Writers of Pro Football Prospectus 2008

Most Recent FO Features

HackenbergChr15.jpg

» SDA: From Death Valley to Happy Valley

While Saturday won't give us an answer to which is the best one-loss team, it is still a big week for conference races across the country.

04 Jan 2010

Week 17 DVOA Ratings

by Aaron Schatz

2009 was a very strange NFL season, at least when it comes to statistics, and it ends with a particularly strange Week 17. Take a season with a lot of good teams packed tightly together in total efficiency, add in a week of unexpected blowouts and teams playing like it was the preseason, and you get some very screwy-looking final 2009 DVOA ratings.

Sitting at the top of our ratings for the entire season are the Baltimore Ravens, who barely made the playoffs with a 9-7 record. The Ravens are the second wild card team to top the ratings, following last year's Eagles; because the Eagles had a tie, the Ravens now set a mark for the worst win-loss record by the team that leads the league in DVOA. As noted in last week's column, DVOA says the Ravens were much better than their 9-7 record even though they don't come off as conventionally "unlucky" in any way. Their schedule was actually a little easier than average. They were reasonably consistent and they excelled at the right times -- in fact, they were also the league's best team according to our estimated wins formula, with 12.1 estimated wins. They just lost close games to good teams. It was a very strange season.

The Ravens actually were one of three wild card teams at the top of the DVOA ratings. The highest division champion was New England, fourth in DVOA despite going just 10-6. You may not have noticed, but actually only two teams in the entire AFC were better than 10-6. The Chargers and Colts are of course the favorites to come out of the AFC for Super Bowl XLIII, but they aren't favored by DVOA -- San Diego because they started slow, Indianapolis because they gave up on their last two games, and both teams because they had more regular wins than Pythagorean wins.

Actually, those Pythagorean win totals give a great demonstration of how little separates the top teams in 2009. For those who don't know, Pythagorean wins are a simple projection of how many games a team would be expected to win based only on points scored and allowed. Green Bay led the league with 11.8 Pythagorean wins, making this just the second 16-game season where no team finished with at least 12 Pythagorean wins. (The other was 2003, when New England and Kansas City tied for the league lead with 11.4 projected wins.) However, eight different teams this year finished between 11.1 and 11.8 Pythagorean wins. Before 2009, there had never been a season where more than six teams had at least 11 Pythagorean wins.

That group of eight, by the way, doesn't include Indianapolis. Because of their colossal mail-in job over the last two weeks, the Colts end up with 10.8 projected wins -- yet they still end up with the fifth highest difference between regular and Pythagorean winning percentage since the AFL-NFL merger. (They were on their way to breaking the record until Week 16.)

The "sitting starters" thing has completely wreaked havoc on the weighted DVOA ratings that we use for the playoff odds simulation. Green Bay comes out as the best team in football right now, lowering the strength of earlier games... but the Colts drop to 14th thanks to the Curtis Painter experience. To make the playoff odds simulation a little bit more accurate, we've adjusted things by using weighted DVOA through Week 16, not Week 17, for four teams whose games were affected by sitting starters: Arizona, Green Bay, New Orleans, and San Diego. We're also using the same "halfway through Week 16" rating we used last week for Indianapolis. These adjustments make for a smaller gap between Green Bay and the rest of the league and move Indianapolis up to sixth, which is much more realistic.

The best team in weighted DVOA that won't be playing in the postseason is Carolina. The Panthers have been sensational since midseason and finish the year fourth in weighted DVOA, but believe it or not they don't end up with the all-time biggest difference between total season DVOA and weighted DVOA. That record belongs to the 2001 Washington Redskins.

