DVOA Analysis

Football Outsiders' revolutionary metrics that break down every single play of the NFL season

Final 2007 DVOA Ratings

by Aaron Schatz

Although they struggled during a few of their later games, the New England Patriots finish 2007 with the best DVOA rating of any team in the 12 years for which we have the play-by-play breakdowns. They also finish with the best offensive rating ever. The Patriots are the first team to ever finish a season with a total DVOA over 50% or an offensive DVOA over 40%, and the first team to finish with more than 14 "estimated wins." They are the fourth team to finish the year number one in both passing and rushing DVOA, along with the 2002 Chiefs, the 2001 Rams, and the wacky mirror-universe 1996 Ravens. (First in offense, 29th in defense. Really.)

2007 NE 51.8%   2007 NE 42.8%
1999 STL 45.8%   2004 IND 39.9%
1996 GB 40.6%   2006 IND 33.7%
2001 STL 38.5%   2002 KC 33.3%
2000 TEN 37.4%   2000 STL 31.7%
2004 NE 35.7%   2000 IND 30.1%
2004 PIT 34.8%   2004 KC 28.7%
1999 JAC 34.8%   2007 IND 28.3%
2002 TB 34.0%   1998 DEN 28.0%
2004 IND 33.8%   2001 STL 27.3%
2005 IND 33.5%   2003 KC 27.0%
2007 IND 33.0%   2005 IND 26.9%

The Patriots also excelled when it came to individual stats, of course. Tom Brady is the first quarterback to ever earn more than 200 DPAR in a season. Randy Moss is the second wide receiver to earn more than 50 DPAR in a season, and his 2007 campaign narrowly edges out Marvin Harrison's 2001 season for the top WR DPAR ever. In addition -- this is discussed further in Quick Reads -- Brady does in fact finish with the greatest quarterback season in modern NFL history according to the formula we used in Pro Football Prospectus 2005. I mistakenly forgot to update opponent adjustments when I wrote last week's Quick Reads, which would have put Brady eighth instead of 11th through last week. Then Brady went and had another great game against the Giants, much more in line with his numbers from the first half of the year, and that was enough to move him past Peyton Manning and Bert Jones.

Speaking of Mr. Jones and Mr. Manning: Clearly, sitting your starters in the final week is a Colts franchise tradition, and once again it affected their DVOA rating this year, dropping the 2007 Colts out of the historical DVOA top 10.

On the other end of the spectrum, the 2007 San Francisco 49ers finish 10th among history's worst total DVOA ratings, and eighth among the worst offensive DVOA ratings. A bad team, but not the historically bad team they looked like a few weeks ago.

Once again this year, teams sitting their starters in Week 17 played havoc with the DVOA ratings. As I've noted in past years, all of the research I've done has shown that removing these "sit starters" games from DVOA doesn't improve the accuracy, and therefore we leave them in. We'll look at this issue again this off-season, but for now, every play of the season is in the ratings. As you can imagine, that means a lot of good teams take a hit. Out of the final top 10 teams, only Green Bay and Tennessee have a higher DVOA after Week 17 than they did after Week 16.

Of course, not every team that had a terrible game in Week 17 was sitting its starters. I'm looking at you, Dallas.

Pittsburgh was one of the teams sitting people, and it cost them the number one spot in defensive DVOA. Instead, Tennessee finishes the year as DVOA's top defense. Last week I wrote about how the Steelers defense is nowhere near as good as the defenses that led the league in DVOA from 1996-2006. After the final game, this year stands out even more as a year when there were a few very good defenses, but no great ones. At -13.5%, the Titans would have finished fourth in 1997 and 2006, and fifth or lower in every other year. This year's best defense ranks 59th among the 375 teams in DVOA history.

Chicago finishes the year with the top special teams. In fact, Chicago finishes with the second-highest special teams rating in DVOA history (9.1%), behind only the 2002 New Orleans Saints. However, that rating comes with an asterisk. Regular readers know that I added a new element to our kickoff ratings at midseason to try to account for teams deliberately short-kicking to avoid Devin Hester (and, eventually, numerous other returners). I have not gone back and standardized this change, or added it to previous years. Using the old formula for special teams, Chicago's special teams rating would only be 7.5%. That would still lead the league in 2007, but it would be far behind from the 2002 Saints.

Another note: I have fiddled with the special teams baselines over the last couple weeks to reflect the leaguewide changes in special teams this year. Punt returns were longer than ever, but that was nothing compared to the colossal jump in kick return yardage. On top of this, kickoffs themselves were longer when kickers weren't deliberately trying to avoid the returner, so everything had to be jiggled to keep it from looking like the 32 worst kickoff coverage teams in NFL history all played in 2007. On top of this, field-goal percentage went up around the NFL for like the zillionth year in a row. I will probably need to do a little more adjusting in the off-season, but the changes I've made over the last couple weeks should help even things out for now.

Despite the problem of teams sitting people and not trying in Week 17, this year's DVOA ratings really do a good job of reflecting the best and worst teams. The top 10 teams in DVOA all made the playoffs, as did a team that is effectively tied for 11th (Washington). The Giants are the one straggler, finishing 16th. Even the Bengals' silly-looking DVOA ratings make sense now. Yes, I personally think the Giants are better than the Bengals, but 15th sounds like a reasonable ranking for a 7-9 team in the superior conference.

