Week 13 DVOA Ratings

by Aaron Schatz

Last night, the New England Patriots were completely and totally outplayed. A win is a win in the standings, but in DVOA, the Patriots took a major hit. Here are the ratings for last night's game, first without opponent adjustments:

TEAM	OFF	DEF	ST	TOTAL
BAL	30.2%	2.6%	5.4%	33.0%
NE	2.6%	30.2%	-3.4%	-31.0%

...and with opponent adjustments:

TEAM	OFF	DEF	ST	TOTAL
BAL	37.8%	-39.6%	5.4%	82.8%
NE	11.5%	42.5%	-3.4%	-34.4%

Prior to last night's game, the Patriots had a DVOA above 45% in every single game this year, and the Ravens had a DVOA below 20% in every single game this year. One of the big reasons for the DVOA difference: Forcing a fumble on an interception return is essentially a fluke play, and isn't counted in DVOA.

Some other changes thanks to last night's game:

New England's overall DVOA fell nearly 10 percentage points, while Baltimore rose from 24th overall to 17th overall.
New England no longer qualifies as the top offense of the DVOA Era, falling behind the 2004 Colts. They still rank as the best overall team of the DVOA Era by a healthy margin.
The Patriots no longer have all three units in the top five, with the defense and special teams falling to eighth and seventh, respectively.
The Ravens rise to second overall in defense, although the gap between Pittsburgh and Baltimore is larger than the gap between second-place Baltimore and tenth-place San Diego.
New England rose from 32nd to 19th in VARIANCE, based on just one game. That's how consistent they were before Monday night.
Primarily due to this change in VARIANCE, the Patriots are no longer perfect in "estimated wins," going from 11.0 to 11.3 instead of 12.0.
In two games, New England's pythagorean winning percentage [1] has gone from .907 to .887 to .872. Two weeks ago, they had the highest pythagorean winning percentage of any team since the 1969 Minnesota Vikings (.923). They have fallen behind seven teams since the 1970 merger, and numerous teams from before that. Most of those teams are from the seventies, however; the only team to ever have a pythagorean winning percentage above .870 in a 16-game season is the 1985 Bears (.878).

I know a lot of people will want to talk about the specifics of this game. I posted an Extra Point last night with my thoughts [2], and I would prefer that people use that thread for any arguments specific to that one game (playcalling, timeouts, officiating, etc.). That keeps this thread free to talk about the season in general.

The Patriots' rise in VARIANCE has another strange result: Based on DVOA, the most consistent team of 2007 is now the Jacksonville Jaguars. Yes, the same team that last year was ridiculously inconsistent, blowing out playoff teams and losing to also-rans. Last year at this time, the Jaguars had six games with DVOA above 40% and three games with DVOA below -40%. This year, the Jaguars have one game with DVOA above 40% and no games with DVOA below -40%. Jacksonville has put up a DVOA rating between 25% and 40% in five of their last six games. (The exception was the loss to New Orleans with Quinn Gray at quarterback, which is their worst game of the year.)

One other note: Cincinnati rises another three spots this week, from 15th to 12th, despite losing to Pittsburgh in a game that was not really close. I'm totally at a loss here. I'll try to look at more splits to figure out why Cincinnati is so high, but it doesn't seem to make any sense.

* * * * *

Here are the Football Outsiders team efficiency ratings through 13 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 [3].) 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.

