Writers of Pro Football Prospectus 2008

Most Recent FO Features

HarvinPer09.jpg

» Impact of the NFL's Kickoff Rule Change

After three NFL seasons of kicking off from the 35-yard line, what has been the impact on touchbacks, returns, field position, scoring and injuries? Also, is this rule responsible for a record number of big comebacks?

20 Oct 2004

Week 6 DVOA Ratings

by Aaron Schatz

Here are the Football Outsiders team efficiency ratings through Week 6 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 UpGB (32 to 24), DEN (10 to 4), WAS (27 to 21), BUF (18 to 12), 

Moving DownDET (8 to 18), TEN (22 to 28)

There was no Football Outsiders in 2002.  But if there was, we would have spent the season tracking two teams, one in each conference, that jumped out in front of the rest of the NFL very early on and maintained DVOA ratings far beyond their conference foes for the entire season.  Every week from Week 4 of 2002 until the end of the season, Tampa Bay was the top-ranked NFC team every week, while Oakland was the top-ranked AFC team every week.  These two teams, Tampa Bay and Oakland, eventually met in the Super Bowl.

This year, two teams have similarly established themselves early on as the best teams in the NFL, and while there is a lot of football left to be played between now and February, the path is set for Philadelphia and New England to play in Super Bowl XXXIX.

Of course, the Patriots have a number of tests in front of them, starting this Sunday when they take on the other undefeated team besides themselves and the Eagles, the New York Jets.  Wow, who would have expected such a fast start from the Jets!  Oh yes, that's right, us.  The AFC East is shaping up to be one of the strangest divisions in the history of the NFL.  The Pats and Jets, of course, are undefeated, while the Dolphins are challenging the metaphor machine to come up with similar examples where sustained greatness suddenly turned to rubbish.  The Clash album Cut the Crap?  The last season of Homicide after Pembleton left the force?  The 1999 Chicago Bulls?  Rocky V?  The arrival of "Cousin Pam" on The Cosby Show?  How about that season of Saturday Night Live with Chris Elliott and Jeaneane Garafalo?  Free free to nominate your own examples in the comments.

Then there is the fourth team in the division, the Buffalo Bills.  They're horrible, right?  The general consensus is that they only have a win because when two winless teams play each other, someone has to come out the victor.  But this consensus may not be accurate.  This week, with Kansas City's loss to Jacksonville, Buffalo takes over the title of the highest-ranked losing team according to DVOA.  (For those curious, DVOA says that the Chiefs deserved to lose that game even without taking into account Lawrence Tynes' sudden transformation into the Chris Dudley of kicking.)

The idea of Buffalo as an above-average team sounds ludicrous, but three of Buffalo's four losses have come by three points or less.  What's interesting is that they are doing this with balance.  Last year DVOA had the Bills ranked #7 in defense, but #29 in offense.  This season, the defense doesn't seem to be playing quite as well, ranked #13 (-3.3%), but the offense has improved from terrible to average, ranked #17 (-2.0%).  Even more curious is the way the Buffalo offense has gradually improved from game to game this season.  Here is how their offense rates in each game this season, both by DVOA (adjusted for opponent) and VOA (not adjusted):


Week Opponent Offense
DVOA
Offense
VOA
1 JAC -30.8% -25.9%
2 OAK -28.3% -17.6%
4 NE 2.1% -11.0%
5 NYJ 14.2% 10.7%
6 MIA 32.8% 25.5%

The Bills aren't going to the playoffs out of a division with two teams that started 5-0.  But if they beat Baltimore this week, we'll know they are for real.  Not great, but good -- good enough to play an interesting role as spoiler during a four-game stretch starting Week 9 where they get the Jets, Patriots, Rams, and Seahawks and could help decide two division races.