Top 10 Difference Between Total and Weighted DVOA, 1993-2009
Year Team Total
DVOA
Rank Weighted
DVOA
Rank Difference Season
W-L
Next Season
W-L
2001 WAS 0.4% 16 23.0% 5 22.6% 8-8 7-9
2002 NYJ 16.0% 8 36.2% 1 20.2% 9-7 6-10
2009 CAR 10.2% 14 28.9% 4 18.7% 8-8 --
1993 HOIL 17.0% 5 34.4% 1 17.4% 12-4 2-14
1996 TB -10.6% 20 6.1% 13 16.7% 6-10 10-6
1997 ATL -5.6% 20 10.3% 8 15.9% 7-9 14-2
2006 TEN -9.6% 22 5.7% 14 15.3% 8-8 10-6
2005 MIN -18.0% 24 -3.0% 18 15.0% 9-7 6-10
1993 TB -32.9% 27 -18.2% 23 14.7% 5-11 6-10
2000 JAC 3.2% 16 17.8% 8 14.6% 7-9 6-10

As you can see, these big finishes didn't necessarily guarantee a playoff run the following season. The 1997-1998 Atlanta Falcons are the example the Panthers hope they can follow, but most of these teams actually won fewer games the next year. And yes, that table does include DVOA ratings for the 1993 Houston Oilers and Tampa Bay Buccaneers. That's a little addition to our historical data that I'm saving for the dead week after the conference championship games.

Of course, while the top of the league was well-balanced in 2009, the bottom of the league was anything but. Perhaps the best statement about the current state of the league's worst franchises is that the five teams that finished at the bottom of the 2008 ratings also finished at the bottom of the 2009 ratings in slightly different order. It would have been the bottom six teams, except Cleveland's end-year winning streak moved the Browns into 26th place for the year, slightly ahead of Tampa Bay. Detroit becomes just the second team in DVOA history to finish with a rating below -50%, actually coming out worse than they did during their 0-16 season of 2008, and worse than every team since 1993 except for the 2005 San Francisco 49ers. St. Louis is 31st in DVOA but last in Pythagorean wins, with just 1.6 projected wins. That's the third-lowest Pythagorean total for any team in a 16-game season, ahead of only the 1991 Colts (1.4) and the 2000 Browns (1.5).

Two units this year come close to the best of the DVOA Era, starting with the San Diego pass offense. San Diego ends up with 62.2% pass offense DVOA, which trails only the 2007 Patriots and the 2004 Colts. Astonishingly, the Chargers did this despite finishing dead last in the league in run offense. They completely obliterate the previous record for the biggest gap between a team's pass offense DVOA and run offense DVOA, which belonged to the 2003 Tennessee Titans. That team had a gap of 64.7% DVOA; the Chargers have a gap of 73.4% DVOA.

The other historically great unit in 2009 is Cleveland's special teams. Cleveland's rating of 8.8% is twice as high as any other team, and currently ranks as the third-highest rating of the DVOA Era behind the 2002 Saints and the 2007 Bears. The Vikings were higher than Cleveland as of a few weeks ago, but slipped in the last few weeks, and no longer have the best year-to-year improvement of any special teams in the DVOA Era. Still, the Vikings go from 32nd last year to an impressive third this year.

And yes, the Pittsburgh Steelers finish with the worst net kickoff value of any team in our database. They aren't dead last in special teams because they actually were positive in three of the other four areas of special teams that make up our ratings.

When it comes to individual stats, 2009 was emphatically a year for quarterbacks. Passing stats have never been better around the NFL. Tom Brady ends up leading the league with 2,170 passing DYAR, the fourth-highest total of the DVOA Era. Peyton Manning is second with 1,932 passing DYAR... and that's the fifth-highest total of the DVOA Era. In total, six quarterbacks from 2009 end up in the all-time top 15 for passing DYAR.

Top 15 Passing DYAR, 1993-2009
Year Player Team DYAR
2007 T.Brady NE 2,788
2004 P.Manning IND 2,493
2006 P.Manning IND 2,308
2009 T.Brady NE 2,170
2009 P.Manning IND 1,932
2004 D.Culpepper MIN 1,929
2008 D.Brees NO 1,921
2009 P.Rivers SD 1,919
2009 D.Brees NO 1,846
2007 P.Manning IND 1,845
2009 B.Favre MIN 1,803
2009 M.Schaub HOU 1,789
2008 P.Manning IND 1,783
2000 P.Manning IND 1,766
2003 P.Manning IND 1,757

