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.

23 Dec 2003

Week 16 Team Efficiency Ratings

by Aaron Schatz

Here are the team efficiency ratings after Week 16, 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.)

Let me get this out of the way first.  If you are visiting for the first time and want to know why Tampa rates so high and New England so low, I've already gone over the reasons for this numerous times.  Rather than just restate them I'll point you to this and this and this to explain why Tampa Bay has been a better team this year than the won-loss record shows and I'll point you to this and this to explain that New England doesn't rank on top of our ratings because the Pats are a historical anomaly with an unmatched run of consistently good but not great play.

Speaking of that unmatched run -- for those who didn't catch this statistic during the game, the Patriots were the first team in NFL history to win 10 games in a row, all by 14 points or less.  Then they went out and won their 11th straight game by 14 points or less.  As much as we Pats fans want to think we're going to spank the Bills this week in revenge for Week 1, the Pats will probably go out and win their 12th straight game by 14 points or less.

That revenge ties into the most important reason why the VOA ratings differ from public perception -- the fact that they include all 15 games of the season with equal value.  There are now significant differences between the total season VOA ratings and weighted VOA ratings that gradually give less and less impact to early-season games.  Kansas City remains the top rated team in total season VOA, but they fall out of the top spot in weighted VOA ratings this week due to their recent defensive struggles.  Instead, St. Louis is now the top team in weighted VOA.  Tennessee and New England, who still stand at #7 and #8 in total VOA due to early season struggles, rank #2 and #3 in weighted VOA.

One interesting statistic to look at is the "FOREST INDEX" we introduced a couple weeks ago, which I'm now calling "ESTIMATED WINS" based on a reader suggestion.  This is the statistic that combines general performance with performance in key situations as well as team consistency.  Four teams are now in a virtual tie with 10.4 estimated wins -- Kansas City, New England, Seattle, and Tampa Bay.  But of those four teams, only one has trended upward as the season has moved forward: the Patriots.

One interesting team that ranks far higher in ESTIMATED WINS than it does in either TOTAL VOA or even WEIGHTED VOA is the Philadelphia Eagles.  9.4 estimated wins put them seventh, but they rank at #14 in our total VOA ratings and #11 in the weighted ratings.  They have more estimated wins thanks to good red zone performance as well as consistent play; they aren't particularly good in the other split used to compute estimated wins, second half defense in close games.

49ers has somewhat calmed talk of an Eagles trip to the Super Bowl, but I'm still somewhat mystified by the number of people who feel that they will triumph over the Rams.  The Eagles bandwagon is a valuable lesson in the fact that perception lags behind actual facts by a few months or even years.

Over in the AFC, the conventional wisdom says that the Kansas City Chiefs are toast because of their horrible run defense.  Nobody is quite as pessimistic about the Colts, but everyone agrees that they face the same problem.  The general perception is that the AFC representative in the Super Bowl will be decided by defense.  It will probably be the Patriots, but if anyone is going to beat New England it will be a team that can play defense like Baltimore, Tennessee, or Denver.

But if the inability to stop the run is going to keep the Chiefs and Colts out of the Super Bowl, why are so many people happily aboard the Eagles bandwagon?  Check out DVOA ratings for defense; Kansas City is #32 against the run, Indy is #30 against the run, and right above those two teams are the Philadelphia Eagles ranked #27 with +9.7% DVOA against the run.  This means that the Eagles turn an average running back into Ahman Green.  (If you prefer your stats of the official NFL variety, the Eagles still rank #23 in rushing yards per game allowed.)

Conventional wisdom continues to hold that the Eagles are a defense-oriented team whose only offensive weapon is Donovan McNabb, while the St. Louis Rams are an offensive force with an average defense.  Maybe this was the case in 2001, but it turns out that just the opposite is true.  St. Louis has been a dominant defensive team all year, with the amazing Torry Holt and slightly less-transcendent Isaac Bruce saving the offense from the pick-happy Marc Bulger.  Over in Philadelphia, the pass defense has been overrated and the run defense has been even more overrated, but the offense has been great.  The offensive strength is hidden because so much of it is wrapped up in a three-headed running back committee, which not only ranks very high in our ratings for rushing but also makes up virtually all the success of the Philadelphia passing game.  Those long bombs the Eagles threw against Miami on Monday Night Football may be in everyone's short-term memory but they were very, very out of the norm for the way the Eagles offense has run most of the year, which has been screen after screen after screen -- nearly all of them successful.