The opposite of Buffalo is the lowest-ranked winning team, and this week that changes from Jacksonville to St. Louis.  I have to admit the Rams are confusing the heck out of me.  For the last couple of years, their defense has been stalwart while their offense has gradually declined (with the exception of Torry Holt and Isaac Bruce).  Both the Rams offense and defense led the league in turnovers.  This year, things are switched -- the entire offense is clicking like the Rams of old, with very few turnovers, while the defense has played terrible, with very few turnovers.  The addition of Steven Jackson has been huge; not only is he very good (and would have looked nice in Dallas, don't you think?) but he gives Marshall Faulk rest which makes Faulk more effective.  The question is whether Bulger can keep his low interception total going.  DVOA thinks the Rams have played a very easy schedule, so I'm not so sure.  Their next three games come against three of the top four defenses in DVOA: Miami, New England, and a very angry revenge-seeking Seattle.  At least that Miami game should solve the problem where the St. Louis defense has very few takeaways.

Yes, Seattle still has the top-ranked defense, thanks to their ridiculous performance in the first three games of the season.  Seattle's defensive DVOA before the bye was -42.1%.  Seattle's defensive DVOA since the bye is a more pedestrian -3.0%.  Grant Winstrom is out a few weeks.  Seahawk fans, be worried.

* * * * *

This week we've posted the first offensive line and defensive line statistics for the 2004 season, with adjusted line yards, adjusted sack rate, and all the extra bells and whistles.  Thanks again to John Argentiero, whose amazing Excel macro programming has made producing these statistics each week much easier.

(Before I go into some notes on adjusted line yards, here's a quick explanation: they represent all rushes from running backs, but with all long runs truncated at 10 yards to get an idea of how much of the running game is good blocking and how much is a running back in the open field.  Explained more here.)

Indianapolis and the New York Jets are 1-2 in offensive line yards this season, and both running games are shaped in a similar fashion.  Neither Edgerrin James or Curtis Martin have had many long runs, but they are also hardly ever stuffed at the line of scrimmage.  Both running backs are consistent, high in running back success rate, able to take advantage of holes made by the offensive line but lacking breakaway speed.  Yes, that includes James, who like fastballer-turned-junkballer Frank Tanana has come out of surgery as a completely different yet still highly skilled player.  The Jets are also number one in adjusted sack rate; Chad Pennington has only been sacked six times despite facing two good division defenses, Miami and Buffalo.  The Colts are seventh in adjusted sack rate, although Manning has been sacked even fewer times than Pennington.  These are two very good, very underrated (with the exception of Kevin Mawae) offensive lines.

A very different kind of offensive line belongs to the New York Giants, ranked fourth in offensive line yards.  Notice that the Giants are #1 in 10+ YARDS, meaning an overwhelming number of their rushing yards have come on long runs.  They've only converted one of five chances on 3rd/4th-and-short, worst in the league (although a small sample size).  They are near the bottom of the league in adjusted sack rate.  There's been a lot of talk about an improved offensive line but the Giants' improvement in the ground game is primarily Barber, not his blockers.

On the other side, you have the Houston Texans.  Not the worst offensive line in the league, but certainly this is not where the improvement has been that makes them 3-3.  The Texans are near the bottom in our metrics for measuring both run blocking and pass blocking, and on top of this they have had a grand total of two percent of their rushing yards on long runs (after 10 yards past the line of scrimmage) which is far and away the worst total in the league.

Miami, of course, ranks last in offensive line yards and first in defensive line yards.  A look at the top teams in defensive line yards shows how stopping the run has been instrumental in the turnarounds of the Atlanta and Washington defenses.  Atlanta, 29th in defensive line yards a year ago, is so far second; Washington, 28th in defensive line yards a year ago, is so far fifth.  Their opposite is Dallas, who have had all kinds of problems stopping the run this season and have dropped from fourth in defensive line yards to 24th.

I've always said that I am a writer who uses numbers to tell a story, not a statistician who writes articles, and one team whose numbers tell a pretty clear story is the New England Patriots.  New England fans know that stopping the run has been the Achilles heel of their seemingly perfect team this season, but the numbers give more details.  The Patriots last year were 12th in defensive line yards; this year they've plummeted to 30th.  They are last in the league at stuffing runs at the line of scrimmage or for a loss.  And yet, just like in 2003, they are first in the league at preventing long runs more than 10 yards past the line of scrimmage.  So the problem is clearly with the defensive line, and most people will say that the Patriots miss Ted Washington.  I'm not so sure.  Look at the directional rushing and you see that where the Patriots have been particularly poor is not up the middle but on runs left.  Surprisingly, that's the home of the Patriots' best defender, Richard Seymour.  Teams are double teaming Seymour to take him out of the play and then running that direction.  It's getting them good yardage, but no breakaways.

