Week 9 DVOA Ratings

by Aaron Schatz

In the end, the Game of the Century came out pretty much exactly how DVOA said it would.

Last week, New England led Indianapolis 70.6% to 57.0%.

New England's DVOA for the Game of the Century was 79.8%, only slightly higher than their season DVOA to that point. Indianapolis's DVOA for the Game of the Century was 31.9%, a little bit lower than their season DVOA to that point.

One week later, New England's lead over Indianapolis is only slightly higher: 70.9% to 53.9%. The biggest effect that this game had on DVOA is that New England and Indianapolis are now listed with two of the six hardest schedules in the league so far, and both teams now have higher DVOA than VOA. (Last week, only Indianapolis did.)

Then, as if they had to prove to the Colts and Patriots that they still existed, the Cowboys and Steelers went on national television and opened up a gigantic can of whup-ass on two hated division rivals. The result: A gap between the top four teams and the other 28 teams that is like nothing we've ever seen before. But we'll get to that after I run the numbers...

Here are the Football Outsiders team efficiency ratings through nine weeks of 2007, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here [1].) 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.

Opponent adjustments are currently set at 90% and will increase each week until they are full strength after Week 10.

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

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

	TEAM	TOTAL DVOA	LAST WEEK	NON-ADJ TOT VOA	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
1	NE	70.9%	1	68.7%	9-0	47.0%	1	-17.6%	3	6.3%	4
2	IND	53.9%	2	46.7%	7-1	40.5%	2	-14.1%	4	-0.8%	17
3	DAL	44.0%	3	41.5%	7-1	29.9%	3	-13.8%	5	0.4%	13
4	PIT	36.1%	4	45.5%	6-2	13.6%	6	-23.0%	2	-0.5%	16
5	TEN	20.9%	8	20.9%	6-2	-6.5%	22	-30.7%	1	-3.3%	26
6	TB	19.0%	6	13.1%	5-4	15.8%	4	-5.1%	10	-2.0%	18
7	GB	16.1%	7	22.2%	7-1	11.1%	9	-4.7%	12	0.3%	14
8	MIN	13.7%	16	13.2%	3-5	4.4%	14	-5.2%	9	4.1%	9
9	JAC	10.4%	9	5.8%	5-3	8.6%	11	3.0%	17	4.8%	6
10	SEA	10.1%	10	17.0%	4-4	-2.8%	19	-3.1%	13	9.9%	1
11	NYG	6.3%	12	13.3%	6-2	6.7%	12	-5.0%	11	-5.4%	29
12	PHI	5.0%	11	3.4%	3-5	12.7%	8	5.3%	20	-2.4%	22
13	SD	4.8%	5	4.9%	4-4	3.6%	15	5.1%	19	6.3%	3
14	WAS	0.6%	13	-5.3%	5-3	-10.3%	26	-9.9%	7	0.9%	12
15	CLE	-0.9%	15	-2.3%	5-3	13.1%	7	18.5%	30	4.5%	8
16	BUF	-2.1%	18	-7.5%	4-4	-7.1%	23	1.4%	14	6.4%	2
	TEAM	TOTAL DVOA	LAST WEEK	NON-ADJ TOT VOA	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
17	CIN	-3.8%	17	-3.5%	2-6	15.7%	5	17.4%	29	-2.1%	20
18	DET	-7.0%	22	2.0%	6-2	1.5%	17	2.8%	16	-5.6%	30
19	NO	-7.8%	23	-13.5%	4-4	9.2%	10	11.0%	23	-6.0%	31
20	BAL	-8.3%	14	2.0%	4-4	-19.9%	29	-13.7%	6	-2.1%	19
21	ARI	-9.5%	19	-2.0%	3-5	-3.5%	20	5.9%	21	-0.1%	15
22	KC	-13.0%	21	-5.8%	4-4	-17.3%	27	-6.6%	8	-2.3%	21
23	HOU	-13.6%	24	-11.7%	4-5	-2.7%	18	12.7%	25	1.8%	11
24	DEN	-14.1%	20	-18.9%	3-5	4.7%	13	16.4%	27	-2.4%	23
25	CAR	-15.7%	25	-12.4%	4-4	-7.5%	24	1.7%	15	-6.5%	32
26	ATL	-20.6%	28	-13.3%	2-6	-8.4%	25	9.4%	22	-2.8%	24
27	CHI	-22.3%	26	-26.2%	3-5	-23.4%	31	3.6%	18	4.7%	7
28	MIA	-23.2%	27	-30.7%	0-8	2.8%	16	23.0%	31	-3.1%	25
29	NYJ	-25.7%	30	-27.5%	1-8	-5.1%	21	26.4%	32	5.8%	5
30	OAK	-35.1%	29	-27.2%	2-6	-17.8%	28	12.5%	24	-4.7%	28
31	STL	-41.1%	31	-41.5%	0-8	-22.3%	30	14.5%	26	-4.4%	27
32	SF	-50.4%	32	-42.7%	2-6	-36.2%	32	17.3%	28	3.1%	10