* * * * *

Here are the final Football Outsiders team efficiency ratings for the 2007 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 for opponent and consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season.

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. LAST YEAR represents 2006 rank, while LAST WEEK represents rank in Week 16 of 2007.

To save people some time, we request that you please use the following format for all complaints:

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

1 NE 51.8% 5 1 40.6% 1 16-0 42.8% 1 -6.1% 8 2.9% 7
2 IND 33.0% 7 2 22.2% 4 13-3 28.3% 2 -10.8% 3 -6.1% 32
3 DAL 23.9% 10 5 17.0% 6 13-3 19.0% 4 -5.8% 10 -0.9% 18
4 JAC 23.7% 6 4 31.1% 2 11-5 20.8% 3 -3.3% 12 -0.4% 15
5 GB 21.0% 15 8 19.4% 5 13-3 17.3% 5 -1.3% 15 2.4% 8
6 SD 18.9% 2 5 27.1% 3 11-5 4.8% 15 -9.8% 5 4.2% 4
7 TB 17.8% 31 6 13.5% 9 9-7 7.7% 10 -10.2% 4 -0.1% 12
8 PIT 17.5% 12 7 5.6% 14 10-6 6.8% 12 -12.3% 2 -1.7% 21
9 SEA 11.7% 25 9 13.8% 8 10-6 5.6% 14 -5.4% 11 0.7% 11
10 TEN 8.8% 24 13 2.2% 16 10-6 -4.0% 21 -13.5% 1 -0.6% 17
11 PHI 7.7% 3 10 7.6% 11 8-8 10.2% 8 -3.2% 13 -5.7% 31
12 WAS 7.7% 23 14 10.1% 10 9-7 1.0% 17 -7.2% 6 -0.5% 16
13 CLE 5.8% 30 12 14.7% 7 10-6 6.9% 11 7.6% 22 6.6% 2
14 MIN 4.9% 22 11 6.2% 13 8-8 3.4% 16 0.2% 18 1.7% 9
15 CIN -0.1% 11 15 0.8% 19 7-9 11.0% 6 10.1% 24 -0.9% 19
16 NYG -0.6% 8 16 -1.7% 21 10-6 -2.5% 19 -2.9% 14 -1.0% 20
17 HOU -2.1% 28 21 1.0% 18 8-8 5.7% 13 13.3% 30 5.5% 3
18 DEN -3.2% 18 19 2.1% 17 7-9 7.8% 9 6.3% 21 -4.8% 28
19 BUF -3.8% 16 18 0.5% 20 7-9 -7.6% 22 0.1% 17 3.9% 6
20 CHI -4.4% 4 20 7.3% 12 7-9 -20.4% 31 -6.9% 7 9.1% 1
21 NO -5.1% 9 17 4.0% 15 7-9 10.3% 7 11.4% 27 -3.9% 26
22 BAL -7.0% 1 22 -13.8% 24 5-11 -12.8% 26 -6.1% 9 -0.3% 13
23 ARI -10.1% 26 23 -12.3% 22 8-8 -0.8% 18 5.4% 20 -3.9% 25
24 DET -19.9% 29 24 -25.3% 28 7-9 -2.7% 20 13.9% 31 -3.3% 23
25 NYJ -20.8% 19 26 -12.5% 23 4-12 -10.8% 25 10.9% 25 0.9% 10
26 KC -21.2% 13 25 -25.0% 27 4-12 -18.0% 29 -0.5% 16 -3.7% 24
27 CAR -21.3% 14 27 -19.5% 25 7-9 -15.2% 27 0.7% 19 -5.4% 30
28 ATL -23.1% 21 28 -24.2% 26 4-12 -10.7% 24 12.1% 29 -0.4% 14
29 MIA -28.1% 17 29 -31.4% 31 1-15 -8.8% 23 16.7% 32 -2.7% 22
30 OAK -28.7% 32 30 -27.1% 29 4-12 -15.4% 28 8.3% 23 -5.0% 29
31 STL -33.9% 20 31 -30.9% 30 3-13 -18.7% 30 11.0% 26 -4.3% 27
32 SF -38.0% 27 32 -32.2% 32 5-11 -30.5% 32 11.5% 28 4.1% 5

  • NON-ADJUSTED VOA does not include adjustments for schedule, fumble recovery rate, or (for special teams) weather and altitude.
  • 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.
  • 2007 SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • PYTHAGOREAN WINS represent the number of wins projected from the team's points scored and allowed, as described in this article.
  • VARIANCE (VAR.) measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from least consistent (#1, highest variance) to most consistent (#32, smallest variance).

