 38-21 (The Bears were down 38-14 before scoring a garbage time touchdown. And just last year, the Tebow-Broncos (yeah, I know, not the best example) lost to the 6-10 Bills 40-14.

by Arkaein :: Tue, 01/01/2013 - 10:58am

And in 2010 the 14-2 Patriots lost to the 5-11 Browns 34-14 at mid-season. I remember this particular nugget because 20 points was the same as the combined margin of all 6 of GB's losses that year.

by dmstorm22 :: Tue, 01/01/2013 - 2:34pm

The 2006 Patriots lost 21-0 to the 5-11 Dolphins, and the 2003 Patriots even more infamously lost 31-0 to the 6-10 Bills.

by JIPanick :: Tue, 01/01/2013 - 2:50pm

Pretty sure he meant this year, not ever.

by Andrew Potter :: Tue, 01/01/2013 - 3:37pm

As am I. His guess is still wrong though.

by Ben :: Mon, 12/31/2012 - 9:54pm

I know I'm not following the proper format, but I have a hard time thinking of the Colts as worse than the 5 or so teams in front of them at this point in the season.

I know about the point differential. The Colts got blown out against the Bears week 1. When the Colts had a bunch of rookies, and Chicago had it's full defense. They also got blown out against the Pats (not that that is necessarily uncommon) with 3 return TDs. The certainly laid an egg against the Jets and played the Chiefs and Jags closer than they should have. It just seems like the by virtue of the Colts playing a lot of bottom feeders, there is a feedback loop where the Colts are bad for not beating them by a lot, but those teams don't get any credit for playing well, since it was against the Colts...

I have no problem with them being the lowest rated team in this years playoff, that's hard to argue against. It does seem odd that 3 other teams that have been worse than them historically all won their first game. If the Colts do too, it seems like there might be an issue with at least weighted DVOA, when it comes to playoff time.

by Hummingbird Cyborg :: Tue, 01/01/2013 - 10:35am

It's a tiny sample size, so I don't think it means much for DVOA. DVOA still correlates much better than win loss at predicting future winners, but we all know that sometimes (Reasonably often) the weaker team wins a game that you wouldn't expect them to win.

I mean, as a Broncos fan, I didn't expect the Tebow Broncos to beat the Steelers last year because the Steelers were clearly the better team by a large margin. It didn't mean that the Broncos were better than the Steelers, but that they played better for that game than their average game and the Steelers played worse. Sometimes it also means that maybe a team got a bit lucky with things like recovering fumbles.

by Paddy Pat :: Tue, 01/01/2013 - 12:38pm

I've always assumed that there's a big psychological factor here. First of all, the crappy team usually feels like its playoff birth is a sort of miracle (how the hell did that happen?). This makes coaches less conservative, more inclined to go for forth downs, try for big plays, empty out the playbook, etc. Because hey, what have they got to lose? They're playing with house money. The opponent is also likely to look on the game as an easy win and thus underrate the weak team. When the stronger team falls behind there's a mixture of disbelief and complacency, and frequently a failure to adjust--Hey, we're better than they are; let's keep doing what we're doing and things will settle out. That was definitely the case last year with the Steelers. LeBeau just couldn't bring himself to adjust to Tebow's deep ball. Finally, the weak teams often play with an enormous chip on their shoulders because they know there's a funny sort of stigma that goes with being the team that doesn't belong.

I'm not saying any of that is predictive, but it seems to me to explain why the worst teams often win One playoff game--just one, mind you, and then get creamed the following weak.

by theslothook :: Mon, 12/31/2012 - 9:54pm

I didn't realize this because no one is talking about him, but by conventional statistics, does anyone realize how insane drew brees' numbers are and then the fact that he DIDN't MAKE THE PROBOWL? Despite completing only 63%, he again threw for over 5 thousand yards and led the league with 43 tds. If people really believe in conventional statistics, then Drew Brees is the mvp of football.

by RickD :: Mon, 12/31/2012 - 11:02pm

I think what kills his candidacy more than anything else are his 19 interceptions, which equals the sum of Brady's (8) and Manning's (11).

FO has Brees at #3 among QBs, but he's well behind the other two. He threw a lot of picks, and they hurt the Saints a lot.

by theslothook :: Mon, 12/31/2012 - 11:44pm

Oh I happen to hate most qb stats to begin with so I wasn't trying to pump up Brees or anything. I was merely showing how fickle fans and the media are when it comes to individual accolades or Qb lists. Drew brees by all rights had the same type of season he always has.

