Week 7 DVOA Ratings

by Aaron Schatz

Once again, the Patriots and Colts are 1-2 in this week's DVOA ratings, and the difference between them comes down basically to one unit: special teams. The Colts didn't actually have such a great offensive game against Jacksonville, so their offensive lead over the Patriots is very slim. The Patriots didn't have the best defensive game against Miami, so their defensive lead over the Colts is very slim. At this point, it is clear that the aberration was last year's Colts defense in the regular season, not in the postseason. The Colts are sixth in defense. SIXTH! Who would have ever thought?

Look specifically at special teams [1], and you can see that the Pats are better in kickoff returns, while the Colts are better in punt returns. The biggest difference between these two teams is that New England is far superior on kickoffs, with better coverage and a kicker who can knock it out of the end zone at times. Irony is a bitch, isn't it?

* * * * * *

Here are the Football Outsiders team efficiency ratings through seven 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 [2].) 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 70% and will increase each week until they are full strength after Week 10.

DAVE is an early-season formula that combines early-season performance with our preseason projection to get a more accurate picture of how well teams will play over the course of the entire season. (DAVE stands for "DVOA Adjusted for Variation Early.") In this week's DAVE ratings, for teams with six games, the preseason projection counts for 7.5 percent, and the current DVOA counts for 92.5 percent. For teams with five games, the split is 19/81. In addition, the weight of Weeks 1-3 has been lowered slightly.

This will be the last week for DAVE. Beginning next week, we will no longer consider preseason projections, and we will re-introduce the usual WEIGHTED DVOA formula that drops the importance of early games.

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	DAVE	RANK	NON-ADJ TOT VOA	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
1	NE	63.0%	1	62.2%	1	68.1%	7-0	43.4%	2	-14.4%	5	5.2%	7
2	IND	56.4%	2	56.9%	2	53.6%	6-0	44.0%	1	-13.5%	6	-1.0%	16
3	DAL	35.4%	4	34.4%	3	35.1%	6-1	24.1%	3	-11.7%	7	-0.3%	13
4	PIT	30.6%	3	27.2%	4	39.6%	4-2	10.4%	8	-21.4%	3	-1.2%	17
5	TB	18.9%	5	18.4%	5	13.2%	4-3	19.0%	4	-1.8%	13	-2.0%	19
6	TEN	17.0%	10	17.4%	7	13.4%	4-2	-1.2%	17	-24.1%	2	-5.9%	30
7	GB	16.7%	6	15.6%	9	16.5%	5-1	8.6%	10	-9.1%	8	-1.0%	15
8	JAC	16.6%	7	17.6%	6	11.4%	4-2	11.7%	6	-1.7%	14	3.2%	9
9	SD	13.8%	13	16.5%	8	11.2%	3-3	8.0%	11	0.6%	16	6.5%	5
10	SEA	13.8%	14	13.4%	11	20.1%	4-3	-0.9%	16	-5.3%	10	9.4%	1
11	WAS	13.5%	9	14.7%	10	12.5%	4-2	-13.0%	26	-24.3%	1	2.2%	12
12	NYG	9.8%	12	11.9%	12	12.2%	5-2	11.7%	7	-3.7%	12	-5.6%	28
13	PHI	8.6%	11	6.2%	13	13.2%	2-4	10.1%	9	-1.2%	15	-2.7%	22
14	BAL	3.0%	8	3.5%	14	14.3%	4-3	-11.2%	25	-15.8%	4	-1.6%	18
15	MIN	2.7%	15	2.0%	15	4.0%	2-4	-4.1%	20	-4.3%	11	2.5%	10
16	CIN	-2.1%	19	-3.0%	16	-0.5%	2-4	13.4%	5	9.0%	22	-6.5%	31
	TEAM	TOTAL DVOA	LAST WEEK	DAVE	RANK	NON-ADJ TOT VOA	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
17	ARI	-3.4%	17	-3.7%	17	3.2%	3-4	-1.3%	18	1.7%	17	-0.4%	14
18	CLE	-5.1%	18	-3.9%	18	-10.4%	3-3	3.7%	14	15.8%	30	7.0%	4
19	HOU	-7.6%	16	-9.8%	20	-10.4%	3-4	-4.9%	21	9.8%	25	7.1%	3
20	KC	-9.2%	21	-8.2%	19	-2.3%	4-3	-16.0%	28	-9.0%	9	-2.2%	20
21	DEN	-10.2%	24	-13.0%	23	-10.8%	3-3	7.1%	12	12.8%	29	-4.5%	27
22	CAR	-11.7%	20	-10.7%	22	3.2%	4-2	-2.6%	19	5.3%	19	-3.7%	23
23	BUF	-13.6%	22	-10.5%	21	-20.2%	2-4	-17.0%	29	6.0%	20	9.3%	2
24	DET	-14.9%	23	-14.5%	24	-9.1%	4-2	-0.1%	15	9.0%	23	-5.8%	29
25	CHI	-21.7%	27	-21.0%	26	-25.5%	3-4	-22.7%	30	3.7%	18	4.7%	8
26	NO	-22.6%	28	-20.0%	25	-33.0%	2-4	-8.2%	23	10.1%	26	-4.2%	25
27	ATL	-25.3%	25	-24.8%	27	-21.1%	1-6	-9.4%	24	12.0%	28	-3.9%	24
28	MIA	-25.5%	26	-26.6%	28	-30.0%	0-7	6.3%	13	29.3%	32	-2.5%	21
29	NYJ	-29.3%	30	-28.4%	29	-30.6%	1-6	-7.2%	22	28.4%	31	6.3%	6
30	OAK	-33.9%	31	-32.8%	30	-23.6%	2-4	-15.8%	27	9.6%	24	-8.4%	32
31	STL	-38.9%	29	-38.4%	31	-42.7%	0-7	-27.2%	31	7.4%	21	-4.3%	26
32	SF	-45.2%	32	-46.8%	32	-43.5%	2-4	-36.7%	32	10.8%	27	2.3%	11