RANK 2007
1 NE 51.8% 16-0 52.7% 14.2 1 0.1% 19 13.8 1 14.8% 19
2 IND 33.0% 13-3 31.4% 12.4 2 2.5% 8 12.5 2 16.4% 14
3 DAL 23.9% 13-3 27.2% 10.8 5 -1.5% 21 11.0 6 24.2% 3
4 JAC 23.7% 11-5 22.0% 11.3 3 1.7% 10 10.7 7 15.6% 16
5 GB 21.0% 13-3 27.2% 11.1 4 -5.6% 29 11.5 3 21.7% 6
6 SD 18.9% 11-5 16.7% 9.6 7 0.2% 18 11.3 5 25.0% 2
7 TB 17.8% 9-7 22.3% 10.1 6 -7.4% 30 10.0 9 22.1% 5
8 PIT 17.5% 10-6 19.4% 9.3 9 -3.3% 24 11.4 4 22.7% 4
9 SEA 11.7% 10-6 20.2% 9.1 11 -11.0% 32 10.7 8 12.1% 25
10 TEN 8.8% 10-6 3.2% 9.2 10 1.4% 13 8.1 15 18.3% 10
11 PHI 7.7% 8-8 0.8% 9.5 8 4.3% 5 9.1 11 14.5% 20
12 WAS 7.7% 9-7 0.2% 8.3 15 4.4% 4 8.7 12 15.0% 18
13 CLE 5.8% 10-6 9.1% 8.9 12 -5.1% 28 8.5 14 6.2% 32
14 MIN 4.9% 8-8 7.1% 8.8 13 -3.9% 26 9.5 10 19.5% 9
15 CIN -0.1% 7-9 6.8% 8.5 14 -3.2% 23 7.9 17 8.4% 31
16 NYG -0.6% 10-6 -4.4% 7.9 17 1.1% 14 8.6 13 8.4% 30
RANK 2007
17 HOU -2.1% 8-8 -7.7% 7.7 18 2.7% 7 7.9 18 13.8% 22
18 DEN -3.2% 7-9 0.4% 7.3 21 1.1% 17 5.7 21 34.2% 1
19 BUF -3.8% 7-9 -9.4% 8.3 16 5.1% 2 4.9 26 11.8% 26
20 CHI -4.4% 7-9 -6.5% 7.4 20 1.5% 12 7.6 20 14.4% 21
21 NO -5.1% 7-9 -4.9% 7.5 19 -3.5% 25 7.8 19 13.7% 23
22 BAL -7.0% 5-11 -14.4% 6.9 22 1.7% 11 5.0 25 15.4% 17
23 ARI -10.1% 8-8 -0.7% 6.2 24 -9.3% 31 8.1 16 20.0% 8
24 DET -19.9% 7-9 -17.4% 6.3 23 3.5% 6 5.7 22 18.1% 12
25 NYJ -20.8% 4-12 -23.3% 5.0 28 5.1% 1 5.4 24 11.6% 28
26 KC -21.2% 4-12 -23.5% 5.1 27 1.1% 15 4.5 28 21.5% 7
27 CAR -21.3% 7-9 -22.3% 5.8 25 1.1% 16 5.6 23 16.6% 13
28 ATL -23.1% 4-12 -21.5% 5.5 26 -1.2% 20 4.0 29 15.9% 15
29 MIA -28.1% 1-15 -35.3% 4.1 29 4.9% 3 3.8 30 12.4% 24
30 OAK -28.7% 4-12 -28.3% 4.0 30 2.0% 9 4.9 27 18.3% 11
31 STL -33.9% 3-13 -31.3% 3.6 32 -2.7% 22 3.7 32 9.6% 29
32 SF -38.0% 5-11 -33.2% 3.9 31 -4.7% 27 3.7 31 11.7% 27

For those readers who are sticklers for detail, I should mention that the DVOA ratings on the "best DVOA ever" table may differ slightly from ratings on the stats pages, because they may include small play-by-play fixes that aren't reflected on pages last revised a year or two ago. We'll make sure to update everything to the most current numbers over this off-season.

On to the housekeeping... All the team stats pages and individual stats pages are now updated with final 2007 numbers. Because of New Year's, we're not yet sure when we'll have playoff odds and the DVOA Premium database updated.

There will be no new Football Outsiders content on New Year's Day, and no Any Given Sunday this week. We'll be back Wednesday with the Every Play Counts All-Pro Team, the second-half Loser League review, and the usual DVOA on AOL piece with comments on all 32 teams.

For those interested, I'll be chatting about the playoffs over at BaseballProspectus.com this Friday from noon until 1:30pm Eastern. Feel free to also ask questions about what the other 20 teams should be doing over the off-season to improve.


142 comments, Last at 04 Jan 2008, 11:15am

1 Re: Final 2007 DVOA Ratings

I seem to recall that GB had a very low variance for most of the year. And this week they're ranked #6. I assume that's been slowly climbing over the course of the year, and I just wasn't paying attention to it.

I'll go back and look, but as a general question, have there been many teams that had a large jump in variance (one way or the other) this year? If so, any speculation/commentary on what the causes might have been?

2 Re: Final 2007 DVOA Ratings

First? And does Chi weighted DVOA show future promise for next year or just a big fluke? If injuries revert back to an average number, they might be decent next year. And if Mike Brown somehow doesn't break every bone in his body like every year, they'll be really good again.

3 Re: Final 2007 DVOA Ratings

How close is Brady to Manning in the best QB season ever comparison? Cause since Manning didn't play in week 17 in 2004, maybe we should prorate his first 15 games to 16 and then compare.

4 Re: Final 2007 DVOA Ratings

Is it just me or did the Colts play a comb of some really bad offenses, teams with offensive injuries when they played them, good offensive teams struggling on offense when they played team, JAC week 13 at home and NE at home? Is it really the #3 D in the NFL? I just didn't see any team loaded for bear to knock off the Champs all year.

5 Re: Final 2007 DVOA Ratings

And to answer my own question, apparently I was correct in remembering that GB had a really low variance all year. Until week 16 and the loss to Chicago. Which vaulted them from the mid-to-low 20s in variance up to #5 in week 16, #6 final.