Drew brees throws alot, which helps him accumulate a lot of yards and td passes(i read a stat that showed he throws the most passes on the goal line than any other qb in the nfl).

by Hummingbird Cyborg :: Tue, 01/01/2013 - 12:38pm

People realize that quarterbacks throw more in losses and rack up yardage that way. I think the perception is that's what happened in New Orleans.

by dbostedo :: Tue, 01/01/2013 - 1:15pm

Even given that, I'd guess that if they Saints were 12-4 and had won their division, and Brees had identical stats, you'd hear him talked about in the MVP race.

by LionInAZ :: Tue, 01/01/2013 - 4:52pm

Brees threw 681 times in 2010, 678 times in 2011, and 693 times this year. It's the offense, not the losses.
Strangely, he threw a lot fewer passes in the Super Bowl year.

by Gomer_rs (not verified) :: Wed, 01/02/2013 - 3:23am

Of course Drew Brees threw a lot fewer passes in the super bowl year. Even overly pass crazy teams run the ball with the lead.

by DRohan :: Wed, 01/02/2013 - 2:46pm

The Saints went 13-3 in the Super Bowl season, and followed that up with 11-5 and 13-3, respectively in 2010 and 2011. I'm guessing they weren't playing from behind all that much more often in those two seasons.

by LionInAZ :: Wed, 01/02/2013 - 6:01pm

Exactly. I might also have included that Brees threw 645 passes in 2008. Throwing a lot of passes is the norm for Drew Brees, not the exception.

by DisplacedPackerFan :: Mon, 12/31/2012 - 10:32pm

So all the NFC North teams had positive DVOA, has a division finished the year with all the teams on the plus side before?

by JoeyHarringtonsPiano :: Tue, 01/01/2013 - 1:11pm

Kind of on the same subject, did the Lions this year set the record for fewest wins for a team with a positive DVOA? I'm sure there have been a few teams with 5 wins.

by LionInAZ :: Tue, 01/01/2013 - 5:22pm

The Lions are the third 4-12 team to finish with positive DVOA, and not even the best. The 91 Chargers finished with +1.6% DVOA, while the 96 Ravens had +5.1%. That Baltimore team was 1st in offense and 29th in defense. No team with less than 4 wins finished positive.

by JoeyHarringtonsPiano :: Wed, 01/02/2013 - 6:55am

Ah, thanks for looking that up. I remember that Chargers team losing close game after close game. The next year, Bobby Ross showed up and bad luck left, and they went 11-5.

That '96 Ravens team was bad, but man were they fun to watch.

by Anonymouss (not verified) :: Wed, 01/02/2013 - 2:30pm

Not to change the subject, but for the colts haters, they beat 3 of those 4 nfc north teams.

by theslothook :: Mon, 12/31/2012 - 11:47pm

I also hate the interception stat. It tends to skew things so badly in so many ways. Guess what interceptions correlate strong with? Throwing a lot of deep and medium passes. Guess what else? Throwing when you're behind in the score?

Now, I fully recognize that throwing ints might have led to you having to throw from behind in the first place, but that aside, both of those factors are really circumstances that may be beyond the qb's control. After all, your defense shares a big responsibility for why you are behind and the scheme/receivers/running game also play a big factor in why you have to throw medium and deep so often.

by Perfundle :: Tue, 01/01/2013 - 2:07am

Interceptions gotten by throwing when you're behind in the score could very well be cancelled out by touchdowns gotten by throwing when you're behind in the score because the opposing defense is playing prevent, and getting those chances because the opposing offense is either complacent or trying to run out the clock; see the Patriots-49ers game and last year's Saints-Packers game for examples. It might or might not come out even in the end, but don't make it like throwing when behind has to lead to poorer stats.

by Anonymous1 (not verified) :: Tue, 01/01/2013 - 2:00pm

NE's offensive success against SF in the second half had very little to do with complacency.

It was three parts figuring them out (which they had already done long before SF ran out to a big lead, they just kept putting the ball on the ground) and one part Justin Smith leaving.

by Aaron Brooks Go... :: Wed, 01/02/2013 - 11:05am

That would be why SF switched to a zone after smothering NE with man for the entire game?

by Anonymous1 (not verified) :: Wed, 01/02/2013 - 8:07pm


First off, if you check the game log, you'll see that NE was already moving the ball pretty comfortably, but put the ball on the ground after on long drive and had another where they dropped an easy first down. Credit SF for creating the turnovers, but only in the first quarter did they really "smother" NE.