I know a number of readers disagree with the opponent adjustments that give Brady the season lead in passing value. As I noted in an Extra Points post a couple weeks ago, Brady has played against the hardest schedule of opposing pass defenses of any quarterback in the past 17 years. Six of his games came against the top four pass defenses according to DVOA, with four others against the Ravens (7), Saints (9), and Dolphins (11). Against all of these defenses, except for New Orleans, Brady outplayed most or all of his contemporaries. He had the second and fifth-highest DYAR totals allowed by the Jets this year. He had the highest DYAR total allowed by the Broncos, the highest DYAR total allowed by the Dolphins (in the Week 9 game), and the third-highest DYAR allowed by the Panthers. His 250 DYAR against the Bills in Week 1 was double what any other quarterback did against them for the rest of the season.

Some other individual stat notes from 2009, both good and bad:

  • JaMarcus Russell finishes the season with -755 passing DYAR, which is the fourth-lowest season of all-time. He ends up ahead of David Carr in 2002, Bobby Hoying in 1998, and Alex Smith in 2005.
  • Chris Johnson easily leads the league with 343 rushing DYAR, nearly 100 ahead of every other running back, but his awesome 2,000-yard season doesn't even come close to the top rushing DYAR seasons of all-time because it was so boom-and-bust. (Priest Holmes in 2002 and Terrell Davis in 1998 each had over 500 rushing DYAR.) Johnson finishes with a 45 percent Success Rate, 32nd in the league among backs with at least 100 carries. Ryan Grant passed Ray Rice in the final week to rank second in rushing DYAR and quietly had a very good season. Do you remember anyone even mentioning Ryan Grant this year?
  • Sidney Rice finishes up as the league leader in receiving DYAR for wide receivers, although Vincent Jackson leads qualifying receivers in DVOA and is second in DYAR. Mike Thomas of Jacksonville actually edges out Wes Welker for the league lead in catch rate... albeit on 100 fewer passes. (Thomas caught 77 percent of 62 passes, Welker 76 percent of 162 passes.)
  • Although Darrius Heyward-Bey didn't end up with enough pass targets to make our main ranking list for wide receivers, there has to be some sort of recognition for his mind-blowing 23 percent catch rate on 40 passes. No other player since 1994 has put up a catch rate lower than that with at least 25 passes.
  • Antonio Gates sets a new record for the best DYAR season by a tight end with 358 DYAR. Here are your all-time single-season leaders for DYAR by a tight end, which goes back to 1994 since I haven't had the time to finish the 1993 receiving numbers yet. Gates has four of the top ten seasons, and one other 2009 tight end sneaks onto the list at the bottom:
Top 10 TE Receiving DYAR, 1994-2009
Year Player Team DYAR
2009 A.Gates SD 358
2004 T.Gonzalez KC 329
2004 A.Gates SD 325
2000 T.Gonzalez KC 304
1995 M.Chmura GB 280
2007 A.Gates SD 278
1996 S.Sharpe DEN 278
1999 R.Dudley OAK 278
2005 A.Gates SD 270
2009 D.Clark IND 261

Fun end-season housekeeping: All team and individual stats pages should now be updated with final 2009 numbers. FO Premium will be updated with final 2009 ratings sometime tonight. We'll get all the 2009 stats onto the player pages sometime in the next week or two. Loser League results will be announced in Scramble for the Ball on Wednesday.

* * * * *

These are the Football Outsiders team efficiency ratings through the end of the 2009 regular season, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)

OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. WEIGHTED DVOA is adjusted so that earlier games in the season become gradually less important. It better reflects how well the team is playing right now. LAST WEEK represents rank after Week 16, while LAST YEAR represents rank in 2008.