ESTIMATED WINS [3] 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. (Note: This is based on 2007 performance only. In other words, DVOA, not DAVE.)
VARIANCE [4] 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	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
1	NE	63.0%	7-0	6.9	2	-3.8%	25	5.9%	8	3.4%	30
2	IND	56.4%	6-0	7.0	1	1.7%	14	2.9%	13	6.6%	23
3	DAL	35.4%	6-1	5.3	3	-3.5%	22	1.7%	16	7.9%	19
4	PIT	30.6%	4-2	4.3	8	-9.1%	30	-1.9%	20	32.1%	1
5	TB	18.9%	4-3	4.4	6	-0.1%	18	-12.3%	29	29.8%	2
6	TEN	17.0%	4-2	4.3	7	5.2%	6	-1.9%	19	5.3%	24
7	GB	16.7%	5-1	4.7	4	3.8%	10	-13.0%	30	4.5%	27
8	JAC	16.6%	4-2	4.5	5	3.0%	11	5.3%	12	4.5%	28
9	SD	13.8%	3-3	4.2	9	0.7%	16	2.2%	15	19.2%	7
10	SEA	13.8%	4-3	4.0	12	-9.0%	29	-15.5%	32	23.5%	4
11	WAS	13.5%	4-2	4.0	14	-1.2%	19	12.1%	3	7.6%	21
12	NYG	9.8%	5-2	4.1	11	-3.7%	23	5.3%	11	10.1%	15
13	PHI	8.6%	2-4	4.2	10	-3.7%	24	12.4%	2	15.1%	11
14	BAL	3.0%	4-3	3.4	19	-19.7%	32	19.5%	1	4.9%	25
15	MIN	2.7%	2-4	3.8	15	-2.7%	21	-7.1%	25	3.8%	29
16	CIN	-2.1%	2-4	4.0	13	5.2%	7	-5.6%	24	2.6%	32
	TEAM	TOTAL DVOA	W-L	ESTIM. WINS	RANK	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
17	ARI	-3.4%	3-4	3.1	22	-5.0%	27	-13.5%	31	29.1%	3
18	CLE	-5.1%	3-3	3.6	16	5.0%	9	-10.3%	27	11.0%	13
19	HOU	-7.6%	3-4	3.0	23	2.6%	12	5.7%	9	8.9%	18
20	KC	-9.2%	4-3	3.6	17	-4.6%	26	0.6%	17	10.7%	14
21	DEN	-10.2%	3-3	2.9	24	10.0%	2	-5.1%	23	19.6%	6
22	CAR	-11.7%	4-2	3.3	20	-11.3%	31	9.1%	5	16.8%	9
23	BUF	-13.6%	2-4	3.2	21	13.2%	1	2.7%	14	13.0%	12
24	DET	-14.9%	4-2	3.5	18	-1.7%	20	5.6%	10	20.0%	5
25	CHI	-21.7%	3-4	2.3	26	7.6%	4	-2.8%	21	6.6%	22
26	NO	-22.6%	2-4	2.5	25	9.9%	3	-12.2%	28	9.2%	17
27	ATL	-25.3%	1-6	1.9	28	0.6%	17	-1.5%	18	4.7%	26
28	MIA	-25.5%	0-7	2.0	27	5.2%	8	6.3%	7	7.6%	20
29	NYJ	-29.3%	1-6	1.7	29	6.2%	5	11.8%	4	2.9%	31
30	OAK	-33.9%	2-4	1.3	31	-7.3%	28	8.3%	6	16.5%	10
31	STL	-38.9%	0-7	1.3	30	1.5%	15	-4.7%	22	18.3%	8
32	SF	-45.2%	2-4	1.2	32	2.1%	13	-8.2%	26	9.4%	16