The Eagles have a shot at the Super Bowl, certainly, but the point here is that if defense does indeed win out in the playoffs -- and since the road to Houston goes through the Edward Jones Davis Smith Goldfarb Josefowicz TWA Whatever the Heck it is Called Dome -- you should get ready for a rematch of the 2002 Super Bowl.  This time John Madden better not suggest that the Patriots sit on the ball.

Speaking of the Pats, again, there's been plenty of discussion of Joe Namath doing his best impression of Saturday Night Live's Drunk Girl during the Jets-Pats broadcast, but I haven't yet seen any discussion of the miserable clock management shown by the Jets in the final two minutes.  With 1:25 left, Chad Pennington completed a pass to Jerald Sowell that took the Jets to their own 45-yard line.  Did the Jets then run up to the line and spike the ball to save time?  No, they slowly ambled their way towards the new line of scrimmage.  In the time it took them to get there, Tom Ridge changed the national terror threat level from yellow to orange to mauve to burnt umber to maize to plaid and back to orange again.  The next play didn't begin until 56 seconds remained in the game.  It was an incomplete pass.  Well, hey, if the Jets wanted 2nd-and-10, they could have just had it 25 seconds earlier.  Instead of having 2nd-and-10 with 1:15 left, the Jets had 2nd-and-10 with :50 left.  Then, to cap it all off, they took a timeout even though the clock had already been stopped by the incomplete.  The third down interception by Eugene Wilson rendered this all moot, but if the Jets had run out of time with the ball they would have only had themselves to blame.

Another slow play this week was Terrell Buckley's third quarter interception for touchdown against Buffalo.  Am I the only one who thought that CBS was showing this play in slow motion?  I can't believe nobody from Buffalo even tried to catch up to Buckley; it was as if they all just threw up their hands and said, "Oops, we lost, I'm going to take a nap."

Finally, a discussion of Denver.  Many people are taking the Broncos as the sleeper team to make a run at the Super Bowl in the AFC.  One of the reasons given is that their season record and statistics are worse than they should be, due to the struggles when Jake Plummer was injured.  Our VOA numbers bear this out.  In fact, what's interesting is that every aspect of the Broncos all-around game declined during Weeks 6-9, when Plummer was out of the lineup, except for special teams.  Originally, my plan was to run a table showing the Broncos before, during, and after the Plummer injury.  But this table would be somewhat skewed by the Week 11 demolition of the San Diego Chargers.  Not only does this game stand out in the VOA ratings as the best defensive performance by any team this year, the defense was so good that this game stands out as the best all-around performance of any team this year despite the fact that the offense was below average!

Therefore, I present this table, which makes the same point but without Week 11 throwing the post-injury defensive numbers all out of whack:

DENVER DVOA OFFENSE
(PASS)
OFFENSE
(RUN)
DEFENSE SPECIAL
TEAMS
TOTAL
Weeks 1-5 25.2% 10.4% -13.3% 1.2% 25.9%
Weeks 6-9 (Plummer out) -30.3% -2.3% 8.1% 17.4% -5.3%
Week 11 vs. SDG -51.1% 1.2% -138.7% 6.8% 127.9%
Weeks 12-16 48.7% 23.8% -3.6% -10.6% 26.9%

  • 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 16:


