Writers of Pro Football Prospectus 2008

Most Recent FO Features

KhanSha1.jpg

» Futures: My Expansion Franchise

You've just been awarded an NFL expansion team and must build your personnel department. How would you do it? Matt Waldman takes on the exercise.

09 Nov 2004

Week 9 DVOA Ratings

by Aaron Schatz

(Note: Now fixes earlier error involving ratings for Indianapolis and Minnesota.)

Here are the Football Outsiders team efficiency ratings through Week 9 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 UpCIN (25 to 16), DEN (13 to 9)

Moving DownDAL (16 to 24), NYG (9 to 13)

The first thing you are probably wondering is why Pittsburgh is not #1 after soundly beating both Philadelphia and New England over the past two weeks.  It's important to realize that DVOA doesn't measure which teams beat which other teams, but rather how teams perform on every single play of the season.  While Pittsburgh has been the best team in football over the past two weeks, they haven't been the best team in football over the past nine weeks.  In particular, the offense didn't really come together until Week 5 against Cleveland.  Pittsburgh's offensive DVOA for their first four games was -1.3%.  For the past four games, it is 30.0%, third best in football over the past five weeks behind Kansas City and Indianapolis.

Pittsburgh has steadily climbed up the ratings over the past few weeks.  Here are the total season DVOA ratings for all three top teams as they've developed over the past few weeks:


Total DVOA after Each of the Past Five Weeks

Team Week 5 Week 6 Week 7 Week 8 Week 9
PHI 51.3% 48.3% 44.7% 44.1% 38.0%
NE 39.0% 43.0% 44.7% 31.9% 36.7%
PIT 19.6% 18.8% 19.7% 27.3% 31.7%

You may notice that the Steelers have a higher unadjusted VOA than either the Eagles or Patriots, but adjustments move them down into third place.  The PAST SCHEDULE numbers listed below say that the Steelers and Patriots have had schedules of similar difficulties, but due to differences in game situations and run/pass ratio, the Patriots get a strong adjustment upwards based on their opponents.  (For example, if Houston passes on you, your defensive rating gets adjusted upwards because they have a good passing game; if Houston runs on you, your defensive rating gets adjusted downwards to reflect that the Texans can't run the ball.)  The Patriots also have a better DVOA rating because they've only recovered the ball twice out of their nine fumbles on offense.  An extra couple of turnovers are a big help to your opponents but, as I often point out, the percentage of fumbles you recover is just luck and random ball bounces.

The Eagles lost but this changes the playoff picture in the NFC exactly zippo.  To paraphrase Louis XIV (or perhaps Yasser Arafat), "Apres les Aigles, le deluge."  Out of the top nine teams in DVOA, eight are AFC teams.  Five AFC teams have six wins, but only two NFC teams have six wins, and one of them, Atlanta, is way down at #17 in total DVOA.  I don't think anyone has any idea who is going to play the Eagles for a trip to the Super Bowl, or whether the NFC will even find six teams with winning records.  Over in the AFC, despite all the strength, I have a real hard time picturing anyone other than the Steelers and Patriots in the AFC Championship, because the other teams are either totally inconsistent (Denver), unbalanced in favor of defense (Baltimore), or unbalanced in favor of offense (everyone else).

* * * * *

Six thoughts from Monday Night Football:

I watched the game with Ian (ex-Scramble writer) on his 50-inch HDTV (slobber) and he pointed out that Edgerrin James may have set an NFL record for most times making the first guy miss in one game.  This statistic isn't available, but it would really useful to track how many yards a running back ran before first contact.  Some guys just find the hole and get through it, letting the offensive line do the work, while other guys are able to bounce off tacklers and avoid the defense.  Last night, James was definitely the second guy.

According to this article by Len Pasquarelli, many people now consider Minnesota's Bryant McKinnie the best left tackle in the NFC, and last night's game was an aberration.  I don't get to see many Vikings games but I'll be watching to see how McKinnie plays against the Packers next week because last night he was horrendous.  An absolute turnstile with "Property of Dwight Freeney" tattooed on his ass.

The Colts ran an interesting play at 11:20 of the fourth quarter.  On 1st-and-10 from their own 35-yard line, they ran the usual Indy play fake, except that James and Manning did almost nothing to sell the fake.  It didn't seem right, because Manning is the best play fake quarterback in the NFL -- but it made sense a few seconds later when Manning hit James in the flat and he ran through a good amount of free space for a 12-yard gain.  I think this was a play fake fake.  They ran what looked like a play fake, but so badly that Minnesota would think, "Oh, not another one of these play fakes, we're not biting on it this time."  Since Minnesota didn't come up to stop the run, however, that meant nobody was covering James, freeing him for the reception.

