Writers of Pro Football Prospectus 2008

Most Recent FO Features

JohnsonCal09.jpg

» Scramble Over/Unders: the Norths

The league's northern divisions pose a number of meaty questions, such as: "Is the Bears' offense due for a repeat performance?" "Why do the Lions have such pronounced splits?" and "Has Johnny Manziel made the Cleveland brass even crazier?"

26 Oct 2003

Week 8 Team Efficiency Ratings

by Aaron Schatz

Here are the team efficiency ratings after Week 8, 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.)

We're now at the midpoint of the season.  There's one undefeated team left, Kansas City.  Wouldn't you know it, this week Kansas City moves to the top of the VOA ratings for the first time, passing the Vikings after Minnesota lost to the New York Football Giants.

Moving up: NYG (from #19 to #10), TEN (from #16 to #9), NWE (from #17 to #11), BAL (from #23 to #18, "Thanks Danny Kanell!")
Moving down: BUF (from #11 to #20), PIT (from #12 to #21), NYJ (from #10 to #16), MIN (from #1 to #6)

A quick digression.  For those of you wondering why these ratings took a little longer than usual to publish this week, it's been quite a few days in the Schatz household.  My daughter Mirinae Fay Schatz was born Sunday October 26 at noon, just in time to watch football with daddy.  Six pounds ten ounces.  Both mom and daughter are feeling very well, thanks.  For those wondering about the name, it's pronounced MEER-in-ay, it's Korean, and it means "galaxy."  My wife and I knew this Korean girl named Mirinae and always found her name beautiful, so we gave it to our daughter.  We know she'll never have to be "Mirinae S." in school.

I hope nobody finds it too obnoxious if I use this opportunity to point out that little Paypal donation button on the right side of the screen.  As you can imagine, we all put a lot of time into this website above and beyond our normal employment, particularly the hours I spend breaking down each week's play-by-play statistics and the hours Benjy Rose has spent designing and maintaining the website.  And, wouldn't you know it, Benjy and I are the two Outsiders providing for young infants; Abigail Rose is now nine months old.

Back to the NFL: While we clearly agree with all commentators on the team at number one, I'm guessing we're in disagreement with most other rankings when it comes to number two.  The VOA numbers still say that Tampa Bay is the second-best team in the NFL, especially after their victory over fellow defense-oriented Dallas this week.  Tampa has had one real loss, to San Francisco, and two massive flukes.  Yes, losing a huge lead in the last five minutes says something particularly unappealing about your football team, but it is still a fluke.  Tampa is still a great team despite some defensive injuries, and you'll notice they have the second-easiest remaining schedule.

Tampa's win over Dallas showed which was the better team, but it didn't necessarily show which was the better defense.  Even after the loss, and despite the weekly increase in the adjustment for the easy Cowboy schedule, VOA says that Dallas has the best defense in the league.  Dallas still kept Tampa's offense to a 0% VOA this week, but they lost because Quincy Carter and the gang were humiliated to the tune of a -68% VOA by the Tampa defense.  We said that Dallas' offense was mediocre back when people were writing that they were "the league's #1 rated offense" and now we've been proven correct.

Of course, Tampa won't win the NFC South, because the easiest remaining schedule and a two-game lead belong to the still-overrated Carolina Panthers.  Once again this week, Carolina pulled out a close game in overtime.  Eventually their luck has to run out, and I'm guessing that time is the second round of the playoffs (since they will likely get a bye).  

You may remember that before the season we introduced the Pythagorean Theorem to football.  Introduced by Bill James, it says that a baseball team's record can be estimated using runs scored and allowed.  It turns out this works for the NFL as well, although with a larger margin for error due to the short season.  Anyway, tying this back into Carolina, the 2003 Panthers have scored 141 points and allowed 125; according to the Pythagorean Theorem, this translates to a 9-7 record over a full season, even though they are on pace for 12-13 wins.  Right now Carolina has a winning percentage of .857 even though their performance translates to a winning percentage of .571.  That pace is almost impossible to continue.  Over the past twenty years, only one club has overachieved its Pythagorean projection by more than three wins -- the 1992 Colts, who went 9-7 despite scoring like a 5-11 team -- and right now the Panthers are overachieving even more than that.