The biggest issue in this week's matchup between the undefeated Patriots and undefeated Jets is going to be the ground game.  Both teams have been excellent running the ball and terrible stopping the run.  The difference is that the Patriots have had more long runs, and the Jets have allowed more long runs.  There are very good odds on Corey Dillon taking it to the house with a highlight-quality double-digit carry, and that being the difference in the ballgame.

* * * * *

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 60 percent strength and will gradually increase each week until Week 10.

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 48.9% 1 44.8% 5-0 30.4% 3 -8.8% 9 9.6% 3
2 NE 43.0% 2 37.7% 5-0 30.0% 4 -15.3% 3 -2.3% 22
3 SEA 27.1% 3 42.7% 3-2 3.1% 15 -26.0% 1 -1.9% 21
4 DEN 26.3% 10 31.5% 5-1 11.0% 11 -12.7% 6 2.6% 11
5 NYG 24.8% 5 36.0% 4-1 17.0% 9 -13.0% 5 -5.1% 29
6 NYJ 24.8% 4 42.0% 5-0 28.9% 5 7.6% 21 3.5% 10
7 IND 21.0% 6 20.4% 4-1 34.7% 2 12.6% 25 -1.2% 18
8 PIT 18.8% 7 26.9% 5-1 8.8% 13 -6.1% 11 3.8% 9
9 SD 16.5% 9 20.6% 3-3 19.6% 7 8.6% 23 5.5% 4
10 BAL 15.3% 12 22.1% 3-2 -9.4% 25 -10.9% 7 13.8% 1
11 ATL 7.2% 13 20.4% 5-1 -6.9% 19 -9.5% 8 4.6% 7
12 BUF 5.4% 18 -0.7% 1-4 -2.0% 17 -3.3% 12 4.2% 8
13 KC 5.2% 11 2.1% 1-4 18.7% 8 4.9% 18 -8.5% 32
14 DAL 4.2% 14 -3.6% 2-3 13.0% 10 9.1% 24 0.4% 14
15 HOU 1.4% 16 2.5% 3-3 5.6% 14 5.4% 19 1.3% 13
16 JAC 1.1% 20 -4.6% 4-2 0.3% 16 0.8% 15 1.6% 12
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 MIN -0.7% 15 0.6% 4-1 35.4% 1 30.3% 32 -5.7% 30
18 DET -4.0% 8 -2.5% 3-2 -11.3% 26 -2.5% 13 4.8% 6
19 TB -8.4% 19 -10.2% 1-5 -14.3% 30 -7.7% 10 -1.8% 20
20 STL -9.1% 17 -10.3% 4-2 20.2% 6 25.2% 31 -4.1% 28
21 WAS -10.3% 27 -14.6% 2-4 -22.1% 31 -17.8% 2 -5.9% 31
22 CLE -10.6% 26 -15.8% 3-3 -8.0% 20 7.8% 22 5.3% 5
23 ARI -13.7% 21 -9.5% 1-4 -12.5% 27 -2.0% 14 -3.3% 26
24 GB -14.3% 32 -19.3% 2-4 9.8% 12 23.5% 30 -0.5% 15
25 CAR -14.3% 23 -22.6% 1-4 -8.1% 21 3.1% 17 -3.0% 25
26 NO -14.5% 24 -5.7% 2-4 -2.3% 18 23.0% 29 10.9% 2
27 CHI -15.9% 25 -16.1% 1-4 -13.7% 29 1.3% 16 -0.9% 16
28 TEN -18.2% 22 -9.1% 2-4 -8.4% 22 7.3% 20 -2.5% 23
29 CIN -25.9% 29 -29.7% 1-4 -8.7% 23 13.4% 26 -3.9% 27
30 OAK -30.2% 28 -37.4% 2-4 -13.6% 28 13.6% 27 -3.0% 24
31 SF -30.9% 30 -41.1% 1-5 -9.3% 24 20.2% 28 -1.4% 19
32 MIA -38.8% 31 -43.6% 0-6 -52.6% 32 -14.9% 4 -1.1% 17