Weird. But I guess it's variance, so...yeah. I guess one game playing really poorly would have a big effect on variance if you've played well most of the rest of the year.

6 Re: Final 2007 DVOA Ratings

#1, GB's variance skyrocketed due to their disaster in Chicago

It's also intersting how GB in terms of DVOA is abrely ahead of SD but in estimated wins trump them b 1.5+ wins

*It could be a crazy playoff considering 4 of the Top 7 teams in DVOA rank so highly on the variance list

7 Re: Final 2007 DVOA Ratings

And does Chi weighted DVOA show future promise for next year or just a big fluke?

Yeah, promise that they can look as good as the Eagles and Vikings, and miss the playoffs with a mediocre record. Look forward to it.

8 Re: Final 2007 DVOA Ratings

7: Perhaps it means we should respect the power of the neck beard.

9 Re: Final 2007 DVOA Ratings

All, I've recently started a website with a stats based power rating that I call EPPR (Expectation-Performance Power Rating). I haven't updated for the last week of this season yet, but check out how the Pats compare to other teams of all time (I've rated all teams after each game going back to 1933). Any feedback is welcome. Here's the direct link to the ratings: http://www.quantsportsguy.com/alltimeranks.php. EPPR is based on winning (and by how much) and how good the opposing team is, similar to Elo rankings in chess. I'm hoping to check correlations to the DVOA during the offseason.

An interesting note... the Pats have the 4th best end to the regular season, and the best since the 1968 Colts. Hoping to update with this weeks' numbers this week. Thanks!

10 Re: Final 2007 DVOA Ratings

sure would love to see one of those trend line thingies for the Redskins...and I guess the other playoff teams as well.

11 Re: Final 2007 DVOA Ratings

Should there be any correction to the ratings for resting starters and testing rookies in week 17? I would think that would distort the weighted ratings significantly.

12 Re: Final 2007 DVOA Ratings

Disregard my comment. I somehow skimmed past the part of the post that addressed this.

13 Re: Final 2007 DVOA Ratings

#4 - The Colts were 5-0 in meaningful games against the best division in the NFL, the AFC South. Their defensive backups made Tennessee's starters look inept Sunday night, holding them to 2.5 yards per carry. They led the league in points allowed and didn't allow more than 25 points in any game. They played the Pats without two of three starting LBs and faired okay. This is the best Colts defense that I've ever seen, even without Dwight Freeney.

14 Re: Final 2007 DVOA Ratings

Mostly, the large number of teams resting players hasn't affected much like you said, Aaron, but it sure did hurt the Seahawks' defensive DVOA. Although the starters were in there, longer, except for Rocky Bernard, but even though the coaches were playing to win, the players clearly weren't in it, overlooking a 2nd-straight downtrodden NFC South opponent.

Opponent-adjustmnet is opponent adjustment, and maybe they deserve the significant kick down the ratings for playing those 2 games the way they did, but when you look at scording defense, the drive stats here, number of drives, and time of possession, they probably have the greatest differential between opponent-adjusted and non-adjusted numbers. Of course a lot of that is because their schedule was so dang easy, but in a defensive year that is so incredibly down across the board, they really have been playing like the 2nd best defense in the conference, and the 4th or 5th in the league.

15 Re: Final 2007 DVOA Ratings

I think there's a good possibility that the Pats won't make it out of the AFC playoffs

The Jaguars are on fire. From Weeks 10-16; their DVOA has been about 54%!

Since the Colts have gotten reasonably healthy (after they played KC in week 11), their DVOA from weeks 12-16 was ~41%.

From Weeks 5-16, the Chargers have had a DVOA of ~34%.

This team just reminds me a lot of the 01 Rams and especially the 05 Colts.

16 Re: Final 2007 DVOA Ratings

Can someone explain how the Raiders could set the all time record for net punting average and be 26th this year in punting by DVOA?

I can certainly buy the idea that there are more sophisticated metrics than net punting average and that those metrics would put them much lower than the best of all time.

Still, I would think that net punting average would have a strong correlation with the punting rating. Certainly strong enough that a record breaking season in one would be above average in the other.

I know that they have very few punts from the other team's side of the field, since they (1) do not have a great offense, (2) kick more long FGs than most teams, and (3) go for it on 4th down more than most teams. That is a big boost for net punting average and likely explains a part of this.

That's about the only thing I can think of and that sounds more like a reason that you could be #1 all time in NPA, but only #26 all time or only 5th this year or something like that. It does not seem like a reason for being #1 in history for NPA, while 26th this YEAR in punting.

Has FO found any other performances that were both record breaking in a positive way by conventional stats, but bad by FO stats?

17 Re: Final 2007 DVOA Ratings


18 Re: Final 2007 DVOA Ratings

very bad rankings of the Raiders. No way Raiders sgould be ranked lower than Dolphins. ridiculous rankings system at this site. Maybe I can help make it better for next season.

I give it a try now.

Raiders in top ten

19 Re: Final 2007 DVOA Ratings

Scharf, the thing that struck me about those numbers was the variance. I mean if the predictions were consistently low or high that would be helpful. But from that it just looks like a crap-shoot. It's a good thing I don't use these to bet.

20 Re: Final 2007 DVOA Ratings

Do the Bills have the biggest difference between expected wins and pythagorean wins this year? They play well (ish), they just don't score points or stop the other team from scoring?