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	WEIGHTED DVOA	RANK	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
1	NE	61.4%	1	59.0%	1	12-0	47.0%	1	-9.7%	8	4.7%	7
2	DAL	41.9%	2	43.7%	2	11-1	31.0%	2	-10.0%	7	0.9%	14
3	IND	31.1%	3	24.1%	4	10-2	25.6%	3	-10.6%	6	-5.1%	30
4	PIT	25.1%	5	18.5%	6	9-3	3.3%	14	-24.1%	1	-2.2%	21
5	GB	24.9%	4	26.3%	3	10-2	21.7%	4	-0.6%	15	2.5%	11
6	TB	19.4%	6	20.7%	5	8-4	13.1%	8	-8.1%	9	-1.8%	18
7	SD	16.4%	7	18.5%	7	7-5	5.4%	13	-4.8%	10	6.2%	3
8	JAC	15.8%	8	17.2%	8	8-4	15.3%	6	3.3%	19	3.7%	9
9	MIN	15.2%	10	16.5%	9	6-6	7.6%	10	-3.9%	12	3.8%	8
10	SEA	15.1%	9	13.9%	10	8-4	1.2%	15	-11.1%	4	2.8%	10
11	CLE	7.1%	13	12.6%	11	7-5	13.5%	7	13.7%	30	7.3%	2
12	CIN	5.6%	15	5.1%	12	4-8	16.4%	5	10.1%	25	-0.7%	15
13	PHI	3.4%	12	0.0%	15	5-7	9.6%	9	2.6%	17	-3.6%	27
14	WAS	1.1%	14	0.3%	14	5-7	-4.2%	20	-3.3%	13	2.0%	12
15	NYG	1.0%	11	0.8%	13	8-4	-0.7%	16	-4.0%	11	-2.3%	22
16	TEN	-2.8%	17	-6.6%	20	7-5	-10.4%	24	-11.5%	3	-3.9%	28
	TEAM	TOTAL DVOA	LAST WEEK	WEIGHTED DVOA	RANK	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
17	BAL	-3.3%	24	-3.5%	17	4-8	-15.0%	27	-14.4%	2	-2.7%	23
18	BUF	-4.3%	19	-1.2%	16	6-6	-5.3%	21	4.3%	20	5.3%	5
19	ARI	-6.4%	18	-7.4%	21	6-6	-2.1%	18	1.2%	16	-3.1%	24
20	DEN	-7.9%	16	-8.9%	22	5-7	5.5%	12	10.0%	24	-3.4%	26
21	CHI	-8.9%	25	-5.4%	19	5-7	-19.4%	30	-1.5%	14	9.0%	1
22	NO	-10.1%	21	-4.0%	18	5-7	5.8%	11	12.7%	27	-3.2%	25
23	KC	-11.7%	23	-9.2%	23	4-8	-20.6%	31	-10.8%	5	-1.9%	19
24	HOU	-12.7%	20	-15.9%	25	5-7	-1.1%	17	13.2%	28	1.6%	13
25	NYJ	-16.4%	27	-11.7%	24	3-9	-6.3%	22	15.3%	32	5.3%	6
26	DET	-17.2%	22	-18.7%	26	6-6	-2.5%	19	9.1%	21	-5.6%	31
27	OAK	-23.9%	29	-20.9%	27	4-8	-12.7%	25	9.8%	23	-1.4%	17
28	ATL	-24.7%	28	-24.4%	28	3-9	-14.1%	26	9.6%	22	-1.0%	16
29	MIA	-25.1%	26	-27.4%	29	0-12	-9.6%	23	13.5%	29	-2.0%	20
30	CAR	-27.7%	30	-28.9%	30	5-7	-17.8%	28	2.8%	18	-7.2%	32
31	STL	-32.3%	31	-30.8%	31	3-9	-18.1%	29	10.2%	26	-3.9%	29
32	SF	-44.7%	32	-45.9%	32	3-9	-34.8%	32	15.3%	31	5.4%	4

ESTIMATED WINS [4] uses a statistic known as "Forest Index" that emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close. It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles.
PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
VARIANCE [5] 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	NON-ADJ TOT VOA	ESTIM. WINS	RANK	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
1	NE	61.4%	12-0	59.6%	11.3	1	4.1%	8	-5.1%	19	11.8%	19
2	DAL	41.9%	11-1	42.6%	9.9	2	1.6%	15	-13.5%	22	9.4%	23
3	IND	31.1%	10-2	30.5%	9.5	3	2.4%	11	-14.2%	23	14.3%	12
4	PIT	25.1%	9-3	30.6%	7.5	8	-5.2%	25	13.9%	12	28.5%	1
5	GB	24.9%	10-2	27.0%	8.6	4	1.6%	17	-27.4%	30	9.5%	22
6	TB	19.4%	8-4	21.5%	8.0	6	-5.2%	26	-36.6%	32	16.4%	9
7	SD	16.4%	7-5	11.2%	7.0	11	5.2%	6	-17.2%	25	21.6%	6
8	JAC	15.8%	8-4	14.4%	8.1	5	1.6%	16	-13.1%	21	5.3%	32
9	MIN	15.2%	6-6	15.1%	7.6	7	0.7%	19	-20.1%	28	24.0%	4
10	SEA	15.1%	8-4	22.6%	7.4	9	-9.1%	32	-20.7%	29	11.9%	18
11	CLE	7.1%	7-5	4.1%	6.8	12	2.0%	12	-19.9%	27	7.5%	29
12	CIN	5.6%	4-8	6.2%	7.1	10	6.6%	4	-31.7%	31	7.7%	28
13	PHI	3.4%	5-7	-4.0%	6.6	13	7.3%	2	9.5%	14	13.5%	15
14	WAS	1.1%	5-7	-9.8%	5.9	16	6.6%	3	16.4%	9	9.4%	25
15	NYG	1.0%	8-4	2.7%	6.3	14	-0.6%	20	20.6%	4	6.8%	30
16	TEN	-2.8%	7-5	-2.8%	5.9	17	-2.5%	21	6.4%	15	18.8%	7
	TEAM	TOTAL DVOA	W-L	NON-ADJ TOT VOA	ESTIM. WINS	RANK	PAST SCHED	RANK	FUTURE SCHED	RANK	VAR.	RANK
17	BAL	-3.3%	4-8	-8.5%	5.5	18	1.9%	14	15.4%	11	14.3%	13
18	BUF	-4.3%	6-6	-15.1%	6.1	15	11.0%	1	-4.5%	18	12.5%	16
19	ARI	-6.4%	6-6	1.9%	5.0	23	-7.4%	31	-17.3%	26	27.0%	3
20	DEN	-7.9%	5-7	-6.1%	5.4	20	1.6%	18	2.4%	17	28.2%	2
21	CHI	-8.9%	5-7	-11.1%	5.3	21	3.1%	10	10.4%	13	7.7%	27
22	NO	-10.1%	5-7	-8.1%	5.4	19	-5.5%	27	-12.2%	20	14.6%	11
23	KC	-11.7%	4-8	-14.0%	4.6	24	3.7%	9	-14.7%	24	14.1%	14
24	HOU	-12.7%	5-7	-11.4%	4.6	25	-4.5%	22	19.5%	6	11.8%	20
25	NYJ	-16.4%	3-9	-19.4%	4.3	26	6.0%	5	18.0%	7	12.1%	17
26	DET	-17.2%	6-6	-12.5%	5.2	22	1.9%	13	23.8%	3	22.4%	5
27	OAK	-23.9%	4-8	-15.9%	3.3	30	-5.2%	24	29.4%	2	14.9%	10
28	ATL	-24.7%	3-9	-18.5%	3.8	27	-5.8%	28	6.0%	16	6.2%	31
29	MIA	-25.1%	0-12	-32.4%	3.3	29	5.2%	7	19.8%	5	9.0%	26
30	CAR	-27.7%	5-7	-19.4%	3.6	28	-7.1%	30	30.8%	1	17.3%	8
31	STL	-32.3%	3-9	-26.4%	2.8	31	-4.8%	23	16.4%	8	10.8%	21
32	SF	-44.7%	3-9	-37.3%	2.3	32	-6.7%	29	15.8%	10	9.4%	24