  • 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 48.9% 5-0 4.7 1 -2.0% 21 -1.2% 21 8.5% 24
2 NE 43.0% 5-0 4.7 2 0.2% 16 -1.9% 22 14.4% 12
3 SEA 27.1% 3-2 3.1 11 -4.0% 24 -7.2% 31 10.4% 21
4 DEN 26.3% 5-1 4.1 3 -5.0% 26 -7.6% 32 12.7% 16
5 NYG 24.8% 4-1 3.8 6 3.6% 13 2.2% 14 9.8% 22
6 NYJ 24.8% 5-0 4.0 4 -14.7% 32 7.4% 6 2.8% 31
7 IND 21.0% 4-1 3.5 8 -3.7% 23 2.6% 13 4.2% 28
8 PIT 18.8% 5-1 3.7 7 -14.3% 31 11.7% 2 15.3% 11
9 SD 16.5% 3-3 3.8 5 7.1% 5 -5.0% 28 22.8% 5
10 BAL 15.3% 3-2 3.0 15 -4.6% 25 10.5% 4 22.4% 6
11 ATL 7.2% 5-1 3.4 9 -9.2% 30 -0.7% 20 21.4% 7
12 BUF 5.4% 1-4 3.0 14 0.0% 17 0.0% 17 13.7% 13
13 KC 5.2% 1-4 2.9 17 6.0% 6 2.6% 12 11.2% 19
14 DAL 4.2% 2-3 3.0 12 4.4% 10 7.3% 7 12.9% 15
15 HOU 1.4% 3-3 3.0 13 -5.2% 27 3.6% 10 16.4% 9
16 JAC 1.1% 4-2 3.2 10 9.4% 3 -4.1% 27 7.8% 25
TEAM TOTAL
VOA
W-L ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VARIANCE RANK
17 MIN -0.7% 4-1 3.0 16 4.8% 8 -0.6% 19 7.1% 26
18 DET -4.0% 3-2 2.6 21 5.5% 7 -2.0% 23 32.0% 3
19 TB -8.4% 1-5 2.6 20 -1.8% 20 -6.7% 29 11.7% 18
20 STL -9.1% 4-2 2.7 19 -5.5% 28 3.7% 9 10.4% 20
21 WAS -10.3% 2-4 2.1 25 1.6% 15 7.0% 8 13.6% 14
22 CLE -10.6% 3-3 2.7 18 4.5% 9 10.9% 3 20.5% 8
23 ARI -13.7% 1-4 1.9 26 -0.8% 18 0.3% 16 12.5% 17
24 GB -14.3% 2-4 2.5 22 -1.1% 19 1.5% 15 35.4% 2
25 CAR -14.3% 1-4 2.5 23 14.7% 1 -7.2% 30 4.1% 29
26 NO -14.5% 2-4 2.3 24 -5.9% 29 -0.1% 18 4.6% 27
27 CHI -15.9% 1-4 1.8 27 3.9% 11 -2.2% 24 1.8% 32
28 TEN -18.2% 2-4 1.6 31 -2.2% 22 -2.2% 25 38.3% 1
29 CIN -25.9% 1-4 1.6 29 1.9% 14 13.4% 1 3.5% 30
30 OAK -30.2% 2-4 1.8 28 10.8% 2 3.1% 11 25.2% 4
31 SF -30.9% 1-5 1.6 30 3.6% 12 -3.5% 26 15.5% 10
32 MIA -38.8% 0-6 0.7 32 8.0% 4 7.8% 5 9.1% 23

Posted by: Aaron Schatz on 20 Oct 2004

1 comment, Last at 18 Oct 2005, 10:39pm by Shred Nard

Comments

1
by Shred Nard (not verified) :: Tue, 10/18/2005 - 10:39pm

This week's fox sports nfl power rating fail dearly. Your methods of ranking suck ass!