Beyond that, the whole prevent thing is wildly overstated. A few analysts who reviewed the tape said that they didn't see much in the way of changes from SF, and certainly they weren't playing prevent once NE had gotten the ball back only down 31-17.

As I said before, NE's second half success was two parts figuring them out (which had already happened) and one part Justin Smith leaving the game. The zone/prevent discussion is misguided.

by Karl Cuba :: Wed, 01/02/2013 - 8:42pm


by Anonymous1 (not verified) :: Wed, 01/02/2013 - 10:38pm

So someone with a valid opposing view is a troll?

Man, the arrogance on this site is astounding. I've enjoyed FO since pretty much day one, and the comments used to be a place a real insight. Now it is pretty much a circle jerk of people who like to feel intellectually superior.

by Will Allen :: Wed, 01/02/2013 - 10:56pm

You may wish to examine your tone in your last post, and given yourself an irony-check. I say this as someone who is sympathetic to your view, at least in regard to the centrality of Justin Smith to the Niners' defensive performance.

by Anonymous1 (not verified) :: Wed, 01/02/2013 - 11:09pm

Mea Culpa on the tone, then. It wasn't my intention to ruffle any feathers (initially, at least ;-) ) just to try and point out that a lot more went into NE's offensive explosion that simply "SF stopped trying".

by Karl Cuba :: Thu, 01/03/2013 - 9:30am

I'm not sure that backing off into a safe zone is the same as stopping trying. I thought the niners miscalculated and got worn down by the Pats' no huddle and rapid tempo. They thought their lead was safe but that it wasn't is also a function of the Pats' abilities.

by Anonymous1 (not verified) :: Thu, 01/03/2013 - 10:42am

Fair enough. I just want to point out that the dominance in the first quarter seems to overshadow the fact that NE was finding success much earlier than it became clear on the scoreboard.

I've also heard whispers that NE feels like they found something for a possible rematch that makes them very confident in their chances.

by Will Allen :: Thu, 01/03/2013 - 11:11am

It pains me to say it, but I fear that unless Justin Smith can be Justin Smith, the Niners aren't getting that far. It's what I hate about football the most, beyond the toll the injuries take on the player's lives; who gets injured and when they get injured plays a gigantic role in who experiences ulitmate success.

by theslothook :: Wed, 01/02/2013 - 9:06pm

Actually, I charted this game. The big difference was a few things but notably, all of NE's td drives came off big plays. It should be noted that throughout that comeback, there were many times the pats were stuffed on first and second down before converting on third downs or 4th downs. They converted three 4th downs in their comeback- drives that might have ended otherwise. The big plays meant they didn't have to drive the length of the field. Again, we can read into these plays however we want, but it isn't like the machine of NE just ripped apart the defense. It was close, even the big plays were all very tough plays to execute. Two went to Lloyd despite being in terrific coverage. 1 Went to Houamanawanui of which he almost lost and was again in tight coverage. Another came on a ticky tack Pi penalty in the end zone. On two of the drives where they had the ball at the goal line. Again, these were tough plays which the pats executed but they weren't exactly specifically designed where brady looked off the defense and the receiver burned the corners.

Finally- the other big highlight was the fact that the 49ers pass rush virtually disappeared in the second half. ESPN's logs credit the 49ers with I think 2 or 3 total pressures before the second to last drive, both of which I remember were schemed by the 49ers to get free rushers in. Besides those, brady had forever and a day to sit back and wait for guys to get open. And finally, on the second to last drive, Ray Mcdonald annihilated Volmer and Wendell to get two back to back sacks(one went to Francois).

Again, the big difference between the 1st half and second half were big plays and pass rush. Hardly Ne figuring out sf and clobbering them.

by Anonymous1 (not verified) :: Wed, 01/02/2013 - 10:53pm

I don't contest anything you wrote, just your concusions.

The lack of pressure was primarily due to NE adjusting to SF's defense with a dash of Justin Smith's absence added in. I agree that big plays were a major factor, but that information doesn't contradict my post, it supplements it. Yours just goes into more detail about the approach.

I'll agree that the 4th down plays are an important factor. Had NE not been in such a dire situation, they may have punted on a few occasions. However, they may also have approached things differently if they didn't view every possession as 4 down territory.

The bigger issues that gets overlooked though, is that NE was moving the ball as early as their first possession in the second quarter. NE had a 16 play, 62 yard FG drive, followed by a 3 and out that was only short because a wide open Lloyd dropped an easy first down. NE's next possession was an easy 50 yard drive in 4 plays that ended with a Ridley fumble. 3 drives, 3 times NE moved the ball well (or only didn't due to their own error).