Best DVOA Ever Watch

BEST TOTAL DVOA						BEST OFFENSIVE DVOA
AFTER 7 WEEKS			AFTER 7 GAMES			AFTER 7 WEEKS			AFTER 7 GAMES
YEAR	TEAM	DVOA	YEAR	TEAM	DVOA	YEAR	TEAM	DVOA	YEAR	TEAM	DVOA
1999	STL	64.8%	2007	NE	63.0%	2000	STL	53.6%	2000	STL	45.5%
2007	NE	63.0%	1996	GB	55.9%	2007	IND	44.0%	2000	IND	44.9%
2007	IND	56.4%	1999	STL	55.9%	2007	NE	43.4%	2007	NE	43.4%
1996	GB	55.9%	2006	CHI	47.4%	2000	IND	40.3%	2004	IND	37.8%
1998	DEN	45.5%	1998	DEN	43.6%	1999	WAS	39.1%	1999	WAS	36.5%
2006	CHI	45.0%	2001	STL	43.4%	2004	IND	38.6%	1998	DEN	35.0%
2001	STL	43.4%	2005	IND	40.8%	1999	STL	36.1%	2006	IND	34.8%
2005	IND	40.8%	1999	JAC	39.9%	1998	DEN	35.0%	2002	KC	34.4%
1997	SF	40.4%	2002	TB	38.6%	2002	KC	34.4%	2002	SF	34.3%
2000	MIA	40.2%	1997	SF	38.5%	2002	SF	32.4%	2004	KC	31.6%

Yes, the Patriots are now the best team in DVOA history through seven games -- and yet, they could lose that title by next week if the Colts destroy the Panthers. (Check out the difference between Carolina's non-adjusted VOA and DVOA, by the way. Hello, strength of schedule adjustments! Baltimore has a similar issue.)