Did anybody catch the fact that on the Nate Burleson punt return touchdown, Vikings wide receiver Kelly Campbell couldn't catch Colts punter Hunter Smith from behind as Smith was racing to tackle Burleson?  What kind of a wide receiver can't catch up to a punter in the open field?

Peyton Manning is a better left-handed quarterback than Jake Plummer.

Jermaine Wiggins could use a salad.

* * * * *

I'll have many more thoughts on current DVOA ratings and the prospects for the remainder of the season in two big midseason reports coming over the next two weeks.  This Friday I'll review all 16 NFC teams, next Friday all 16 AFC teams.  I have even built a midseason projection system similar to the preseason system that picked out the Jets and Chargers as teams to watch, so I'll have projected DVOAs for the remainder of the season as well as projected won-loss records.  If you would want an early glimpse of the midseason previews, they are running one division per day from Tuesday through Friday in the New York Sun both this week and next.

Also, look for a Loser League Midseason Review from Patrick coming later today or tomorrow along with the final results of Loser League Part I and signups for Loser League Part II.

* * * * *

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.  Opponent adjustments are currently at 90 percent strength.  Beginning next week, opponent adjustments will be at full strength, and we'll introduce WEIGHTED DVOA which gives more importance to recent games.

As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.  Player stats pages will be updated later today.


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 38.0% 1 30.7% 7-1 31.2% 3 4.8% 20 11.5% 2
2 NE 36.7% 2 23.5% 7-1 27.0% 6 -7.6% 6 2.2% 10
3 PIT 31.7% 3 39.9% 7-1 15.1% 9 -14.8% 2 1.7% 11
4 BAL 23.0% 6 21.4% 5-3 -5.9% 23 -20.9% 1 8.0% 3
5 NYJ 19.2% 5 34.5% 6-2 29.4% 5 10.8% 23 0.6% 16
6 SD 18.0% 7 32.2% 6-3 20.4% 8 4.0% 18 1.6% 13
7 KC 17.8% 4 11.8% 3-5 32.3% 2 7.1% 22 -7.5% 32
8 IND 15.5% 8 13.7% 5-3 39.2% 1 18.1% 25 -5.6% 30
9 DEN 10.1% 13 13.2% 6-3 8.7% 11 -2.6% 12 -1.1% 20
10 SEA 9.9% 11 29.9% 5-3 1.9% 15 -12.2% 4 -4.2% 29
11 HOU 7.1% 12 2.6% 4-4 8.6% 12 2.8% 17 1.3% 15
12 DET 5.6% 10 4.8% 4-4 5.8% 14 -2.0% 13 -2.2% 23
13 NYG 5.1% 9 20.4% 5-3 -0.8% 17 -5.9% 8 0.0% 18
14 MIN 4.7% 14 1.6% 5-3 30.2% 4 22.1% 28 -3.5% 27
15 BUF 3.1% 18 1.5% 3-5 -8.4% 26 -4.7% 9 6.8% 4
16 CIN 0.7% 25 -3.1% 3-5 -2.0% 18 1.3% 16 4.0% 7
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 ATL 0.4% 15 3.1% 6-2 -5.0% 21 1.1% 15 6.5% 5
18 TB -0.6% 19 -1.9% 3-5 -5.0% 22 -7.2% 7 -2.8% 26
19 JAC -1.3% 17 -7.1% 5-3 -3.3% 19 0.4% 14 2.4% 9
20 WAS -5.7% 20 -6.2% 3-5 -17.9% 31 -14.7% 3 -2.5% 24
21 CHI -6.7% 23 -5.1% 3-5 -17.4% 30 -10.2% 5 0.4% 17
22 GB -8.4% 21 -11.6% 4-4 13.8% 10 23.6% 29 1.4% 14
23 CLE -8.8% 22 -14.1% 3-5 -0.5% 16 14.0% 24 5.7% 6
24 DAL -12.2% 16 -16.8% 3-5 6.8% 13 21.5% 27 2.5% 8
25 CAR -13.8% 24 -19.5% 1-7 -6.8% 24 4.5% 19 -2.5% 25
26 ARI -14.8% 26 -6.1% 3-5 -13.9% 29 -3.1% 10 -4.0% 28
27 NO -14.9% 27 -7.0% 3-5 -3.3% 20 23.9% 30 12.3% 1
28 STL -19.0% 28 -16.8% 4-4 20.5% 7 32.3% 32 -7.2% 31
29 TEN -20.3% 29 -13.8% 3-5 -13.4% 28 5.8% 21 -1.1% 19
30 MIA -30.2% 32 -39.4% 1-8 -34.6% 32 -2.8% 11 1.7% 12
31 OAK -30.6% 31 -33.8% 3-6 -8.9% 27 20.1% 26 -1.6% 21
32 SF -34.1% 30 -43.5% 1-7 -7.6% 25 24.6% 31 -2.0% 22