ESTIMATED WINS [2] uses a statistic known as "Forest Index" that emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close. It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles. Teams that have had their bye week are projected as if they had played one game per week.
PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
VARIANCE [3] measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from least consistent (#1, highest variance) to most consistent (#32, smallest variance).

	TEAM	TOTAL DVOA	W-L	ESTIM. WINS	RANK	WEIGHTED DVOA	RANK	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
1	NE	70.9%	9-0	9.0	1	71.2%	1	5.3%	6	-1.7%	17	2.9%	31
2	IND	53.9%	7-1	9.0	1	53.1%	2	7.8%	3	-7.8%	23	5.9%	22
3	DAL	44.0%	7-1	7.4	3	44.5%	3	0.8%	14	-2.8%	19	10.2%	15
4	PIT	36.1%	6-2	5.9	6	33.4%	4	-8.8%	31	-3.1%	20	34.1%	1
5	TEN	20.9%	6-2	5.9	5	21.4%	5	-1.1%	19	-0.2%	15	5.2%	28
6	TB	19.0%	5-4	5.8	7	19.2%	6	1.6%	13	-18.3%	31	22.7%	4
7	GB	16.1%	7-1	5.9	4	15.6%	7	-2.1%	22	-10.1%	25	3.8%	29
8	MIN	13.7%	3-5	5.6	9	13.8%	8	0.8%	15	-15.1%	29	18.0%	8
9	JAC	10.4%	5-3	5.6	8	10.3%	9	2.7%	12	7.0%	10	5.6%	24
10	SEA	10.1%	4-4	4.9	12	9.2%	10	-6.5%	30	-23.3%	32	13.7%	11
11	NYG	6.3%	6-2	5.1	10	7.7%	11	-6.0%	28	14.7%	4	6.2%	21
12	PHI	5.0%	3-5	5.1	11	2.7%	13	2.9%	11	14.1%	5	18.4%	6
13	SD	4.8%	4-4	4.6	17	5.6%	12	0.3%	17	1.1%	14	25.0%	3
14	WAS	0.6%	5-3	4.4	18	0.2%	16	3.7%	9	15.4%	3	9.9%	16
15	CLE	-0.9%	5-3	4.8	16	0.8%	15	0.6%	16	-11.0%	27	5.5%	25
16	BUF	-2.1%	4-4	4.8	14	1.0%	14	8.2%	2	6.6%	11	11.8%	13
	TEAM	TOTAL DVOA	W-L	ESTIM. WINS	RANK	WEIGHTED DVOA	RANK	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
17	CIN	-3.8%	2-6	4.9	13	-4.7%	18	7.5%	4	-10.9%	26	1.8%	32
18	DET	-7.0%	6-2	4.8	15	-6.0%	19	-6.2%	29	11.2%	7	19.9%	5
19	NO	-7.8%	4-4	4.3	19	-4.5%	17	3.1%	10	-14.1%	28	9.2%	17
20	BAL	-8.3%	4-4	3.6	24	-8.3%	20	-10.8%	32	21.1%	1	6.2%	20
21	ARI	-9.5%	3-5	3.6	25	-9.9%	21	-5.5%	27	-17.4%	30	28.4%	2
22	KC	-13.0%	4-4	3.8	21	-11.6%	22	-3.3%	23	-2.3%	18	8.9%	18
23	HOU	-13.6%	4-5	3.6	23	-16.3%	24	-2.0%	21	11.6%	6	14.5%	10
24	DEN	-14.1%	3-5	3.7	22	-16.5%	25	8.6%	1	-8.2%	24	18.4%	7
25	CAR	-15.7%	4-4	3.9	20	-14.5%	23	0.1%	18	3.0%	13	11.6%	14
26	ATL	-20.6%	2-6	3.0	27	-19.9%	26	-4.0%	25	4.0%	12	3.2%	30
27	CHI	-22.3%	3-5	3.1	26	-21.5%	27	6.3%	5	-1.4%	16	5.6%	23
28	MIA	-23.2%	0-8	2.6	28	-24.5%	28	5.2%	7	10.0%	8	5.4%	26
29	NYJ	-25.7%	1-8	2.6	29	-24.6%	29	4.8%	8	19.3%	2	5.4%	27
30	OAK	-35.1%	2-6	1.5	31	-34.7%	30	-5.1%	26	7.1%	9	15.0%	9
31	STL	-41.1%	0-8	1.6	30	-41.1%	31	-1.3%	20	-4.3%	22	13.1%	12
32	SF	-50.4%	2-6	1.2	32	-51.9%	32	-3.9%	24	-4.0%	21	8.9%	19