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

To save people some time, please use the following format for all complaints: <team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>


TEAM TOTAL
DVOA
LAST
WEEK
LAST
YEAR
WEIGHTED
DVOA
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
1 BAL 32.5% 2 3 31.4% 2 9-7 16.9% 8 -13.6% 3 2.0% 8
2 GB 30.3% 4 11 37.5% 1 11-5 22.5% 4 -14.0% 2 -6.2% 32
3 PHI 28.6% 1 1 20.6% 7 11-5 13.3% 10 -11.0% 6 4.3% 2
4 NE 28.3% 3 9 27.6% 5 10-6 29.7% 1 2.6% 17 1.2% 13
5 DAL 26.0% 6 19 31.2% 3 11-5 24.2% 3 -0.7% 10 1.0% 14
6 NO 23.4% 5 13 13.5% 12 13-3 27.6% 2 1.3% 14 -2.8% 28
7 MIN 17.7% 10 18 13.4% 13 12-4 15.2% 9 1.7% 15 4.2% 3
8 IND 17.1% 7 8 9.8% 14 14-2 19.6% 6 1.8% 16 -0.6% 20
9 PIT 16.4% 11 2 17.6% 10 9-7 17.6% 7 -2.9% 9 -4.1% 30
10 NYJ 16.3% 13 17 19.6% 8 9-7 -9.7% 22 -23.4% 1 2.6% 6
11 SD 13.4% 8 7 22.8% 6 13-3 21.8% 5 8.8% 23 0.4% 16
12 DEN 11.3% 9 22 3.8% 16 8-8 3.5% 18 -7.9% 7 -0.1% 18
13 ARI 10.5% 12 21 13.9% 11 10-6 8.8% 13 -0.1% 12 1.6% 11
14 CAR 10.2% 18 6 28.9% 4 8-8 1.5% 20 -11.6% 5 -2.9% 29
15 HOU 9.9% 14 23 18.9% 9 9-7 12.9% 11 5.2% 20 2.2% 7
16 MIA 6.0% 16 14 3.5% 17 7-9 7.3% 16 3.1% 18 1.8% 9
TEAM TOTAL
DVOA
LAST
WEEK
LAST
YEAR
WEIGHTED
DVOA
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
17 NYG 5.0% 15 4 -5.8% 23 8-8 11.3% 12 4.8% 19 -1.5% 27
18 ATL 2.3% 19 16 -3.5% 21 9-7 8.7% 14 5.4% 21 -1.0% 22
19 SF 1.1% 20 25 8.3% 15 8-8 -10.6% 23 -12.3% 4 -0.6% 19
20 CIN 0.4% 17 26 -4.0% 22 10-6 1.7% 19 0.4% 13 -0.9% 21
21 WAS -5.3% 22 12 -1.5% 19 4-12 -4.7% 21 -0.7% 11 -1.3% 23
22 TEN -6.1% 21 5 -0.3% 18 8-8 7.6% 15 12.3% 27 -1.3% 24
23 BUF -7.9% 24 24 -1.7% 20 6-10 -16.4% 27 -7.2% 8 1.3% 12
24 JAC -9.4% 23 20 -13.2% 24 7-9 6.0% 17 14.1% 28 -1.4% 25
25 CHI -20.0% 25 15 -23.7% 27 7-9 -17.3% 28 6.7% 22 4.0% 4
26 CLE -23.6% 27 27 -17.0% 25 5-11 -13.5% 24 18.8% 30 8.8% 1
27 TB -23.9% 26 10 -19.0% 26 3-13 -15.9% 26 11.1% 25 3.1% 5
28 KC -28.4% 29 30 -28.2% 29 4-12 -15.0% 25 11.9% 26 -1.5% 26
29 SEA -31.3% 28 28 -42.5% 30 5-11 -17.9% 29 14.3% 29 0.9% 15
30 OAK -32.8% 30 29 -24.9% 28 5-11 -22.4% 30 10.3% 24 -0.1% 17
31 STL -44.4% 31 31 -43.9% 31 1-15 -26.8% 32 19.2% 31 1.6% 10
32 DET -51.3% 32 32 -49.1% 32 2-14 -26.3% 31 20.6% 32 -4.4% 31

  • 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.
  • NON-ADJ TOTAL VOA lists performance without adjustments for fumble recovery, weather/altitude, or opponent strength.
  • 2009 SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative).
  • PYTHAGOREAN WINS represent a projection of the team's expected wins based solely on points scored and allowed.
  • VARIANCE measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from most consistent (#1, lowest variance) to least consistent (#32, highest variance).


TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK 2009
SCHED
RANK PYTH
WINS
RANK VAR. RANK
1 BAL 32.5% 9-7 34.0% 12.1 1 -1.8% 23 11.6 4 12.8% 12
2 GB 30.3% 11-5 39.6% 11.0 5 -9.0% 31 11.8 1 17.9% 20
3 PHI 28.6% 11-5 30.4% 10.9 6 0.5% 18 10.2 10 21.1% 25
4 NE 28.3% 10-6 21.7% 11.2 4 6.0% 7 11.6 3 20.0% 24
5 DAL 26.0% 11-5 23.4% 11.3 3 1.9% 13 11.3 7 8.1% 3
6 NO 23.4% 13-3 26.8% 11.6 2 -1.4% 22 11.6 5 16.2% 16
7 MIN 17.7% 12-4 23.1% 10.1 9 -6.6% 29 11.6 2 11.4% 10
8 IND 17.1% 14-2 19.3% 10.9 7 0.7% 17 10.8 9 16.8% 17
9 PIT 16.4% 9-7 22.5% 10.4 8 -2.6% 26 9.2 14 11.3% 9
10 NYJ 16.3% 9-7 18.8% 9.3 14 2.7% 10 11.4 6 22.0% 26
11 SD 13.4% 13-3 17.1% 10.1 10 -1.3% 21 11.1 8 7.3% 1
12 DEN 11.3% 8-8 12.4% 9.4 13 1.9% 14 8.1 18 17.5% 19
13 ARI 10.5% 10-6 15.5% 10.0 11 -9.1% 32 9.3 13 22.0% 27
14 CAR 10.2% 8-8 3.7% 9.0 15 8.0% 2 8.2 17 24.0% 30
15 HOU 9.9% 9-7 15.1% 9.6 12 -3.2% 27 9.4 12 8.9% 4
16 MIA 6.0% 7-9 1.6% 8.8 16 7.9% 3 7.2 20 9.6% 5
TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK 2009
SCHED
RANK PYTH
WINS
RANK VAR. RANK
17 NYG 5.0% 8-8 -4.6% 8.2 19 6.4% 4 7.4 19 26.9% 31
18 ATL 2.3% 9-7 1.5% 8.6 17 6.1% 6 9.0 15 10.6% 8
19 SF 1.1% 8-8 6.3% 7.9 20 -7.0% 30 9.5 11 10.2% 7
20 CIN 0.4% 10-6 1.7% 8.2 18 1.1% 16 8.4 16 23.2% 28
21 WAS -5.3% 4-12 -5.8% 7.4 22 -0.1% 19 5.8 23 13.6% 13
22 TEN -6.1% 8-8 -11.8% 7.6 21 2.7% 11 6.8 21 32.9% 32
23 BUF -7.9% 6-10 -11.0% 7.1 24 4.5% 8 5.8 24 23.4% 29
24 JAC -9.4% 7-9 -7.6% 7.2 23 -2.0% 24 5.5 25 19.2% 23
25 CHI -20.0% 7-9 -19.2% 5.6 25 -0.9% 20 6.7 22 14.2% 14
26 CLE -23.6% 5-11 -27.1% 5.2 26 1.3% 15 4.3 28 15.7% 15
27 TB -23.9% 3-13 -29.4% 5.1 27 10.5% 1 3.8 29 17.2% 18
28 KC -28.4% 4-12 -25.6% 4.5 28 2.9% 9 4.7 27 11.8% 11
29 SEA -31.3% 5-11 -26.2% 3.6 30 -4.7% 28 5.0 26 18.7% 22
30 OAK -32.8% 5-11 -35.4% 4.4 29 6.2% 5 2.8 31 18.1% 21
31 STL -44.4% 1-15 -41.4% 2.5 31 -2.1% 25 1.6 32 10.1% 6
32 DET -51.3% 2-14 -51.1% 2.2 32 2.2% 12 2.9 30 7.5% 2

Posted by: Aaron Schatz on 04 Jan 2010

183 comments, Last at 15 Feb 2013, 6:57am by shankar

Comments

1
by Key19 :: Mon, 01/04/2010 - 10:46pm

Dallas defense FINALLY cracks the top 10. Hooray.

