Writers of Pro Football Prospectus 2008

printable version

DVOA Analysis

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

Need Help with KUBIAK or Premium?

blog advertising is good for you


Most Recent FO Features

DavisTer95-2.jpg

» Word of Muth: Church of the Outside Zone

Word of Muth breaks down film of Alex Gibbs coaching and speaking over a cut-up tape. Find out the secrets of the man who's built big seasons for everyone from Terrell Davis to Warrick Dunn.

21 Dec 2004

Week 15 DVOA Ratings

by Aaron Schatz

Here are the Football Outsiders team efficiency ratings through Week 15 of 2004, measured by our proprietary Value Over Average (VOA) system that breaks down every single play and compares a team's performance to the league averaged based on situation in order to determine value over average.  (Explained further here.)

Moving Up: HOU (20 to 15), plus a lot of teams moved up one or two spots.

Moving Down: NE (1 to 4), TB (13 to 16), DET (15 to 18), SEA (17 to 21), DAL (24 to 27), CLE (23 to 28)

You've probably heard this one a thousand times before: "There's a Chinese curse that says 'May you live in interesting times.'"  Well, the Eagles and Patriots are certainly living in interesting times.  The Steelers came very close to living in those times themselves.

This week was "give a scare to the 12-1 teams" week.  The Patriots drop the farthest in the DVOA ratings, because they actually lost to a bad team instead of just winning close against a bad team.  But all three teams fall in the ratings, while the Indianapolis Colts move up with their win over Baltimore.  Philadelphia is now the number one team, but it is difficult not to look at Philadelphia's numbers without seeing an asterisk attached: *before T.O. injury.

In fact, after all that Manning debate, with gnashing of teeth and rending of garments, the Indianapolis Colts are the probably the best team in football right now.  They have moved into the top slot in WEIGHTED DVOA, which measures recent performance stronger than early performance.  More importantly, the defense is starting to prove itself.  As I wrote in my preview of Sunday's game:

We know that the Colts can score against most teams, and we know that the Ravens can stop almost anyone. But if the Colts' defense can significantly outplay the Ravens' offense, it will be a very positive sign for their playoff hopes.

Well, the defense did outplay the Baltimore offense, and the Indianapolis defense is now rates as slightly above average.  An even better sign for Indianapolis fans is that the Colts defense now ranks #9 in weighted defensive DVOA, which is better than Philadelphia (#10) or New England (#11).  It isn't all good news, though; they still struggle against the run, and the top four other AFC teams are all strong running the ball this season.

At the risk of rehashing my comments in last night's open discussion thread (too late), the biggest problem for the Colts is that they are currently the fourth seed and it will be nearly impossible for them to get into the second seed so that they can have that important first round bye.  The same goes for the team they play this week, the team that currently has the third seed: San Diego.  Because they hold tiebreakers over both teams, New England clinches the bye with a win at the Jets this week.  The Colts would still need to win out and have the Patriots lose to both the Jets and the 49ers to get the bye because of a head-to-head loss.  (Because one of their losses was to the NFC, the Chargers could get the bye if they win out and the Patriots lose to the Jets, even if the Patriots beat the 49ers.)

Getting home field advantage throughout the playoffs, it turns out, hasn't been very important over the past few years.  Home teams are 13-11 in the last 12 years of conference championship games.  But getting the first round bye is critical.  Since the playoffs went to six teams per conference in 1990, home teams are 45-11 in the second round.

The other thing that should be noted about Indy's number one ranking in weighted DVOA is how much of it is dependent on a four-game stretch where they just crushed lesser teams, the same four-game stretch that drove Manning towards the touchdown record.  Week 10 vs. Houston, Week 11 at Chicago, Week 12 at Detroit, and Week 13 vs. Tennessee.  Time for another one of those fun DVOA graphs, this time featuring the Indianapolis Colts.  The thick line is the Colts' total DVOA each week.  The two thin lines represent offense (red) and defense (black).  In order to make it easier to understand the relative strengths of offense and defense, I've switched the sign on defense (in other words, for the purposes of this graph only, defense is better when it is positive, just like offense).