Best and Worst DVOA Ever Watch

BEST TOTAL DVOA AFTER WEEK 13			BEST OFFENSIVE DVOA AFTER WEEK 13			WORST TOTAL DVOA AFTER WEEK 13			WORST OFFENSIVE DVOA AFTER WEEK 13
2007	NE	61.4%	2004	IND	47.3%	2005	SF	-58.3%	1997	NO	-43.8%
1999	STL	48.5%	2007	NE	47.0%	1999	CLE	-52.6%	2005	SF	-41.9%
2001	STL	45.4%	2002	KC	37.2%	2004	SF	-47.9%	2002	HOU	-36.5%
1998	DEN	45.2%	1998	DEN	34.3%	2000	CIN	-47.4%	2007	SF	-34.8%
2004	NE	43.0%	2000	STL	33.3%	2007	SF	-44.7%	2004	CHI	-34.7%
2007	DAL	41.9%	2000	IND	32.5%	1998	PHI	-43.8%	2004	MIA	-34.2%
2004	PHI	41.8%	2005	IND	32.3%	2000	ARI	-43.6%	1998	PHI	-32.7%
2004	PIT	41.4%	2006	IND	32.2%	2002	ARI	-42.9%	2006	OAK	-32.0%
2005	IND	41.3%	2007	DAL	31.0%	2003	ARI	-41.0%	2000	CLE	-31.3%
2004	IND	40.5%	2004	KC	29.8%	2000	CLE	-39.2%	2001	CAR	-30.2%

No, your eyes do not deceive you. There are four teams from 2004 ranked among the top 10 teams after Week 13. I remember writing at some point in late 2004 that the top four teams were all ranked higher than the top team of the previous year (Kansas City). The Patriots had to face the best set of opponents of any team to ever win the Super Bowl (by win-loss record: 40-8, .833).

(The Steelers had it even harder the next year. Their opponents were 51-13, which is a .797 percentage. That's the fourth-highest of any Super Bowl champion, but of course the Steelers had to play an additional game, and 52 of their 53 players had to play all four games on the road.)

Housekeeping: Team stats pages, the DVOA premium database [6], and the playoff odds [7] are all now updated. Individual stats pages and Loser League [8] will be updated later today. TUESDAY NIGHT UPDATE: BOTH ARE NOW CURRENT.

For (short) comments on every team, remember to look for DVOA on AOL, every Wednesday. (This will be linked on the FO Goes Mainstream [9] page.)

Readers who bought access to the premium DVOA database, or who are interested in access, might want to know about a new view we've added to the page: "Head to Head Matchups." This page will give you a chance to see different splits for two opponents on the same page. Here is a free sample for everyone [10], featuring splits for this week's Philadelphia-New York game.