2
by Temo :: Mon, 01/04/2010 - 10:52pm

Yea, I gotta give Wade some credit. This defense looked terrible there in the beginning but has really turned it around the past few weeks.

I still hate Carpenter and hold my breath whenever anyone tests Newman deep, but they're playing real well right now.

3
by Key19 :: Mon, 01/04/2010 - 10:57pm

Yes, Newman is quickly becoming the 3rd best corner on the team. I thought Alan Ball played at least as well as Newman when he filled in for him for that series, and he's played pretty well all season.

I really hope Jason Williams makes a big jump this offseason. He seems fast enough that he should at least be competent in coverage. But I'd still like to see them draft a good cover ILB this year.

I honestly feel like putting in Colombo over Free this week might be a mistake. I love how Free has been playing, and I'm not sure Colombo is really better at this point. Maybe he is, but it was Colombo who was in there when we were struggling early this season. We've been hot with Free. I dunno. But Flo has definitely been kickin it lately. And not literally, Justin Tuck.

7
by Or (not verified) :: Mon, 01/04/2010 - 11:05pm

I expect to see Flo released this offseason. It depends on the cap situation, Halsell would know a lot more than I about how that contract works... But I think he's clearly the 3rd best Tackle on this team, while being the oldest and most highly paid.
That said, you could do worse. The Bears would probably bring him in with no hesitation.

16
by Key19 :: Mon, 01/04/2010 - 11:28pm

Trent Cole might beat you up if you told him he got stonewalled by the 3rd-best tackle on the Cowboys' line.

Because he did.

148
by tuluse :: Wed, 01/06/2010 - 5:44pm

Wow, Alan Ball. I always liked him in college and wondered if he could play in the NFL. Really cool that he was worked his way to being a productive player.

4
by Lloyd Christmas (not verified) :: Mon, 01/04/2010 - 11:01pm

I feel like a great deal of the Dallas-Eagles game was won on the lines. Dallas did a great job of making McNabb feel rushed (often more rushed than he actually was), and they did a good job of opening up holes for Marion Barber. in contrast, the Eagles' lines were held together by spit and bailing twine.

I wonder if there's a strong correlation between line play and overall success?

9
by Morton G. (not verified) :: Mon, 01/04/2010 - 11:10pm

Honestly, beating a mediocre team like the Eagles, riddled with injuries, playing with a patchwork O-Line and patchwork defense shouldn't make anyone too proud. Cowboys are not a great team - I predict no more than one playoff win for this team.

They don't beat the Vikings in their dome, and they probably don't beat the Saints in a rematch when they get some of their injured players back. And Green Bay defense would put the clamps on the Dallas O.

13
by Lloyd Christmas (not verified) :: Mon, 01/04/2010 - 11:19pm

...beating a mediocre team like the Eagles...

You's trollin'!

57
by Rick (not verified) :: Tue, 01/05/2010 - 10:12am

I think his point is that the Eagles are a very good team, but with the injuries they are dealing with at key positions (not least of which the loss of their center who hadn't missed a game in 2 years or so), the Eagles dip a bit in performance.

Fact is, the Eagles really didn't beat any great teams this year. In their losses, they played alot of teams close until the Dallas game.

But the playoffs are a different story. As much as December is a different story for Dallas, until this year, the question is are the playoffs a different story for Dallas except for this year?

One is inclined to say yes, it is. They are playing very differently than in the past this time of year.

But I'm still wondering if Wade is capable of winning a playoff game. Sometimes it's not the players, it's the coach and how the team is prepped. Wade has a bad history in this regard.

Coming off an overwhelming win like the one last week, I'd be hard pressed to say the Eagles will win in the coming week. But, on the other hand.......coming off an overwhelming win like the one last week it's possible Dallas players almost expect the same thing next week.