SF overwhelmed NE at first. But NE had caught up only to be undone by a couple mistakes, which SF did well to capitalize one. "Figured out" is simply shorthand for all of that.

by Karl Cuba :: Thu, 01/03/2013 - 9:23am

The disappearing 49er pass rush has slightly more to it than Justin Smith's absence. Aldon Smith has been struggling with a bad shoulder for over a month now and he wears down as the game goes on, though missing his bro from another mo does make it easier to chip and double him.

by Anonymous1 (not verified) :: Thu, 01/03/2013 - 10:43am

Thanks for the input, I wasn't aware of that.

by coremill (not verified) :: Thu, 01/03/2013 - 12:55pm

Fatigue played a big factor too. The Niner offense went 3 and out 3 straight times and the SF defense ended up being on the field for 92 plays. During NE's comeback NE ran 35 plays in just over 18 minutes of elapsed game time. The pass rushers just started to run out of gas in the second half.

by DRohan :: Wed, 01/02/2013 - 2:50pm

I think using the eye test it was evident that Brees was not as sharp this season as in the past. I don't think the increased INT's were driven by situational effects.

by rageon :: Tue, 01/01/2013 - 1:31am

It really doesn't feel like this was the Bears best defense in the past ten years, does it?

by Chip :: Tue, 01/01/2013 - 10:53am

My thought exactly.

That said, they were so consistent in terms of success rate all year. Advanced NFL stats also has them as the #1 DEF by a decent margin as well. The #1 ratings on both sites were maintained week-to-week as well.

This team will miss Lovie.

by Steve in WI :: Wed, 01/02/2013 - 12:24pm

I don't know. Looking back at the whole season, I can believe this was a great Bears defense. The only loss I'd put just as squarely on the defense as on the offense would be San Francisco...well, maybe Seattle, too. Every other loss, I'd argue that the defense either had a great game (Houston) or that the offense played so poorly that the defense was put in an impossible position (see the 2nd Minnesota game, where one touchdown was a returned Cutler interception that was in no way the fault of the defense, and another was set up by a Cutler interception that set Minnesota up in the red zone).

I think firing Lovie was ultimately the right move, but I'm basing that largely because I feel like a defensive meltdown is inevitable given how many great players are aging. Might as well bring in someone who can lead a good offense.

by Ken Hoygan (not verified) :: Tue, 01/01/2013 - 2:12am

Bengals vs San Fran. The two are intertwined, quantum entanglement.

by Paul M (not verified) :: Tue, 01/01/2013 - 3:27am

We all know the recent low seed/WC history but I would be shocked if anyone other than DEN or NE comes out of the AFC. Indy is simply a substandard playoff team; the Bengals not a whole lot better; Baltimore living off the reputation of a fading defense; Houston nowhere near the team they were two or three months ago.

The fascinating aspect of this year's playoffs are the four NFC teams that might emerge to face off against Manning or Brady-- likely the two best QBs since Montana and very possibly two of the best five (or four, or three) QBs in history (and given the venue-- New Orleans-- don't you think Manning makes the most Hollywood sense??)

GB is the one NFC team that could win a shootout; SEA and SF are the two defenses that might be able to pressure Manning/Brady enough to cause a mistake or two or throw off the passing game; I can't for the life of me figure out how Atlanta can win, but maybe we'll all be fooled.

Would expect the AFC winner-- despite this being a pretty solid NFC dominant season, and the weird NFL wheel turning to the NFC (2007, 09-11 winners)-- to be at least a FG favorite but have to admit the game will be quite enticing no matter which of the 8 matchups occurs. Brady would be playing for greatest QB ever and maybe Belichick same for coach; Manning to surpass his boss and take down the #3 spot historically (or at least post Walsh and the installation of the West Coast offense); Rodgers to clearly occupy the top seat post-Brady/Manning; Ryan to gain true elite status; and Wilson and Kapernick just to have some fun. Can't wait.

by jonnyblazin :: Tue, 01/01/2013 - 12:03pm

I think the Ravens and Bengals could do some damage to the top teams. The Ravens beat NE already this year. And the Bengals have the Giants blueprint of having a deep and talented D-line. They're also white hot (18% weighted DVOA vs. 6% total DVOA, plus the 3rd ranked D in weighted DVOA). Both are suited to cold weather, whereas the aerial attacks of the Pats and Broncos could suffer in cold/wind.