(You might also note this comment from that same preseason article: "Kansas City fans have the biggest reason to celebrate.  Not only were the Chiefs about 1.5 wins worse than their projection, they get to face the NFC North and AFC North this year.  Lions and Bengals and Bears, oh my!  As in: Oh my, those teams suck."  On the other hand, you also might remember that Dick Vermeil has a history of underperforming teams.)

I've been thinking lately about one of Bill Simmons' old traditions, figuring out the "Second Half Team."  Every year, one team plays unexpectedly well in the second half, occasionally even making a playoff run.  More often, this team simply leads to making some money because the Vegas line takes the rest of the season to catch up to their new level of play.  Think 2002 Giants, 2001 Patriots, 2000 Jaguars.

In my opinion, the best possibility for the 2003 Second Half Team is Cincinnati.  Now, I understand that this goes against BSG's other rule, "Never Bet on the Bengals."  But there are some good positive indicators for the Bengals.  Not only have they won three of the last four, they won those games with much higher VOAs than they were getting at the beginning of the season.  Here are the opponent-adjusted VOAs for the Bengals so far in 2003, by week:

Week 1 vs. DEN: -14% (L)
Week 2 at OAK: -16% (L)
Week 3 vs. PIT: -56% (L)
Week 4 at CLE: -2% (W)
Week 5 at BUF: -27% (L)
Week 7 vs. BAL: +37% (W)
Week 8 vs. SEA: +46% (W)

Now, that looks to me like a team that is improving, despite the loss in Week 5 to the Buffalo Schizophrenics.  Add to this that Cincinnati has a middle-of-the-pack schedule, and the next few games are set up for them to go on a roll: Arizona, Houston, a game against Kansas City that screams "back door cover," San Diego, and the struggling Pittsburgh Steelers.  It isn't really that ridiculous to imagine the Bengals 7-4 and riding into Baltimore for a December 7 game that could determine the AFC North title.  Ye gods.  One problem holding Cincinnati back: the worst special teams in the league.  My estimates say that the Bengals have lost roughly 17 points based on poor field position from kick and punt returns, and another nine points based on poor punting.  Shayne Graham is actually positive, worth approximately two points on kickoffs and four on field goals hit compared to average kickers.

The other possibilities for second half team?  Well, I would consider the Packers, but they haven't really been getting better as the season goes along.  The Jets seem to be underrated according to VOA, and they are getting Chad Pennington back, but more and more it looks like their higher-than-expected rating is mostly the result of their Week 6 steamrolling of the Buffalo Schizophrenics.  (Buffalo is really messing up these ratings.)  That leaves the San Francisco 49ers, who are much better than their record indicates and don't have a tough schedule remaining, but I'm not quite sure what to believe about a team that can beat Tampa Bay and then lose to Arizona one week later.  Well, other than perhaps they should be forced to play Buffalo in what would be the least predictable game in NFL history.

  • Opponent adjustments are at 80% strength.
  • As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.
  • All numbers are adjusted for opponent quality except for NON-ADJ TOTAL VOA.

Here are the ratings through Week 8:
 