21 Re: Final 2007 DVOA Ratings

14, I'm not sure if the coaches were in it, either. A lot of post game articles noted that the Hawks played the same base defense the entire game. That sort of sounds like the D-Coordinator wanted to get the game the hell over with.

22 Re: Final 2007 DVOA Ratings

First, my apologies to Aaron and staff, and their various ancestors whom I maligned when I saw the preseason projections of the Jags and the Colts. Jags win total was right on the mark. Considerably off on Indy, but that just means you and I were each half right, and you probably did it without spluttering all over your keyboard. Kudos.

#3 If you are going to prorate, prorate by quarters. Manning sat out about 8+ quarters in 04, Brady maybe 1, 2 at most this year.

#4 Otis, Mr. Sanders had it about right and he sees more of the Colts than I do. I DID think they caught a few breaks such as knocking Garrard out of the first Jags game (but then again, they did that with Leftwich last year--is it lucky if you knock their guy out? ot just hard football? And I think an RB in the Tampa game.). But they have also played short-handed most of the year, so I'd call that a wash at best. The MLB and secondary have been healthy, but not 6 of the front 7, to say the least.

In terms of timing teams when their opponents are hot/cold, they had SD when the Chargers started to turn their season around, as opposed to playing them in the first few weeks like the Pats did (who would have beaten SD regardless of when they played, let's face it). At the time they lost to NE, it was the closest game NE had by two scores and they held them to 24, which at that point was their lowest output by two scores again! So all in all, I'd say Indy's D is for real. Even the backups last night were beating the snot out of a Tenn team clawing for a layoff spot. They are finally a real scary D--they hit hard and tackle well, and are fast. They lack size up front and looked a little gassed at the end of the NE game, but all in all, even with subs in, they are a very good D. Better than 2005, and indicating that 2006 was the aberration.

23 Re: Final 2007 DVOA Ratings

How much of a boost did TEN's defense get from playing against 2+ quarters of the impressive Sorgi-led Indy offense, instead of the Manning-led one that DVOA thought they were playing?

24 Re: Final 2007 DVOA Ratings

Looks like the Hawks and Skins will be a dogfight. And, as an early adopter of the Packers, I'm very skeptical of their DVOA. I'm a big believer that deep passes are one of the least consistent abilities a team can have, and for those offenses predicated on the long bomb, you may look unstoppable one week, and like Eli Manning the next.

25 Re: Final 2007 DVOA Ratings


According to DVOA, the current method of ranking teams and determining playoff seeding ALMOST got everything right. Each division winner had the best DVOA in their division, the remaining two best teams in each conference were JAC, TEN, PHI, and WAS.

So the only place where DVOA disagrees with the standings is that the Giants have no business being in the postseason ahead of the Eagles (or Minnesota).

But where is Cincy's reasonable DVOA coming from? Anyone figured that out yet?

26 Re: Final 2007 DVOA Ratings

RE: #22

There is no question the Colts defense is for real.

There is one difference between this year's tournament and last year's. Last year, the Colts good defense came out of the blue and none of their opponents had any game film. This year, there is game film of the good Colts defense.

I think it is impossible to handicap the Colts chances in the tournament until we really find out what games Polian is playing with all the injured players -- specifically Harrison and the d-line.

27 Re: Final 2007 DVOA Ratings

"But where is Cincy’s reasonable DVOA coming from? Anyone figured that out yet?"

What do you mean? The Bengals have been about average in DVOA all year, and in fact, they are about average, having gone 7-9 against a decent set of opponents. Of course, earlier in the season, everyone was saying that DVOA must be broken since it said they weren't terrible. Yeah, funny how that turned out.

28 Re: Final 2007 DVOA Ratings

Yeah, Indy's defense is definately for real. And I'm a big supporter of Bob Sanders for DPOY (I'm a Pats fan, btw). People mention his height all the time, but they overlook the fact that it's really an enormous advantage for him - the low man wins, in football. It's so difficult for o-linemen to try and block a short guy who gets down so low and is so explosive in diagnosing and pursuing plays. He creates mismatches just by being on the field and does a lot for that defense. Incredible player.

29 Re: Final 2007 DVOA Ratings

NB. i'd like their defense a lot more if they had a healthy freeney and mathis, but i guess that goes without saying

30 Re: Final 2007 DVOA Ratings

I'm not sure if it can be fixed, but Maroney's play on Saturday evidences the problems with DPAR. He was awful, dancing around instead of moving the pile. Yet, the Pats give him the ball in situations that are very DPAR-friendly.

31 Re: Final 2007 DVOA Ratings

Colts defense from my observation

The biggest jump i have noticed in the colts defense is that the entire secondary now tackles exactly like Bob Sanders. The missle dive to the thigh knee area, from watching the Jags play the other teams in the league then playing the colts thats the biggest difference, other teams attempt to arm tackle Taylor and Jones-Drew in the secondary and they drag them all over the place. But every db on the colts goes low, you will notice the receivers all trying to leap unsuccesfully to get over the tacklers to no avail. On most teams you have one player that plays like that, Polamalu, Ed Reed, Sanders before this year.

That is my unprofessional opinion on the biggest difference I have seen, don't know if I'm crazy or not, but i'm sure someone will let me know

32 Re: Final 2007 DVOA Ratings

As to the Colts this year: In past years, say 2004-2005, I didn't give much credence to the "soft" label people placed on the Colts. Yes, they didn't beat NE, but who did? I still think those teams were average in terms of being physical, but watching the backups last night against Tennessee, it's pretty obvious that the Colts now are very physical. I can't believe more people weren't hurt in last night's game (on both sides).