The Ravens are also the healthiest they've been all year, especially on defense. They should have Lewis and Ellerbe at full strength in the middle, Ngata has fought through his MCL strain and shoulder injury and seems to be rounding into form, Suggs took last week off to heal his biceps and achilles, and Pollard will return from his rib injury. It also seems like their OLB Kruger and Upshaw have found their roles, with Kruger becoming a very good pass rusher and Upshaw becoming a great run stopper. At no point in the season were all these players healthy and playing together.

by Anonymous1 (not verified) :: Tue, 01/01/2013 - 2:03pm

The Ravens were dominated in just about every facet of that game, other than the officiating. Not saying they are incapable of winning, but pointing at that game as evidence that they could isn't very compelling.

by jonnyblazin :: Tue, 01/01/2013 - 2:35pm

"The Ravens were dominated in just about every facet of that game, other than the officiating. Not saying they are incapable of winning, but pointing at that game as evidence that they could isn't very compelling."

OK, lets look at some of those facets.

Yards per Play:
Ravens 7.7
Pats 5.1

Ravens 503
Pats 396

Ravens 14/135
Pats 10/83

Ravens 70.6%
Pats -20.5%

I think you're memory is maybe playing tricks on you. Name one way in which the Pats dominated the game. Flacco threw a pick, so I guess the Pats dominated the turnover battle at 1/0.

by Anonymous1 (not verified) :: Wed, 01/02/2013 - 8:15pm

My memory is not playing tricks on me.

When teams have three drives extended due to erroneous penalties, they tend to have inflated stats. Here are just a few of the wrong (not questionable, outright wrong) calls and what they led to.

1/2) A hold on Gronk and a OPI on Edelman that stunted a promising NE drive. Both were wrong and both wiped red zone first downs off the board.

3) A hold on Mayo that erased a third down stop on Baltimore's subsequent drive. Baltimore, to their credit, then drove down the field and scored a TD.

Instead of punting from their 30 down 17-0, Baltimore was kicking off down 13-7 with 20 less yards allowed and 70 more gained then they earned.

4) A IC call that erased a 2nd and 14 incompletion. Yet another drive extended that led to a TD, though it is possible they would have converted on 3rd and 14.

5) A sack of Flacco was erased when Spikes was flagged for IC on what was actually OPI. Instead of 3rd and goal from the 20, Baltimore had a first down on the 5.

I am quite sure there were a handful more that I am forgetting. And this isn't one of those "it all evens out" instances. Baltimore's only bad call was the flag on Harbaugh that gave NE 15 free yards and led to a FG.

With neutral ref involvement, NE would have won that game something along the lines of 37-17.

by dmstorm22 :: Tue, 01/01/2013 - 2:37pm

I find it hard to believe that a team that outgained their opponent 503-396 and lost the turnover battle by one can be "dominated in just about every facet of that game".

by Anonymousse (not verified) :: Wed, 01/02/2013 - 12:55pm

I'm not saying its the case in this game, but often teams outgain their opponents when they either go behind early, or when they give up points to returns. IE, its tough for your opponent to get a lot of yards when you give up a pick-6 or a kickoff return for TD

by dmstorm22 :: Wed, 01/02/2013 - 1:18pm

You're right, and in this game I think the Pats returned the one turnover (a Flacco pick) close to the goal line, but still the conventional reasoning on the game seems to be that the Ravens were outplayed but won just because of some bad calls, when the Ravens were down early but then took the lead early in the 3rd quarter. There were a bunch of bad calls on all sides (including a bizarre 15-yard penalty on John Harbaugh). It was a mess of a game (but still damn dramatic) but one the Ravens were very competitive in.

by Anonymous1 (not verified) :: Wed, 01/02/2013 - 8:19pm

No, there was on bad call in NE's favor. One.

There were at least a half dozen in Baltimore's favor, including two that extended drives and another that turned a FG attempt into a TD.

No need to sugarcoat a one sided ref job. It is what it is.

by bomp (not verified) :: Wed, 01/02/2013 - 2:41am

So 3.5 quarters of crap officiating in the Pats' favor is offset by one bad call that helped the Ravens because you remember that call better than the 10 that preceded it? And no, the bad call I'm referring to isn't the final field goal, because I'm not blind. Doesn't change the fact that the Pats will probably bang the Ravens if the two teams play, but the Ravens did out-play them four months ago....

by Anonymous1 (not verified) :: Wed, 01/02/2013 - 8:16pm

You didn't watch the game did you? When exactly were those "3.5 quarters of crap officiating in the Pats' favor"?