61
by t.d. :: Tue, 01/05/2010 - 10:58am

I don't love Wade, but since he's taken over the defense, they've moved from 'good' to 'elite'. I didn't want this matchup, but, with their injuries, the Eagles are as vulnerable as they're going to get.

63
by C (not verified) :: Tue, 01/05/2010 - 11:35am

Everybody dislikes Wade...

Can he win a playoff game? We'll Jim Mora could win playoff games ( he sucks), Norv Turner could beat Peyton Manning in the playoffs ( and people strongly dislike Norv), guys can play vanilla conservative football in the playoffs and win...

Wade is an easy target. He's fat, he talks funny, he probably broke into the business at least partially because of his dad ( and not his own merits). I think people make Wade out to be an absolute baffoon ( asking if he can he win a playoff game?). Average coaches have won playoff games, crappy coaches have won playoff games... Mike Tice is an example of an absolute baffoon, Wade Phillips is probably more like a good DC and an average head coach... and guys like him have won playoff games before.

The trendy thing to say right now is that it's (hard to beat a team 3 times in a season), and that's exactly what Dallas would have to do to Philly. In reality, teams that won the previous 2 match ups that play a 3rd time in the playoffs are 12-7 in the last 19 appearances. In reality though, if you beat a team twice you are probably a better team, and if you meet in the post season you are probably playing at home... You'd probably want to be better than 12-7 or a 63% win percentage so there might be "some" truth to the beating a team 3 times in a season but it's probably overrated by our national hype network known as ESPN.

117
by t.d. :: Tue, 01/05/2010 - 5:58pm

Mike Tice is actually a terrific line coach

19
by Key19 :: Mon, 01/04/2010 - 11:33pm

Well, it makes me proud. No one made excuses for us when we lost 44-6 last year when we had our 3rd string RB, an injured QB, the worst backup Center in the league, and Roy Williams starting at WR. And we still have Roy starting at WR!

9
by HostileGospel :: Mon, 01/04/2010 - 11:10pm

Quantifying line play is sort of tricky, isn't it? Difficult to apply serious statistical analysis there.

--
Overall, I'd be kind of embarrassed to critique something when I didn't know what the hell I was talking about, but then, oh yeah, my NAME is on what I write, isn't it?

-Les Bowen

15
by Lloyd Christmas (not verified) :: Mon, 01/04/2010 - 11:21pm
20
by HostileGospel :: Mon, 01/04/2010 - 11:43pm

Sure, but ALY is what I meant by "quantifying," which you need to do before you can look for any sort of correlation. I don't think anyone, FO included, considers ALY a definitive measure of line play such that you could apply mathematical analysis to it and hope to learn anything definitive. It's good for ranking and comparing, but to say "Well, we ran the numbers and run-blocking as measured by ALY doesn't have a meaningful correlation to winning percentage," is dumb, which is probably why no one is doing it.

--
Overall, I'd be kind of embarrassed to critique something when I didn't know what the hell I was talking about, but then, oh yeah, my NAME is on what I write, isn't it?

-Les Bowen

69
by R O (not verified) :: Tue, 01/05/2010 - 12:22pm

A truer statement was never written.

The Broncos o-line looked like the best in the league last year with that zone blocking scheme and Cutler moving all around the pocket making beautiful throws on the run.

Give them an ordinary (statue-like) QB and more standard run blocking schemes and they defined ordinary this year.

77
by C (not verified) :: Tue, 01/05/2010 - 12:41pm

Yes A QB is responsible for sacks but it's not so easy as statue bad/Mobile good...

A lot of mobile quarterbacks run into sacks and create them, while some statue quarterbacks read plays quickly and have fast releases. You also don't need a 4.6 40 to be able to slide and move in the pocket.

85
by Eddo :: Tue, 01/05/2010 - 12:53pm

R O didn't mention sacks; rather, he was discussing line play. A mobile quarterback running into sacks has nothing to do with line play.