Wow, is that a peak.  Even after taking into account an adjustment because they were playing poor teams, the Colts went nuts during those four weeks.  They've settled down in the last two weeks, although they are still a very good team.  They proved a lot by beating Baltimore, but is it possible the Colts peaked too early?

I don't have anything to say about the NFC that hasn't already been said in the last 24 hours.  The injury to Terrell Owens doesn't quite throw things wide open, but when combined with the injuries on the defensive line it certainly does.  I've been dismissing Atlanta all season long as overrated, but right now they are the third-best team in the NFC according to DVOA and unlike Carolina they get that first round bye.  You have to have run defense to stop the Falcons.  That's the biggest weakness of the Panthers right now, and especially after the defensive line injuries it is also the biggest weakness of the Eagles.  Michael Vick in the Super Bowl?  Perhaps.  Michael Vick getting completely embarrassed in the Super Bowl by the Steelers or Patriots?  Probably.

The AFC contender that matches up worst with the Falcons is the AFC contender with poor run defense and special teams.  Are we ready to add "lost to Michael Vick in the Super Bowl" to the never-ending discussion about Peyton Manning?

* * * * *

Other fun facts from WEIGHTED DVOA:

  • Going into their big game next Sunday, the Patriots and Jets are virtually tied, meaning that my system interprets them as being equal teams right now.
  • If it wasn't for the fact that they have to beat Pittsburgh in Week 17, Buffalo would be the clear favorite for the final AFC playoff spot.  They are now #6 in weighted DVOA, which puts them ahead of the other 8-6 AFC contenders as well as San Diego.
  • Miami is now ranked #21 in weighted DVOA, higher than a number of teams in the playoff hunt including Seattle and New Orleans.
  • St. Louis might make the playoffs even though they are last in the league in defense, last in special teams, and have now fallen to #14 in offensive DVOA and #21 in weighted offensive DVOA.
  • Tampa Bay is #24 in offensive DVOA but #16 in weighted offensive DVOA.

* * * * *

A few weeks ago, I got an e-mail accusing me of anti-Brady bias because he was missing from my ESPN.com quarterback ratings after his game against Kansas City.  What is funny about this is that

a) I'm a Patriots fan, and
b) the game was Monday night, and the ratings are published Monday morning.

Now I get a couple of these e-mails each week, accusing me of bias against one team or another because I didn't include the Monday quarterbacks in the list that was published before they actually played.  Well, for fun, here are the quarterback ratings from last night.  A.J. Feeley had 4.7 DPAR, which would have ranked 16th between Jamie Martin and Drew Bledsoe.  And Tom Brady had -7.5 DPAR which would have ranked just above Luke McCown and Chad Hutchinson.  As this comment pointed out, the conventional NFL passer rating has the two quarterbacks a lot closer to each other.  I think DPAR does a better job than conventional passer rating of telling us who had value in last night's game.

* * * * *

OFFENSE and DEFENSE DVOA are adjusted based on strength of opponent as well as to consider all fumbles, kept or lost, as equal value.  SPECIAL DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season.  NON-ADJ TOTAL VOA does not include these adjustments.  DVOA is a better indicator of team quality.  VOA is a better indicator of wins so far.

As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.