Every week, I've been running a little feature tracking how the Pats and Colts were doing compared to the best teams ever in DVOA, going back to 1996. This week, we're going to change it a little bit. First of all, we'll do just one list -- the best teams through Week 9, not nine games. After next week, the bye won't mean the Pats and Colts have played a different number of games. We're also going to extend the lists a bit past the Top 10 to demonstrate just how incredible this season is. We're also adding a third list, because while nobody was looking, the Tennessee Titans have built what might be a 2006 Bears/2000 Ravens-level defense. Really.

Best DVOA Ever Watch

BEST TOTAL DVOA AFTER WEEK 9			BEST OFFENSIVE DVOA AFTER WEEK 9			BEST DEFENSIVE DVOA AFTER WEEK 9
2007	NE	70.9%	2007	NE	47.0%	2002	TB	-41.2%
2007	IND	53.8%	2000	STL	43.7%	1996	GB	-38.4%
1996	GB	49.4%	2000	IND	43.1%	1998	OAK	-32.7%
2001	STL	46.8%	2007	IND	40.5%	1999	JAC	-31.3%
1999	STL	46.5%	2004	IND	40.3%	2007	TEN	-30.7%
2001	PHI	46.5%	1999	WAS	36.3%	1997	SF	-29.9%
2007	DAL	44.0%	1998	DEN	33.0%	2006	BAL	-29.0%
2000	TEN	43.7%	2006	IND	32.7%	1996	DEN	-28.7%
1999	JAC	43.1%	2002	SF	32.1%	2000	BAL	-27.1%
2002	TB	42.6%	2004	KC	30.9%	2003	MIA	-27.0%
2005	IND	40.1%	2005	DEN	30.4%	2000	TB	-26.9%
2003	KC	39.9%	2002	KC	30.3%	2006	CHI	-26.6%
1997	SF	39.1%	2001	STL	30.2%	2000	TEN	-25.6%
1998	DEN	38.7%	2007	DAL	29.9%	2001	PHI	-25.2%
2007	PIT	36.1%	1998	MIN	29.7%	2003	TB	-25.1%

How good is the Tennessee defense? Only two teams have ever had a defensive DVOA of -30% or better for a full season: the 2002 Bucs and the 2000 Ravens. I don't know if the Titans can keep this up. This is the first week all year where they've had a rating better than -30%. Still, it is amazing that we even have to start the conversation. Nobody is talking about Tennessee. Even hardcore fans like those who read FO probably can't name more than three members of their starting lineup. You all know Keith Bulluck and Albert Haynesworth. Maybe you said Chris Hope? Nick Harper? Kyle Vanden Bosch? If you read the Top 25 Prospects list in PFP 2007, you may remember Cortland Finnegan and Stephen Tulloch (although Tulloch is not starting). Even more incredible, the Titans are doing this with their most talented defensive player suspended for the season! And yes, Pacman Jones is that good, and I do not doubt that this team would be even better on defense with him around.

The Titans defense does not stand out in any one area. The Titans rank fifth or better against the pass, against the run, on first down, on second down, on third down, in the first quarter, in the third quarter, and in the fourth quarter. They are seventh in the second quarter and eighth in the red zone.

And yet, despite having a defense that good, the Titans can't even sniff the top four teams in overall DVOA because a) the Tennessee passing game is so bad and b) the other four teams are so good. Which brings us to the next issue on our agenda...

All year long, I've been trying to get across the idea that the Patriots were not the only team enjoying a season of historical greatness. I've talked a lot about the Colts. Now it is time to talk about the Cowboys and Steelers as well.