The defense is for real, and that will show against just about everyone, except for You-Know-Who. What the Colts D does very well -- attack the ball carrier, keep passes in front of them and expect the QB to eventually make a mistake -- will likely not work against Those-Who-Must-Not-Be-Named. The Colts have for several years played a risk defense, allowing many medium gains but hoping to cause a mistake, a sack, a 3rd-and-long, a turnover. It works against most teams, and most offense. Against the historical Team-That-Must-Not-Be-Named, well, we'll see.

As a Colts fan, I think the biggest problem for the Colts in the first meeting this year wasn't the defense giving up 4th quarter touchdowns, but that the offense couldn't generate any points in the last few drives.

33 Re: Final 2007 DVOA Ratings

This site is the bomb diggity, yo. I love it how you guys create uour own stats to make it appear the Eagles are better than the Giants

34 Re: Final 2007 DVOA Ratings

I posted this in the DVOA predictions thread as areas where I disagreed with the projections. Let's see how I did.

Some observations
-I don’t really understand why NE has a better projected offense than Indy

Well, I do now. Though, Indy's offense was where I thought they'd be (with a couple more injuries), NE's was much better.

-Doesn’t DVOA penalize for lack of depth? I’d think that would drop Washington

Score one for DVOA

-Tampa Bay is not the 6th best team in the NFL, etc

Score another for DVOA

-San Diego’s ‘bad’ projection is fueled entirely by the 29th ranked projected defense. It’s hard to imagine that a team with that defensive line, Merriman, Phillips, and a couple decent corners (Jammer and Cromartie) could be that bad.

SD's defense is 5th in the league. I was right about that.

-The Jets come out pretty well. Does the system think Pennington will stay healthy the whole year?

Jets pessimism rewarded!

-I think Manning & co’s ability to consistently move the chains may not be dealt with effectively by the system because of the lack of comparable reference points.

I still think this; the Colts exceeded their projection despite a rash of injuries.

-The system is surprisingly optimistic about Atlanta, especially the defense.

Maybe it shouldn't have been.

-Has there been a ‘Bengal arrest factor’ addition to DVOA? The offense seems under-predicted even after accounting for oline problems. It’s not like there weren’t problems last year, and they were still pretty good.

DVOA undershot them by 4%. Not bad.

-I wouldn’t be surprised to see NO come close to that prediction (within 5%). This is a team that still employs Fred Thomas, after all.

Me and DVOA were right on this one.

-Why is Detroit predicted to have an above-average defense? Are they the Cardinals of ‘04 (underrated D, aging qb, good young receivers, no oline)?

I still have no idea.

-No love for Oakland’s defense? I thought that with their young talent, especially concentrated in pass defense, they would be one of the few teams expected to overcome defense variability.

DVOA was right on that.

-DVOA hates Matt Leinart, apparently.

Maybe with good reason.

-This has been said before, but the STL offense should be better than that (even though I still think it’s a 6-7 win team).

Without all the injuries, who knows? DVOA predicted what did happen

-The projections (especially on the defensive side of the ball) are very clustered.

With good reason, apparently.

Well, the DVOA projection system got the majority of those predictions relatively correct.

35 Re: Final 2007 DVOA Ratings

This was just posted on NFL.com; it gives a slightly different perspective than DVOA:


The NFL draft calculus positions teams in draft according to the the strength- of- schedule, that is the records opposing teams. In this light the teams that faced the stiffest opposition during the past year were the Eagles (opponents' record:.563, 144 - 112) and the Redskins (.555, 142 - 142).

Other teams in the playoffs such as the Steelers (.453, 116 - 140), the Cowboys (.496, 127 - 129), had less difficult opponents. The Bucs, Packers and New England likewise cruised with easy schedules: (.469, 120 - 136). The Seahawks should quit right now: (.410, 106 - 150).

Too bad about Brian Billick. He deserved to ve fired, but the problems with the team go past the coaching staff. The birds need a decent quarterback and some people to throw to. Next year is rebuilding: they should finish 6 - 10 next year with Monte Kiffin as head coach. (Joke)


36 Re: Final 2007 DVOA Ratings

Said it before, I'll say it again.

Not sure what DVOA is missing, but it is missing something regarding the Giants.

They aren't the best team around, or even top-5. But DVOA consistently underrates them.

Let's give a really simple example. How many other teams could have given the Pats that kind of game?

If the answer is 17, then the Pats aren't what they seem to be. If the answer isn't at least 17, then the Giants aren't what DVOA measures them.

It's the best measurement system around, but it is missing the mark there.

37 Re: Final 2007 DVOA Ratings

re #22:

Looking at the Sorgi gamelogs (the fastest way to see when Manning didn't play), you could give him 1 game in 2004, 2 in 2005, and most of one in 2007. That's it.

And that's not enough to get him past Brady (though it would push 2004 to the #2 spot).