TEAM TOTAL
DVOA 		LAST
WEEK 		NON-ADJ
TOTAL VOA 		W-L OFFENSE
DVOA 		OFF.
RANK 		DEFENSE
DVOA 		DEF.
RANK 		SPECIAL
DVOA 		S.T.
RANK
1 PHI 41.1% 2 38.9% 13-1 27.3% 4 -4.5% 11 9.3% 2
2 IND 39.7% 4 44.0% 11-3 42.3% 1 -0.7% 15 -3.4% 25
3 PIT 36.3% 3 39.3% 13-1 16.4% 7 -17.5% 4 2.3% 9
4 NE 34.6% 1 28.9% 12-2 25.9% 5 -8.5% 8 0.1% 17
5 NYJ 27.8% 6 34.9% 10-4 27.4% 3 1.9% 18 2.3% 10
6 BAL 22.7% 5 15.0% 8-6 -4.4% 21 -21.0% 1 6.1% 4
7 BUF 20.0% 7 22.3% 8-6 -10.7% 26 -20.1% 2 10.6% 1
8 KC 16.7% 9 8.3% 6-8 33.2% 2 16.4% 26 -0.1% 18
9 SD 16.0% 8 24.8% 11-3 15.5% 8 -3.1% 13 -2.6% 23
10 DEN 8.6% 10 7.8% 8-6 1.8% 13 -9.5% 7 -2.8% 24
11 CAR 5.0% 11 10.5% 6-8 -1.7% 17 -11.2% 5 -4.6% 30
12 CIN 3.6% 12 -5.5% 6-8 9.3% 10 7.4% 23 1.7% 14
13 JAC 2.4% 14 -3.0% 8-6 1.0% 15 -3.5% 12 -2.1% 22
14 ATL 1.9% 16 6.8% 11-3 -3.6% 18 0.4% 17 5.9% 5
15 HOU 0.4% 20 -8.8% 6-8 0.6% 16 -1.5% 14 -1.7% 20
16 TB 0.1% 13 5.5% 5-9 -6.5% 24 -10.6% 6 -4.0% 27
TEAM TOTAL
DVOA 		LAST
WEEK 		NON-ADJ
TOTAL VOA 		W-L OFFENSE
DVOA 		OFF.
RANK 		DEFENSE
DVOA 		DEF.
RANK 		SPECIAL
DVOA 		S.T.
RANK
17 WAS -0.9% 19 0.6% 5-9 -14.4% 28 -17.6% 3 -4.0% 28
18 DET -0.9% 15 -2.1% 5-9 -3.6% 19 0.0% 16 2.7% 8
19 MIN -1.6% 21 2.4% 8-6 24.4% 6 22.0% 30 -4.0% 29
20 GB -5.2% 18 -5.7% 8-6 11.5% 9 18.8% 28 2.0% 11
21 SEA -6.2% 17 5.1% 7-7 3.7% 11 6.2% 21 -3.6% 26
22 NYG -11.0% 22 -8.0% 5-9 -5.6% 22 7.1% 22 1.7% 15
23 TEN -17.4% 25 -15.7% 4-10 -10.0% 25 2.5% 19 -4.9% 31
24 OAK -17.4% 26 -27.4% 5-9 3.3% 12 18.7% 27 -2.1% 21
25 NO -17.5% 27 -9.2% 6-8 -5.9% 23 19.6% 29 8.1% 3
26 MIA -17.5% 28 -23.8% 3-11 -30.8% 31 -8.5% 9 4.8% 6
27 DAL -17.6% 24 -19.6% 5-9 -4.1% 20 13.7% 25 0.3% 16
28 CLE -18.1% 23 -32.7% 3-11 -11.9% 27 8.1% 24 1.9% 12
29 ARI -23.0% 30 -13.2% 5-9 -18.1% 29 4.1% 20 -0.8% 19
30 CHI -29.9% 29 -30.5% 5-9 -40.9% 32 -7.2% 10 3.8% 7
31 STL -33.7% 31 -29.6% 6-8 1.6% 14 26.6% 32 -8.7% 32
32 SF -41.9% 32 -39.8% 2-12 -19.3% 30 24.4% 31 1.8% 13