  • PAST SCHEDULE lists average DVOA of opponents played so far, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative).
  • FUTURE SCHEDULE lists average DVOA of opponents remaining on the schedule, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative).
  • 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).
  • 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.


TEAM TOTAL
VOA
W-L ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VARIANCE RANK
1 PHI 38.0% 7-1 6.7 2 5.1% 7 -7.2% 29 8.9% 26
2 NE 36.7% 7-1 7.0 1 1.9% 13 -1.2% 18 17.5% 13
3 PIT 31.7% 7-1 5.5 4 2.1% 12 4.4% 11 19.1% 12
4 BAL 23.0% 5-3 5.3 6 8.5% 2 8.3% 5 20.0% 9
5 NYJ 19.2% 6-2 5.5 5 -4.1% 26 8.2% 6 14.1% 20
6 SD 18.0% 6-3 5.9 3 -4.9% 28 3.0% 13 14.3% 19
7 KC 17.8% 3-5 4.9 9 5.1% 8 -1.7% 22 25.8% 5
8 IND 15.5% 5-3 5.1 7 -0.3% 21 5.5% 10 2.9% 31
9 DEN 10.1% 6-3 5.1 8 -0.3% 20 -6.4% 28 20.0% 8
10 SEA 9.9% 5-3 4.1 17 -11.8% 32 -6.1% 27 9.3% 25
11 HOU 7.1% 4-4 4.7 11 0.5% 19 0.6% 15 6.5% 28
12 DET 5.6% 4-4 4.6 12 2.2% 11 -3.3% 24 32.9% 1
13 NYG 5.1% 5-3 4.2 15 0.8% 16 7.6% 7 23.9% 6
14 MIN 4.7% 5-3 4.4 13 1.5% 15 -1.2% 19 15.3% 17
15 BUF 3.1% 3-5 4.7 10 2.6% 10 -1.6% 21 14.9% 18
16 CIN 0.7% 3-5 3.9 20 1.6% 14 15.4% 1 17.0% 15
TEAM TOTAL
VOA
W-L ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VARIANCE RANK
17 ATL 0.4% 6-2 4.2 16 -3.8% 24 -7.5% 30 26.4% 4
18 TB -0.6% 3-5 4.2 14 -4.9% 27 -9.1% 31 12.5% 22
19 JAC -1.3% 5-3 4.0 19 8.4% 3 -2.1% 23 6.1% 29
20 WAS -5.7% 3-5 3.1 25 -0.4% 22 9.0% 4 13.7% 21
21 CHI -6.7% 3-5 3.5 24 0.6% 18 -1.2% 17 5.1% 30
22 GB -8.4% 4-4 3.8 21 -4.1% 25 4.1% 12 28.4% 3
23 CLE -8.8% 3-5 3.5 23 12.9% 1 10.8% 3 19.8% 10
24 DAL -12.2% 3-5 4.0 18 3.1% 9 10.8% 2 15.9% 16
25 CAR -13.8% 1-7 3.6 22 6.9% 4 -12.3% 32 8.8% 27
26 ARI -14.8% 3-5 3.0 28 -6.0% 30 -3.5% 25 19.5% 11
27 NO -14.9% 3-5 3.1 26 -8.3% 31 -1.5% 20 2.8% 32
28 STL -19.0% 4-4 3.0 27 -5.9% 29 -0.1% 16 10.9% 24
29 TEN -20.3% 3-5 2.5 30 0.8% 17 2.2% 14 29.7% 2
30 MIA -30.2% 1-8 2.1 31 6.3% 5 5.7% 9 17.2% 14
31 OAK -30.6% 3-6 2.5 29 6.2% 6 6.1% 8 21.6% 7
32 SF -34.1% 1-7 2.1 32 -2.0% 23 -5.5% 26 12.0% 23

Posted by: Aaron Schatz on 09 Nov 2004

comments