38 Re: Final 2007 DVOA Ratings

I know there was a Giants thread a few weeks back regarding their ranking, but my comments never got submitted, so here goes:

As a Giants fan, I am frustrated by seeing them behind teams like Philly and Washington when the Giants can, should and have beaten both teams. But the way the Giants have played in general, DVOA is right on correctly. They are a sloppy team. Just look at their -9 turnover differential. They actually won 4 games with a negative differential, including a -4 (Chi), -2 (Wash), -2 (Atl), -1 (Phi). That doesnt scream elite. It screams luck, and potential. If they would have played crisply in any of those games, they (probably) would have crushed their opponents [and had a much higher DVOA].
Instead, they turn the ball over too much, take too many ill-timed penalties (though overall penalty numbers look good), have kickoffs out of bounds... just play sloppy. Do they have potential? Sure. They actually won those aforementioned games while playing poorly. With a few breaks they could have beaten Dal, GB and New England. But they deserve to be ranked right in the middle of the pack. They haven't shown they can play a full 60 minutes of solid football. If they do, they'll be a force in the playoffs.

Looking at the year, their opponenets can be broken down into 4 (subjective) categories- The Giants have

1. Beaten ALL the teams that they clearly should have beaten including the Jets, SF, Atl, Miami [The Miami win was a bit sloppy as was the Jets win].

2. Beaten ALL the mediocre teams including Chicago, Detroit and Buffalo. All of those wins were sloppy. Too many turnovers (TEN in 3 games!), lost drives inside the redzone, kickoffs out of bounds...

3. Gone 3-2 vs the next tier including Philly, Washington and Minnesota. Even in their 3 wins, they were down 14 to the Skins, and barely held off the Eagles in their second meeting.
And they got demolished in the 2 losses.

4. Gone 0-4 vs the top teams including Dallas, GB and New England. But the Giants played well and had a chance to win ALL of those games. (yes, both Dal games were 3 point games in the 4th Q and the GB game was a one point game at the half).

Do I think they could beat most teams ahead of them on this DVOA rankings list? Of course. If they play a crisp, clean game and win the turnover battle, they can beat Tampa, Dallas and GB. They have shown that they are a couple of plays away from doing so. But they have played too many sloppy games to think they can pull out 3 (or 4) in a row.

Here's to hoping the Pats game gets the team to really focus and play a full game.

39 Re: Final 2007 DVOA Ratings

Nate, I think you and I see the team the same way. They are a lower top 10 team. Not a top 5, by any stretch. But you put them against a team in the bottom half of the league, they will win.

DVOA has them in the bottom half.

That is where DVOA measures them wrong. Not sure what it is missing, or if DVOA would be more accurate overall if it had them pegged better (Probably not). But DVOA does not have them pegged right IMO.

Now watch them get skunked next week. :-)

40 Re: Final 2007 DVOA Ratings

Definitely agree with 2007 having "no great defenses". Teams like the Colts, Bucs, Steelers and Pats are all close at the top in defensive points allowed, but I don't see any of these defenses being anywhere as good as the top D's in recent seasons. The 07 Steelers have the stats, but with my eyes I know this defense is inferior to the one they had in 2004 and 2005 (especially the way they are playing now). The Patriots still make plays when they need them and can shut down the Lemon/Losman offenses, but does anyone see them as being as good as the defense they had in 03 and 04? This is the best Indy defense in the Manning era, but do you expect them to stop the Patriots when it counts in a few weeks? I don't.

And with so many bad teams and bad defenses in general, teams have taken advantage of it all year, especially against the pass. The top 4 seeds in the league (NE, IND, DAL, GB) all featured QB's with over 4000 yards passing.

Some records set in 2007:
- League-wide amount of points scored this year (11,104)
- 8 different teams scored at least 400 points
- *61.16% of all passes were completed. The previous record was 59.78% in 2006.
- Drew Brees completed 443 passes (previous record was 418 by Rich Gannon in 2002)
- In fact, 6 of the top 38 completion marks in a season were in 2007 (3 in the top 14).
- *The cummulative 82.6 passer rating on the season was the 2nd highest ever behind 2004 (82.8).
- *720 TD passes were thrown, 2nd only to 2004's amount of 732
- *Four QB's threw at least 31 TD passes in the same season for the first time ever. If Manning and Roethlisberger didn't basically take week 17 off, we'd probably be saying 4 QB's with at least 33 or so.
- *7 QB's threw for at least 4000 yards (previous record: 5 in three different seasons). Hasselbeck came up 34 yards short and didn't play the second half against Atlanta

**The amazing thing about these passing records were that QB's like Kyle Orton, A.J. Feeley, Brodie Croyle, Tarvaris Jackson, Vince Young, Kellen Clemens, Rex Grossman, John Beck, etc. threw a significant amount of passes. If teams were healthier at QB and didn't have to start a non-strike record amount of QB's in a season, the numbers would have been even better.

RB stats suffered for multiple reasons:
- Too many injuries to key guys and a lot of underperformers
- Teams found it was so much easier to pass the ball
- First time since 1993 a RB didn't top 1500 yards rushing

Receiver stats were crazy
- 23 receivers had a 1000 yard season, and I believe 9 of them were first-timers to the mark. That's not a record (there were 26 in 1999 and a few seasons with 24), but that's still a lot and some surprising names to boot (Brandon Marshall with over 1300 yards?)
- 11 different receivers had at least 10 TD catches, tying the mark set in 1995 and 2004 (the two other pass-heavy seasons in league history)

Offense will win another championship in 2007

41 Re: Final 2007 DVOA Ratings

Let’s give a really simple example. How many other teams could have given the Pats that kind of game?