by jonnyblazin :: Thu, 01/03/2013 - 1:37am

I mean, the DVOA of the game, as I stated before is:

BAL 70%
NE -20%

Baltimore gained more than 2.5 yards per play than the Pats. You can point out 5 or 6 or 7 calls, the bottom line is Baltimore significantly outplayed the Pats on a per play basis. Go ahead and argue against that, not a handful of calls. Because its the per play efficiency that is most predictive.

by jonnyblazin :: Thu, 01/03/2013 - 1:53am

If you want to say that the Pats dominated the game, what aspect did they dominate? Did the pass rush bother Flacco? He went 28-39 for 9.8 Y/A with 0 sacks. Did they shut down the running game? Rice rushed for 5.1 yards per rush. Did they run at will vs. the Ravens D? The Pats averaged 2.3 yards per rush. Did they shred the secondary? The Pats averaged 7.4 per pass, which is good but close to their average (7.0).

by Anonymous1 (not verified) :: Thu, 01/03/2013 - 8:20am

Asked and answered above.

In a fairly officiated game, Baltimore is punting down 17-0 in the second quarter and the game is already getting out of hand. Two of Baltimore's TD drives (and the resulting yards/DVOA bump) should be wiped off the board, and NE should have another red zone stop. Since Baltimore had two extra drives erroneously extended, that is now two extra possessions NE should have had.

I'm not saying Baltimore didn't take advantage of their opportunities, but they most certainly got them.

FWIW, I don't blame the refs for every loss. The replacement guys blew a call that wiped off a game winning Woodhead TD in the Arizona defeat, but NE played poorly enough to put themselves in that situation. In Baltimore, they dominated the opening of the game and, as I said before, should have been getting the ball back already up 17-0. Using stats that include all the extra yards and possessions Baltimore was handed doesn't remotely tell the story of that game.

by jonnyblazin :: Thu, 01/03/2013 - 11:12am

Right, but the question isn't who should have won the game. Its whether the game indicates that the Ravens and Pats are a close matchup. From a DVOA standpoint and traditional stats standpoint, it would seem like the Ravens matchup well. Baltimore had its 2nd best DVOA game of the season and the Pats had their 2nd worse. And if I remember correctly, their were plenty of plays in which the Pats fans were whining about and then went I went back and looked at the game again from the all 22 it was a legitimate call. If you want to go back and point all the calls out via replay thats fine, but its probably a waste of time. Because there were 160+ plays in that game and the Ravens played much better than the Pats in the majority of those plays, hence the huge DVOA discrepancy. It is irrelevant whether those plays should have been run or not: they happened, and the Ravens outperformed the Pats in them.

by Anonymous1 (not verified) :: Thu, 01/03/2013 - 11:33am

Someone wrote up (using the all 22) on the game shortly afterward and confirmed every call I mentioned above as being inaccurate, but FWIW, I don't disagree with your larger point. Baltimore does match up against NE very well, even if you ignore the September game.

NE's offense was rolling last year, and somehow Baltimore gave them fits even though Brady wasn't under much pressure at all. 2007 is another clear example of a game when Baltimore gave NE more trouble than it seemed they would. You can point to 2009, but that was a pretty lousy Patriot team that probably was going to lose their first game to anyone, though perhaps not in such inglorious fashion.

If your point is that Baltimore historically proves to give NE trouble, I don't disagree. Just don't mention that damn game! :-)

by jonnyblazin :: Fri, 01/04/2013 - 8:59pm

OK, I think I get the message! I just watched the condensed replay the other night, the officiating was about as laughable as could be.

by Purds :: Tue, 01/01/2013 - 1:55pm

"Brady would be playing for greatest QB ever"

It would be an interesting Brady legacy Super Bowl if NE gets there, with a bit on the line either way:

A NE win, as you say, might put him in the greatest QB ever slot. (I find it hard to anoint anyone "the best ever" as it's so hard to compare different eras.)

A NE loss might begin, after three SB losses and a 3-3 record in the big game, to tarnish the once impeccable reputation. In really only the last SB win, 2004 season, was Brady what we consider Brady, a top-5 DVOA DYAR guy. In many respects, the SB wins were team wins. But, he's been asked to win the recent SB's pretty much on his own, as "the guy," the undisputed king of the team. It hasn't gone as well, even if you throw in the unlikely things that went wrong. A third loss makes especially the conventional narrative much more difficult to tell easily -- he won three, but then he lost three.