One place where the Pats are definitely better than the Colts this year is point differential, thanks in large part to their easier schedule so far. Let's look at one more list, the top Pythagorean winning percentages [5] in the NFL since 1950:

Year	Team	W-L	PF	PA	Pyth
1969	MIN	12-2	379	133	.923
1962	GB	13-1	415	148	.920
1968	BAL	13-1	402	144	.919
1976	PIT	10-4	342	138	.896
1970	MIN	12-2	335	143	.883
2007	NE	7-0	279	120	.881
1968	DAL	12-2	431	186	.880
1975	LARM	12-2	312	135	.879
1985	CHI	15-1	456	198	.878
1975	PIT	12-2	373	162	.878

Right now the Pats would have the best Pythagorean winning percentage since the liberalization of passing rules and expansion to the 16-game schedule in 1978. The biggest stumbling block to that record is, of course, Indianapolis.

We're stuck in this strange situation where the press is over-hyping the "everyone is ignoring the Colts" angle -- but that doesn't stop most everyone from predicting a New England victory in two weeks. Don't be so sure. If I were setting the Vegas line right now based solely on DVOA, the Colts would be the favored by a point or two. The Pats are better, but not better enough to be considered favorites given home-field advantage.

There's no question that this game is going to destroy the hype meter. These aren't just two great teams. These aren't just two teams with a long-standing, bitter rivalry. These are two teams who each have a chance to be considered the greatest team in NFL history -- as long as that team can beat the other during the regular season and probably then again in the postseason. These are also two teams who make otherwise intelligent fans turn into completely irrational idiots, and with that in mind, we've created a special discussion thread on our message board [6] for this game, two weeks ahead of time. Please do us a favor. Unless we run articles specifically related to the actual on-field strategy of this game, please keep discussion of Patriots-Colts in the thread on the board. (Apologies to the Redskins and Panthers, who could mess up the story by actually winning next week.)

Meanwhile, around the rest of the league... perhaps the strangest element in this week's ratings is the ascencion of the Seattle special teams to number one. Seattle fans are used to having bad special teams. The Seahawks finished between 18th and 20th for three straight seasons from 2003 through 2005. Then they improved to seventh last year, and this year they are number one.

Oddly, the one guy having a bad year is the guy who was such a big part of last year's improvement: punter Ryan Plackemeier, who is currently 31st in raw punt value, costing the Seahawks 3.8 points worth of field position. Good punt coverage improves the Seahawks to 20th. In every other category of special teams, they are sixth or higher. Josh Brown is a machine this year. Only Jay Feeley has been better on field goals, and only Neil Rackers and Stephen Gostkowski have been better on kickoffs. Seattle is also getting good returns from Josh Wilson and Nate Burleson.

The rest of the FO staff couldn't figure out why our two Seahawks fans, Doug Farrar and Ben Riley, were so negative in our Audibles e-mails after Seattle's ass-whupping of St. Louis this week. The reason is that it is hard to notice when you win a game with special teams. This week, Seattle came very close to breaking that "top 10 special teams games" list from PFP 2006, getting an estimated 12.2 points above average from the special teams. Brown hit all four field goals, including three from 43 yards or more, and his kickoffs averaged 68 yards. Plackemeier had his best game of the year, as not one of his five punts was returned. Nate Burleson had a 91-yard kickoff return for a touchdown. The only problem was the punt Burleson misplayed, which turned into an 80-yard punt for Donnie Jones of the Rams.

For all our talk about the Seahawks being the best team in a bad division, DVOA does not believe they are a bad team. They are part of a pack of reasonably good teams ranked between fifth and 13th. The top teams in the NFC are pretty much all in that pack: Tampa Bay, Green Bay, and the three NFC East teams that are not Dallas. Honestly, how many games can the Eagles play where they are slightly above-average and lose by a small margin?

One more note on special teams: I ran out of time today to address the "how to squibs to Devin Hester affect Chicago's DVOA" question, so I'll do that one in a separate article later this week.

* * * * * *

For (short) comments on every team, look for DVOA on AOL, every Wednesday. (This will be linked on the FO Goes Mainstream [7] page.)

Individual stats pages, offensive line, and defensive front seven are all now updated through Week 7, along with these three pages:

The Playoff Odds Report now also lists the chances of the Patriots or Colts going undefeated, or the Rams or Dolphins going winless.