Despite losing to the Pats in the Game of the Century, the Colts are still the second-best team we've ever measured through Week 9. No team had ever put up a DVOA over 50% through Week 9, but the Colts are doing that and the Pats are going far beyond just doing that. Right below them are Dallas and Pittsburgh. This year we are seeing four of the top 15 teams ever measured by DVOA through half a season.

The 2007 Cowboys would have led the league in DVOA at midseason any of the past five seasons, and are the best team ever ranked third through Week 9. The 2007 Steelers would have been leading the league in DVOA last year at this time, and they are the best team ever ranked fourth through Week 9. (Ironically, because the top of the league was so bunched up at midseason last year, the second-best teams ever ranked third and fourth through Week 9 are the 2006 Chargers and 2006 Giants.)

There's a colossal gap between that and the Titans. The 2007 Titans are the NINTH-best team to be ranked fifth after Week 9. (The best was the 2005 Bengals.) The Bucs and Packers are the seventh-best teams to ever be ranked sixth and seventh after Week 9.

In any other year, where the Pats and Colts were just regular ol' good teams, the main conversation in football would be whether any team other than Green Bay could derail yet another classic Dallas-Pittsburgh matchup in the Super Bowl. NFL Network would be digging out tape from Super Bowl XIII and Super Bowl XXX. Terry Bradshaw would be razzing Jimmy Johnson about having "only" two rings. ESPN would be interviewing Tony Dorsett and Lynn Swann and even Neil O'Donnell.

This domination by four teams does not seem to signify an overall decline in parity across the league. The whole NFL hasn't spread out, away from 0% DVOA. At the bottom, only the 49ers are jousting with historical ineptitude. The Jets, the Dolphins, and even the Rams are just run-of-the-mill bad teams, no worse than the worst teams of past seasons (or at least, past midseasons).

Just to show the domination of the top four in one other way, here is a table giving the average DVOA rating of the teams ranked 1-4 and the teams ranked 5-8 after Week 9 of every season for which we have the play-by-play breakdowns.

YEAR	TOP 4	NEXT 4	DIF	YEAR	TOP 4	NEXT 4	DIF
2007	51.2%	17.4%	33.8%	2003	29.4%	16.7%	12.7%
1996	36.0%	16.9%	19.1%	2000	35.0%	22.4%	12.7%
2001	33.9%	15.4%	18.5%	2006	34.4%	22.0%	12.4%
1997	27.7%	12.8%	15.0%	1998	29.9%	18.4%	11.6%
1999	35.4%	20.5%	14.9%	2002	30.3%	19.0%	11.3%
2004	30.9%	16.2%	14.7%	2005	32.9%	26.8%	6.1%

The greatness of the top four teams is screwing with our perceptions of the rest of the league. Let me give you an example: the Philadelphia Eagles. Every week, the Eagles seem to look worse and worse and yet they stay high in DVOA. This is the second straight year this has happened, but all the declining veterans are a year older now and it sure doesn't seem like there is a late-season surge coming. DVOA ranks the Eagles 12th, and that team I saw on the field at the Linc Sunday night sure didn't look like one of the top dozen teams in football.

However, the Eagles looked that bad in large part because the Cowboys are so good. The middle of the league is pretty bunched together right now. It seems ridiculous to have them 12th, but look at the teams ranked below them and ask yourself: which of these teams would you really favor to beat the Eagles on a neutral field? The Chargers, maybe... but the Chargers are ahead of the Eagles in weighted DVOA anyway. The Redskins? Jason Campbell is struggling as much as Donovan McNabb lately. I don't think the Eagles are really better than the Browns. Brian Westbrook is the kind of player who should kill the Bills defense. The Bengals don't even have a defense. The Lions? Didn't the Eagles totally beat them down a few weeks ago? Remember, that game counts in the ratings too.

Once you look closer at the rest of the league, Philadelphia being 12th in regular DVOA and 13th in weighted DVOA really isn't that strange.

Housekeeping: The premium database is updated, the team stats pages are updated, the individual stats pages are updated, and the playoff odds are updated (Pats: 22.8 percent chance of going undefeated). The Loser League will be updated after we announce the winner of Loser League Part I in tomorrow's Scramble for the Ball. You can also now register your team for Loser League Part II [4].

For (short) comments on every team, look for DVOA on AOL, every Wednesday. (This will be linked on the FO Goes Mainstream [5] page.)

(And people doubted me when I said that eventually I would go back to writing the really long, rambling DVOA commentaries...)