(Yes, I am a Colts fan and Manning fan, so take this with a grain of salt. But, I do think much of the early Brady praise was based on team SB wins, and now that he really is great, those team wins are harder to come by.)

by Anonymous1 (not verified) :: Tue, 01/01/2013 - 2:05pm

I'd have to agree. Brady would be the worst amongst the QBs who made it to the SB six times. :-)

by Purds :: Tue, 01/01/2013 - 2:33pm

Cute, but no one talks about Fran Tarkenton being a great QB, in large part because he lost three SB's.

Fair? no.

Conventional wisdom? Yes.


Let me add to this: I think nothing would better demonstrate the team nature of football than a Brady loss #3 in the SB. To me, he is an infinitely better QB now than he was at the start of his career, but yet if you were to look purely at playoff/championship records, it would seem to indicate that he's gotten much worse. I don't see it that way, but I do think the talent on those early teams was way undervalued.

by Ender (not verified) :: Tue, 01/01/2013 - 10:41pm

I love when people bring up super bowl wins as an argument for how good a QB is. I know that they aren't really worth discussing football with at that point and can just sort of smile and nod. Especially if that is the first thing they go to in the discussion.

by jedmarshall :: Wed, 01/02/2013 - 8:27am

Caveat of me being a Colts fan too, but teams win Super Bowls, not QB's. Brady is an infinitely better QB the past 8 years while not winning the SB than he was when he "won" 3 early in his career. Flip his career around and you have John Elway (in terms of team success, not individual stats). Turns out 50 other players have to support a good QB and usually get a little luck to win.

by Aaron Brooks Go... :: Wed, 01/02/2013 - 11:10am

"In really only the last SB win, 2004 season, was Brady what we consider Brady, a top-5 DVOA DYAR guy."

Weren't the Panthers the worst SB team of the DVOA era?

by Eddo :: Wed, 01/02/2013 - 12:22pm

What do the 2003 Panthers have to do with the 2004 Patriots and Tom Brady?

by Aaron Brooks Go... :: Wed, 01/02/2013 - 2:21pm

I flipped the Eagles and the Panthers seasons. I was thinking in terms of the 2004 Super Bowl, not the 2004 season.

by RickD :: Tue, 01/01/2013 - 9:53am

Clearly, in terms of all-time best teams, DVOA has accurately captured the glorious nature of my favorite team. The other teams also listed obviously have benefited from weak schedules, media hype, replacement refs, global warming, and the moon's gravitational pull.

by Ryan D. :: Tue, 01/01/2013 - 10:21am

Well, the competition portion of this thread is now over. Everyone else can continue to jockey for the title of the second-best comment in this thread, should they feel the need.

by MudShark (not verified) :: Wed, 01/02/2013 - 2:03am

Too true! As a Seahawks fan I concur! Damn, overrated other 31 teams....

by ammek :: Tue, 01/01/2013 - 12:49pm

Congratulations to Mark Sanchez, who has finally achieved the Dilfer Double of ranking dead last in passing DYAR and dead last in rushing DYAR for quarterbacks. The previous two years, someone has come agonizingly close, but both Jimmy Clausen in 2010 and Blaine Gabbert in 2011 finished next-to-last in rushing, Gabbert by a mere 1 DYAR thanks to Tim Tebow.

Even Trent Dilfer himself never managed the Dilfer Double in the same season, though he came close. As a result, I think the achievement ought to be named the Sanchez Sweep.

by Pass to Set Up ... :: Tue, 01/01/2013 - 6:36pm

Your chart has a typo. It lists last year's rankings for the Jets as #10 and Eagles as #9. This should be reversed, you ranked the Eagles #10 and Jets #9.

by Aaron Schatz :: Tue, 01/01/2013 - 7:47pm

The ratings listed for "last year" are based on the current version of DVOA, not the version we were using last December, so there are some small differences.

by Anonymousse (not verified) :: Wed, 01/02/2013 - 1:02pm

Clear proof that DVOA was overrating the eagles.

by Paul R :: Wed, 01/02/2013 - 3:08am

It seems to me that the teams which perform best in the playoffs tend to be the teams with the most "balanced" DVOA, not necessarily the teams with high rankings.
As I recall, last years Giants were never at the top of the heap, yet their offense and defensive DVOA rankings were always similar.

Obviously, a team with a 32nd ranked offense and 32nd ranked defense is not one to bet on, but perhaps a team with a balanced--but not necessarily overpowering--attack is a tougher opponent to face because their opponent has to play well on both sides of the ball.