TEAM
ADJUSTED
TOTAL VOA
LAST
WEEK
NON-ADJ
TOTAL VOA
W-L
OFFENSE
DVOA
OFF.
RANK
DEFENSE
OVOA
DEF.
RANK
SPECIAL
VOA
S.T.
RANK
1 KAN 39.3% 4 42.9% 8-0 17.7% 3 -12.7% 6 8.9% 1
2 TAM 36.3% 6 37.0% 4-3 12.8% 7 -24.0% 3 -0.4% 23
3 IND 35.8% 3 33.9% 6-1 24.9% 1 -8.7% 9 2.2% 9
4 STL 33.6% 8 37.6% 5-2 9.3% 9 -24.3% 2 0.1% 21
5 SEA 26.0% 2 30.9% 5-2 14.4% 5 -10.3% 8 1.2% 12
6 MIN 25.6% 1 37.0% 6-1 23.2% 2 -4.2% 15 -1.8% 28
7 DAL 20.7% 5 32.6% 5-2 -6.1% 19 -26.2% 1 0.5% 19
8 SFO 15.7% 7 15.9% 3-5 7.5% 11 -10.5% 7 -2.4% 30
9 TEN 9.4% 16 10.7% 6-2 16.7% 4 6.7% 22 -0.6% 24
10 NYG 8.6% 19 0.5% 3-4 2.7% 13 -7.8% 12 -1.9% 29
11 NWE 8.4% 17 13.0% 6-2 -0.9% 17 -8.1% 10 1.2% 13
12 OAK 7.5% 15 5.2% 2-5 8.2% 10 2.1% 19 1.3% 10
13 DEN 7.3% 9 8.2% 5-3 2.1% 15 -4.1% 16 1.1% 15
14 MIA 4.5% 13 11.1% 5-2 -16.1% 26 -21.8% 4 -1.2% 26
15 GNB 3.6% 18 3.2% 3-4 12.7% 8 9.8% 24 0.7% 18
16 NYJ 3.6% 10 6.2% 2-5 14.2% 6 14.7% 26 4.1% 3
TEAM ADJUSTED
TOTAL VOA
LAST
WEEK
NON-ADJ
TOTAL VOA
W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
OVOA
DEF.
RANK
SPECIAL
VOA
S.T.
RANK
17 CLE -0.3% 14 -8.1% 3-5 -7.5% 21 -8.1% 11 -0.9% 25
18 BAL -2.3% 23 4.6% 4-3 -17.5% 27 -12.7% 5 2.5% 6
19 CAR -3.1% 22 -4.5% 6-1 -9.5% 22 0.2% 17 6.6% 2
20 BUF -3.1% 11 -4.7% 4-4 -10.2% 23 -6.3% 14 0.9% 17
21 PIT -3.4% 12 -11.7% 2-5 -10.8% 24 -7.2% 13 0.2% 20
22 CIN -4.5% 20 -3.1% 3-4 2.8% 12 3.0% 20 -4.3% 32
23 SDG -12.7% 26 -17.9% 1-6 2.4% 14 15.1% 28 0.0% 22
24 HOU -12.9% 21 -21.2% 2-5 -7.0% 20 8.2% 23 2.3% 8
25 PHI -17.0% 28 -22.3% 4-3 -18.5% 28 1.6% 18 3.2% 5
26 WAS -17.4% 27 -12.9% 3-4 -0.3% 16 18.0% 30 0.9% 16
27 NOR -18.8% 24 -14.7% 3-5 -5.3% 18 14.7% 27 1.3% 11
28 JAC -18.9% 25 -17.0% 1-6 -11.6% 25 4.0% 21 -3.2% 31
29 CHI -32.0% 31 -35.0% 2-5 -24.8% 32 10.4% 25 3.2% 4
30 ATL -35.7% 30 -47.1% 1-6 -20.9% 30 17.2% 29 2.5% 7
31 ARI -41.0% 32 -48.0% 2-5 -19.0% 29 20.6% 32 -1.4% 27
32 DET -41.9% 29 -42.6% 1-6 -24.5% 31 18.6% 31 1.2% 14

  • PAST SCHEDULE lists average VOA of past opponents, while FUTURE SCHEDULE lists average VOA of upcoming opponents.  Teams are ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative).
  • The PAST SCHEDULE number will differ from the difference between ADJUSTED VOA and NON-ADJ VOA because schedule strength is based on the opponent's total efficiency rating, while opponent adjustments to VOA take into account the situations faced within each specific game.
  • WEIGHTED VOA combines the team's adjusted VOA performance from each game.  The past four weeks are each weighted at 100%, while each week before that is weighted 5% lower, beginning with Week 1 at 80%.
  • VARIANCE measures the statistical variance of the team's weekly adjusted VOA performance.  Teams are ranked from least consistent (#1, highest variance) to most consistent (#32, smallest variance).