TEAM
TOTAL
DVOA
LAST
WEEK
NON-ADJ
TOTAL VOA
W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
SPECIAL
VOA
S.T.
RANK
1 KAN 26.6% 1 30.6% 12-3 27.7% 1 5.8% 22 4.7% 1
2 STL 24.7% 4 27.4% 12-3 3.0% 12 -24.0% 2 -2.4% 29
3 IND 22.5% 2 18.5% 11-4 25.0% 2 1.3% 16 -1.2% 23
4 TAM 20.9% 3 23.2% 7-8 1.1% 13 -21.1% 3 -1.3% 24
5 DEN 20.3% 7 22.8% 10-5 10.1% 8 -9.3% 8 0.9% 11
6 SEA 20.1% 5 22.8% 9-6 17.8% 4 -1.2% 15 1.1% 8
7 TEN 17.5% 6 21.3% 11-4 16.9% 5 -1.3% 13 -0.7% 20
8 NWE 16.8% 8 18.2% 13-2 -1.5% 15 -16.7% 6 1.5% 6
9 MIN 16.8% 11 20.4% 9-6 20.6% 3 2.0% 18 -1.7% 26
10 SFO 16.5% 9 18.0% 7-8 14.4% 6 -4.1% 12 -2.0% 27
11 GNB 13.1% 13 16.1% 8-6 5.9% 11 -6.2% 11 0.9% 10
12 DAL 10.9% 10 18.8% 10-5 -7.1% 21 -17.6% 5 0.4% 14
13 BAL 9.4% 14 9.9% 9-6 -22.4% 31 -27.9% 1 3.9% 2
14 PHI 6.5% 12 7.1% 11-4 13.3% 7 8.4% 24 1.6% 5
15 PIT 5.2% 15 7.6% 6-9 -3.8% 18 -6.3% 10 2.7% 3
16 JAC 2.7% 17 3.0% 5-10 -2.4% 17 -7.3% 9 -2.2% 28
TEAM TOTAL
DVOA
LAST
WEEK
NON-ADJ
TOTAL VOA
W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
SPECIAL
VOA
S.T.
RANK
17 MIA 1.2% 18 1.6% 9-6 -16.9% 29 -17.8% 4 0.3% 15
18 BUF -1.8% 16 -5.6% 6-9 -14.0% 25 -11.8% 7 0.3% 16
19 CIN -8.1% 20 -5.4% 8-7 8.6% 9 13.3% 28 -3.4% 32
20 CAR -9.2% 22 -7.1% 10-5 -6.9% 20 2.0% 17 -0.2% 19
21 NYJ -10.4% 21 -11.9% 6-9 8.0% 10 17.1% 30 -1.3% 25
22 OAK -11.1% 19 -15.4% 4-10 -7.5% 23 4.9% 21 1.3% 7
23 NOR -11.2% 24 -9.5% 7-8 -0.7% 14 11.2% 27 0.7% 12
24 CLE -12.9% 23 -22.0% 4-11 -14.3% 26 -1.3% 14 0.1% 18
25 SDG -16.0% 25 -22.2% 3-12 -2.0% 16 14.0% 29 0.1% 17
26 WAS -16.5% 26 -14.5% 5-10 -4.6% 19 9.6% 26 -2.4% 30
27 NYG -17.4% 27 -19.9% 4-11 -7.2% 22 9.1% 25 -1.1% 22
28 CHI -17.9% 28 -18.7% 7-8 -15.7% 28 4.0% 19 1.8% 4
29 ATL -19.7% 29 -23.6% 4-11 -14.4% 27 5.8% 23 0.5% 13
30 DET -27.6% 30 -31.0% 4-11 -22.6% 32 4.2% 20 -0.7% 21
31 HOU -29.6% 31 -35.2% 5-10 -12.7% 24 17.9% 31 1.0% 9
32 ARI -42.7% 32 -47.3% 3-12 -19.7% 30 20.4% 32 -2.6% 31