For example, using the current DVOA, if the Patriots (1st offense, 15th defense) played the Falcons (12th, 12th) tomorrow, I wouldn't bet the farm on the Patriots.

by Alternator :: Wed, 01/02/2013 - 5:10am

New England and Atlanta are extremely close in defensive DVOA. You do realize that you have just said, "Because Atlanta is worse on offense, I think they'd have an excellent chance to beat the Patriots," right?

by Anonymousse (not verified) :: Wed, 01/02/2013 - 1:00pm

I'd expect a Patriots/Falcons game (on a neutral field, in a non-playoff situation) the same way I'd expect a Patriots/Houston game to play out: A great team handily beating a good team.

I would expect the Patriots to be roughly 10 point favorites. Just as I would expect Seattle, or SF

by Paul R :: Wed, 01/02/2013 - 4:13pm

"New England and Atlanta are extremely close in defensive DVOA. You do realize that you have just said, "Because Atlanta is worse on offense, I think they'd have an excellent chance to beat the Patriots," right?"

Yes, that's what I'm saying, sort of. I might not go so far as to call it an "excellent" chance, but I am saying I wouldn't bet the kids' college fund on New England.

I've been to busy/lazy to look up the stats, but it seems to me that when a lower-ranked (in DVOA) team upsets a higher-ranked team, the lower-ranked team has more balanced DVOA ratings (i.e. Atlanta's 12th offense and 12th defense) than the higher-ranked team. (New England's 1st offense, 15th defense)

My hypothesis is that perhaps the even-handedness of a team, consistent above-average play on both sides of the ball, has a value that the DVOA rankings don't seem to show.

by Eddo :: Wed, 01/02/2013 - 4:57pm

You might be right, but Atlanta and New England are not the best examples to use.

Rather than looking at ordinal rankings, look at the actual DVOA values.

New England has +30.8% ODVOA, +1.3% DDVOA. Atlanta has +6.1% ODVOA, -2.9% DDVOA.

The Patriots have a very good offense and an average defense. The Falcons have a slightly good offense and an average defense.

This year isn't a great year to test your theory. The top teams are actually pretty balanced, with the exception of New England, and even they are an average defensive team.

Last year was different, as the top three offenses (Green Bay, New England, and New Orleans) all had below-average to bad defenses. And even then, they lost to the Giants, Giants, and 49ers, respectively. The Giants had a good offense but just an average defense, while the 49ers had a very good defense but an average offense.

by Paul R :: Thu, 01/03/2013 - 12:47pm

Do the offensive and defensive DVOA numbers have a similar value?

For example: Is +10% offense as good as -10% defense?

by nat :: Thu, 01/03/2013 - 2:57pm

Do the offensive and defensive DVOA numbers have a similar value?
Almost but not quite. Since 2009, there are certain plays that count for an offense and not for a defense, or vice versa. This changes both what a team is evaluated on, and the baseline they are evaluated against.

Mostly this adds plays to the offensive side: aborted snaps, false starts, stuff like that. These depress the average that offenses are compared against, but also give them opportunities to be dinged for those same unforced errors. It also means that adding 10% of offensive DVOA should have slightly less real world value than 10% of defensive DVOA.

I don't know how big that effect is. I'd say consider it a tie-breaker unless Aaron tells us that the effect is bigger. But really, when two teams in an upcoming game are close in DVOA, consider home field advantage, weather, and specific match ups and injury reports before worrying too much about the two slightly different DVOA scales.

For example, if Denver and New England were to meet this off season, pick Denver as the home team, unless you really like the specific match ups for the Patriots, or think Denver has critical players hurt. But don't be surprised if it goes either way, seeing that the DVOAs are very close.

I hope that answers clearly enough.

by Scott Crowder (not verified) :: Wed, 01/02/2013 - 4:23pm

Is there somewhere on here where you track home/away DVOA? I'd be really interested in seeing how teams perform at home and away. I know there was an article about it a week or so ago, but I don't see anywhere where I can look up those numbers.

by ipsofacto (not verified) :: Wed, 01/02/2013 - 7:07pm

Now we know most of the top teams are playoff teams. Duh.

Now we know no rookie QB has ever won or taken a team to the Super Bowl. Duh.

Now we know no Super Bowl winning QB has ever won a Super Bowl with another team. Duh.

Now we know teams ending the season with an 11-game winning streak have gone to the Super Bowl ever other year and never missed the Super Bowl within two occurrences of one another. Duh.

by Dr Marc (not verified) :: Wed, 01/02/2013 - 11:27pm

I have not been a follower for all that long, but as the team that ended first in total team defense( in terms of total yards) been ranked as low as 13 in defense?