TEAM
ADJUSTED
TOTAL VOA
W-L
PAST
SCHEDULE
RANK FUTURE
SCHEDULE
RANK WEIGHTED
VOA
RANK VARIANCE RANK
1 KAN 39.3% 8-0 -2.0% 25 -6.4% 30 38.0% 1 15.9% 22
2 TAM 36.3% 4-3 -0.1% 18 -9.6% 31 35.3% 3 37.6% 3
3 IND 35.8% 6-1 -1.2% 21 -4.2% 26 34.8% 4 21.4% 15
4 STL 33.6% 5-2 -3.7% 28 -6.1% 27 35.4% 2 22.3% 11
5 SEA 26.0% 5-2 -6.2% 30 -3.5% 24 22.3% 6 22.3% 12
6 MIN 25.6% 6-1 -10.6% 31 -2.0% 22 23.7% 5 22.0% 13
7 DAL 20.7% 5-2 -12.4% 32 -6.1% 28 21.8% 7 28.4% 6
8 SFO 15.7% 3-5 0.8% 15 -0.6% 19 13.0% 8 43.3% 2
9 TEN 9.4% 6-2 -0.7% 20 1.2% 15 10.9% 9 15.7% 24
10 NYG 8.6% 3-4 8.3% 3 -3.8% 25 8.0% 11 17.9% 19
11 NWE 8.4% 6-2 -1.4% 22 4.6% 8 10.4% 10 24.5% 10
12 OAK 7.5% 2-5 0.9% 14 2.1% 13 6.2% 13 5.1% 31
13 DEN 7.3% 5-3 0.9% 13 6.2% 6 5.7% 15 16.2% 21
14 MIA 4.5% 5-2 -3.9% 29 4.2% 9 6.1% 14 19.3% 17
15 GNB 3.6% 3-4 1.4% 12 -1.2% 21 6.4% 12 13.4% 28
16 NYJ 3.6% 2-5 -2.4% 27 5.4% 7 4.4% 16 13.4% 27
TEAM

ADJUSTED
TOTAL VOA

W-L PAST
SCHEDULE
RANK FUTURE
SCHEDULE
RANK WEIGHTED
VOA
RANK VARIANCE RANK
17 CLE -0.3% 3-5 5.5% 7 6.9% 3 -0.1% 18 19.7% 16
18 BAL -2.3% 4-3 -2.2% 26 6.7% 4 1.9% 17 25.4% 9
19 CAR -3.1% 6-1 -1.5% 23 -11.1% 32 -2.0% 19 15.8% 23
20 BUF -3.1% 4-4 -0.2% 19 9.8% 2 -4.7% 21 52.5% 1
21 PIT -3.4% 2-5 11.8% 1 -0.9% 20 -5.2% 22 29.1% 5
22 CIN -4.5% 3-4 4.5% 9 1.8% 14 -3.1% 20 12.8% 29
23 SDG -12.7% 1-6 5.3% 8 0.2% 17 -11.9% 23 3.3% 32
24 HOU -12.9% 2-5 7.8% 6 2.7% 10 -14.3% 24 26.0% 8
25 PHI -17.0% 4-3 8.2% 4 -2.4% 23 -15.6% 25 7.6% 30
26 WAS -17.4% 3-4 0.2% 16 1.1% 16 -19.2% 28 32.2% 4
27 NOR -18.8% 3-5 -1.9% 24 -6.2% 29 -18.1% 26 17.9% 18
28 JAC -18.9% 1-6 2.6% 10 2.6% 11 -19.2% 27 14.0% 26
29 CHI -32.0% 2-5 2.5% 11 -0.4% 18 -28.9% 29 16.4% 20
30 ATL -35.7% 1-6 11.0% 2 2.2% 12 -36.3% 30 26.7% 7
31 ARI -41.0% 2-5 7.9% 5 6.4% 5 -39.1% 31 21.4% 14
32 DET -41.9% 1-6 0.0% 17 9.8% 1 -41.4% 32 14.9% 25

PREVIOUS WEEKS:

Posted by: Aaron Schatz on 26 Oct 2003

comments