Actually, going back to the Indy game in week 9, I would say that every Patriots opponent has thrown everything but the kitchen sink at them. Just off the top of my head, the Eagles, Ravens, Steelers, Jets, and Giants all played very good football against the Pats and were clearly fired up and armed with unexpected or unusual defensive gameplans. Four of those games were nationally televised night games.

The Pats players said Saturday night at the Giants was a playoff-intensity game.

There's a tendency to start thinking that the best teams in the league can just blow out every other team without even breaking a sweat. That's not the way the NFL works. The best teams are the best teams because they beat the other solid teams by a little bit more often -- usually in the fourth quarter.

42 Re: Final 2007 DVOA Ratings


game don't natter what do yuo want Seahwk,s coach to do? he team not going to move up in rankings of sadium wear to keep Godzilla out of picture Too much L jordan. Go with j fargas more in 2008 or draft D Mafadden if not use fargas sign small guy like D sroles, let him throw again becuase this is raferts and this is raidders football. Too drunk to write more. trdae notions of Raiders tomorrow ehn Fx of hardcre drinking effects wear off

43 Re: Final 2007 DVOA Ratings

The Patriots still make plays when they need them and can shut down the Lemon/Losman offenses, but does anyone see them as being as good as the defense they had in 03 and 04?

No. Not as good as 03/04. But, it's not the same game. Since the '03 and '04 seasons, the NFL has effectively banned pass defense and instructed the officials to stop calling offensive holding. What you are seeing is the result of the rules changes.

If you've noticed, the good offenses in the NFL are just eating defenses alive with slot receivers and tight ends. Think about it. How is a safety supposed to cover a 250 pound tight end without touching him? Under the current rules, the smart defensive coaches are just conceding the underneath passes and playing cover 2 to limit the big plays. There isn't much else you can do without giving up PI call after PI call. There's no such thing as a shutdown corner with the current rules.

It's just taken a couple of seasons for the coaching to adjust to the new rules. Wanna see defense again? Take it up with Polian and the competition committee.

It's quite evident that Belichick understood the implications of the ban on pass defense and retooled his offense accordingly. What was it against the Steelers? 34 consecutive called pass plays?

44 Re: Final 2007 DVOA Ratings

#42: Now, if we could all be drunk when we read Raiderjoe's posts, they would all make sense.

45 Re: Final 2007 DVOA Ratings

I mentioned this elsewhere but Packer GM Ted Thompson deserves a lot of credit for the special teams improvement. He has made acquiring players capable of contributing on special teams a priority and it shows. Yes guys have to be coached. But it starts with guys both capable and willing to perform on special teams. And that's Thompson.

46 Re: Final 2007 DVOA Ratings


Very interesting post about all the offensive records broken and the quality of offenses this year. But where do you draw the conclusion that a good offense is what wins the championship?

I'm not saying your wrong (after so many folks elect the Patriots champions anyway), I'm just saying your logic doesn't follow. I could just as strongly make the case that the last man standing will be the one who shuts down the shootouts (any given Sunday).

47 Re: Final 2007 DVOA Ratings

The cynic in me applauds Dungy for "resting his starters" and playing sleight of hand. By letting the Titans into the playoffs, the Colts will get to watch San Diego get injured against Tennessee, and either play the now-injured Chargers, or watch Tennessee try to cripple New England players. A much better Colts outcome than working to let the speedbump Browns into the playoffs.

48 Re: Final 2007 DVOA Ratings

You're info isn't correct. I don't know if there's a problem in looking at Sorgi gamelogs because he's played in several quarters without throwing a pass, but when you look at the play by play here is when Manning sat in his 49 TD season:

2004: 1 qtr vs. CHI, 1 qtr vs. Det, 1 qtr vs. TEN, 4 qtrs vs. DEN (all but one drive) so in total Manning sat for 7 quarters in 2004, and as I've mentioned in a previous post he had 2 fourth quarter TDs in his record breaking year with his team up by 14+ points.

Brady had a great season and the record is a great accomplishment for him, I don't know how close Manning would have gotten in DPAR had he played as many qtrs as Brady. One interesting thing is that they have almost identical DPAR/pass attempts. Manning in 2004 had .33418 DPAR/PA and Brady had .33422 this year. Peyton just narrowly edged him out in DOVA 62.8% to 62.0%. I guess the past 2 years have given proof to two of the what if statements of the irrational debate, that given a good defense (if only for the playoffs) Manning could win a Superbowl and given
a stellar supporting cast on offense Brady could put up stats like Manning could.

49 Re: Final 2007 DVOA Ratings


Let me clarify my statement. I believe the Giants are a top 10 team, but not a top 10 DVOA team. They have played down to their inferior competition, which has brought down their DVOA. Yes they were able to pull out some victories, but DVOA doesnt care whether you win, it cares how you play.
If they had gone out and crushed the Bears, Lions and Bills, they might be a top 10 DVOA team right now. But they didnt. They had too many dropped passes and turnovers. Thus a low DVOA but a W.

Does that mean they can't beat Tampa, Dal and/or GB? Of course not. If the Giants stop dropping passes and win the turnover margin, they should be anyone in the NFC. Why? Becasue they are a top 10 team. Just not in DVOA.

50 Re: Final 2007 DVOA Ratings

#20:Do the Bills have the biggest difference between expected wins and pythagorean wins this year?

No. The Bills finished 1.3 wins behind the Pythagorean projection. The Dolphins were worst at 3.1 wins below projection. The Cowboys were best, at +2.2.