  • 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.  It is based on the number of games a team has played so far, so teams which have not had their bye will appear higher.
  • 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.
  • PAST SCHEDULE lists average DVOA of opponents played so far, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative).  You'll notice that FUTURE SCHEDULE has been removed from the table; with only two games left, it is pretty easy to look up the ratings of each opponent.
  • VARIANCE 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 VARIANCE RANK
1 PHI 41.1% 13-1 12.2 1 39.5% 3 -3.6% 23 17.7% 16
2 IND 39.7% 11-3 11.2 3 47.6% 1 -0.7% 18 11.8% 26
3 PIT 36.3% 13-1 10.5 4 44.0% 2 2.5% 13 19.9% 12
4 NE 34.6% 12-2 11.6 2 29.0% 4 5.0% 8 30.3% 2
5 NYJ 27.8% 10-4 10.1 5 27.0% 5 2.1% 14 16.0% 19
6 BAL 22.7% 8-6 9.2 7 23.1% 7 11.3% 2 24.3% 7
7 BUF 20.0% 8-6 9.4 6 26.4% 6 1.4% 15 28.8% 3
8 KC 16.7% 6-8 8.8 9 16.2% 9 6.3% 7 16.0% 20
9 SD 16.0% 11-3 8.9 8 20.1% 8 -1.2% 21 12.3% 25
10 DEN 8.6% 8-6 8.0 10 -0.5% 18 0.6% 17 16.6% 18
11 CAR 5.0% 6-8 7.9 11 9.0% 12 -4.2% 25 15.2% 22
12 CIN 3.6% 6-8 7.2 17 10.9% 10 8.5% 4 12.6% 24
13 JAC 2.4% 8-6 7.7 14 0.7% 16 7.5% 5 5.1% 32
14 ATL 1.9% 11-3 7.8 12 2.8% 14 -6.7% 27 24.9% 5
15 HOU 0.4% 6-8 7.3 16 2.1% 15 4.3% 9 11.0% 29
16 TB 0.1% 5-9 7.3 15 10.1% 11 -8.2% 28 14.3% 23
TEAM TOTAL
DVOA 		W-L ESTIM.
WINS 		RANK WEIGHTED
DVOA 		RANK PAST
SCHED 		RANK VARIANCE RANK
17 WAS -0.9% 5-9 5.8 22 7.6% 13 0.7% 16 20.4% 11
18 DET -0.9% 5-9 7.2 18 -1.5% 19 -0.7% 19 21.5% 10
19 MIN -1.6% 8-6 7.7 13 0.3% 17 -3.8% 24 15.9% 21
20 GB -5.2% 8-6 6.6 19 -4.3% 20 -1.8% 22 24.6% 6
21 SEA -6.2% 7-7 6.5 20 -16.7% 22 -10.1% 31 11.7% 27
22 NYG -11.0% 5-9 5.8 21 -23.4% 28 3.2% 12 31.6% 1
23 TEN -17.4% 4-10 4.9 28 -19.9% 25 3.5% 11 19.6% 13
24 OAK -17.4% 5-9 5.3 25 -21.0% 26 9.6% 3 18.1% 15
25 NO -17.5% 6-8 5.4 24 -17.3% 23 -6.6% 26 5.2% 31
26 MIA -17.5% 3-11 5.2 26 -12.3% 21 6.5% 6 19.6% 14
27 DAL -17.6% 5-9 5.7 23 -22.8% 27 3.8% 10 11.1% 28
28 CLE -18.1% 3-11 5.0 27 -18.8% 24 16.8% 1 25.7% 4
29 ARI -23.0% 5-9 4.2 29 -23.6% 29 -8.2% 29 22.1% 9
30 CHI -29.9% 5-9 3.8 31 -37.9% 31 -1.0% 20 17.3% 17
31 STL -33.7% 6-8 3.9 30 -37.2% 30 -8.6% 30 23.6% 8
32 SF -41.9% 2-12 2.5 32 -42.8% 32 -11.2% 32 10.5% 30

Posted by: Aaron Schatz on 21 Dec 2004

comments