  • ESTIMATED WINS uses a statistic known as "Forest Index" that takes into account offense, defense, and special teams, as well as consistency, red zone performance, and performance in the second half when the score is close, and then projects a number of wins adjusted to a league-average schedule.
  • WEIGHTED DVOA combines the team's DVOA performance from each game.  The past four weeks are each weighted at 100%, while each week before that is weighted progressively lower, beginning with Week 1 at 22%.
  • PAST SCHEDULE lists average DVOA of past opponents, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative).  This number will differ from the difference between DVOA and (non-adjusted) VOA because schedule strength is based on the opponent's total efficiency rating, while opponent adjustments to VOA take into account situations faced within each specific game.
  • 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
SCHEDULE
RANK VARIANCE RANK
1 KAN 26.6% 12-3 10.4 3 22.7% 4 -3.3% 28 22.9% 17
2 STL 24.7% 12-3 9.5 6 26.8% 1 -2.9% 26 26.4% 11
3 IND 22.5% 11-4 9.8 5 18.5% 6 0.3% 16 20.2% 22
4 TAM 20.9% 7-8 10.4 4 14.4% 9 -4.8% 30 26.3% 12
5 DEN 20.3% 10-5 9.2 8 22.3% 5 0.2% 17 24.1% 13
6 SEA 20.1% 9-6 10.4 1 11.7% 12 -4.6% 29 22.4% 19
7 TEN 17.5% 11-4 9.1 9 23.4% 2 -3.3% 27 19.2% 24
8 NWE 16.8% 13-2 10.4 2 23.1% 3 -1.1% 22 16.4% 27
9 MIN 16.8% 9-6 7.5 14 10.4% 13 -1.4% 24 34.2% 4
10 SFO 16.5% 7-8 6.8 17 13.3% 10 -0.7% 18 36.6% 3
11 GNB 13.1% 8-6 8.7 11 16.7% 7 -0.8% 20 23.6% 16
12 DAL 10.9% 10-5 7.1 15 9.8% 14 -8.5% 32 33.9% 5
13 BAL 9.4% 9-6 9.0 10 14.6% 8 0.4% 15 19.9% 23
14 PHI 6.5% 11-4 9.4 7 12.5% 11 -0.9% 21 15.4% 28
15 PIT 5.2% 6-9 8.5 12 2.4% 17 0.9% 13 22.8% 18
16 JAC 2.7% 5-10 6.2 18 9.2% 15 1.4% 11 21.5% 20
TEAM TOTAL
DVOA
W-L ESTIM.
WINS
RANK WEIGHTED
DVOA
RANK PAST
SCHEDULE
RANK VARIANCE RANK
17 MIA 1.2% 9-6 8.2 13 3.2% 16 0.6% 14 27.8% 10
18 BUF -1.8% 6-9 6.1 20 -6.7% 19 0.9% 12 38.1% 2
19 CIN -8.1% 8-7 5.1 25 -5.1% 18 1.6% 9 14.3% 30
20 CAR -9.2% 10-5 5.6 22 -7.0% 20 -5.1% 31 11.8% 32
21 NYJ -10.4% 6-9 6.1 21 -11.3% 22 1.5% 10 12.7% 31
22 OAK -11.1% 4-10 5.5 23 -16.3% 24 4.2% 5 24.0% 14
23 NOR -11.2% 7-8 7.1 16 -8.6% 21 -1.9% 25 20.3% 21
24 CLE -12.9% 4-11 5.3 24 -20.4% 27 6.6% 1 32.9% 6
25 SDG -16.0% 3-12 4.8 26 -16.3% 25 4.3% 4 18.3% 25
26 WAS -16.5% 5-10 3.3 31 -20.6% 28 -1.2% 23 28.4% 9
27 NYG -17.4% 4-11 4.3 30 -23.3% 29 2.6% 7 30.5% 7
28 CHI -17.9% 7-8 6.2 19 -11.4% 23 -0.8% 19 17.9% 26
29 ATL -19.7% 4-11 4.7 27 -18.0% 26 2.4% 8 28.4% 8
30 DET -27.6% 4-11 4.6 28 -26.8% 30 2.6% 6 14.9% 29
31 HOU -29.6% 5-10 4.3 29 -37.1% 31 4.5% 3 23.7% 15
32 ARI -42.7% 3-12 2.8 32 -42.0% 32 5.7% 2 44.7% 1

PREVIOUS WEEKS:

Posted by: Aaron Schatz on 23 Dec 2003

comments