89
by R O (not verified) :: Tue, 01/05/2010 - 1:02pm

And for all his faults, Cutler is not a scrambler who runs into sacks. Shanahan always had him moving in the pocket in order to give him more time to throw.

That is much more predictable than having a scrambler like say Mike Vick who would often run into sacks. These players can make their line look worse.

The o-line needs to know where the QB is going to be. They had that predicability with Cutler, they don't get it with a typical "scrambling" QB.

91
by C (not verified) :: Tue, 01/05/2010 - 1:12pm

Agreed that Cutler was more "predictable" and didn't run into sacks.

You could move the pocket with him on roll outs, waggle plays, bootlegs, misdirection etc. Having Mike Vick drop back, stare at the DE and then dance around the pocket looking for daylight is harder to block. He could run toward the sidelines, then cut it back etc.

If I'm a coordinator with a below average offensive line, I'm not dropping my QB on straight 3, 5, 7 step drops, I'm moving the point that D-Lineman attack, running waggles, roll outs, bootlegs, screens, draws, and keeping them guessing. I'm also going over blitz breaking plays over and over again.

92
by R O (not verified) :: Tue, 01/05/2010 - 1:18pm

Which is exactly why an average o-line looked spectacular with Shanahan having Cutler do exactly that.

94
by C (not verified) :: Tue, 01/05/2010 - 1:28pm

Shanny's a good coach, with a plan in place, a good system, and he had players to fit his plan ( athletic tackles). I really hope the Redskins don't bring him in.

95
by Rick A. (not verified) :: Tue, 01/05/2010 - 1:32pm

You guys are dead-on. Yet Denver still could not win with all of that. At home they were atrocious, the defense was God awful, and Cutler did loser things reeking of his unrealistic arrogance. His benefactor was fired. Football is a tough business...

5
by theshadowj :: Mon, 01/04/2010 - 11:01pm

I'm trying to figure out how Houston's overall DVOA went down despite outplaying NE, who is now #4.

My only guess is that Indy's plunge the last two weeks (they went from 27.9 in week 15 down to 17.1 now, which then counts for two games against Houston) has drastically reduced the opponent adjustments for those games.

26
by Anonymousnonwewwwf (not verified) :: Tue, 01/05/2010 - 12:43am

I think it also has something to do with the fact that their offense did not even slightly light up a secondary that was easily lit up by the saints or indy. There defence played much better than expected but i dont think they will do well in the playoffs even though i love them. This praise is especially high because i am from new england and drinking a narragansett tall boy (if you have ever read the back of one, you would understand the sentiment) at the moment, so houston better appreciate the fact that i am rooting for them.

45
by Rivers McCown :: Tue, 01/05/2010 - 6:39am

Yes, I bet they're excited to take on the Bengals. Pesky Sunday night results be damned.

6
by lester bangs (not verified) :: Mon, 01/04/2010 - 11:04pm

San Diego 11th overall? Throw the Commodore 64 out.

8
by Or (not verified) :: Mon, 01/04/2010 - 11:06pm

San Diego's low DVOA has long confused me. Especially since FO loved them so much in the preseason.

14
by Temo :: Mon, 01/04/2010 - 11:20pm

Well, I mean in-season ratings have no reliance on pre-season predictions.

As far as their games, they actually only have 3 games with negative DVOA-- Week 1 Oakland, Week 2 Baltimore, and week 6 Denver. But they just don't have as many "domination" games as other teams, plus the week 1 Oakland game is really weighted downwards.

161
by SDfan (not verified) :: Thu, 01/07/2010 - 4:16pm

The Chargers are #1 in variance, so they have been very consistent, so I'm not surprised they only had 3 games with -DVOA scores. You mentioned that they don't have as many domination games as other teams... they did win some close ones, but they also won 4 games by 25 points or more. That sounds pretty dominant to me!

17
by tally :: Mon, 01/04/2010 - 11:29pm

Re #6 & #8:

Have you seen their defense? Or their rush offense? Absolutely putrid. A bunch of stuffs on runs by LDT and Sproles was offset only by an otherworldly pass offense. Their defense has given up huge leads, in part because the Chargers can't run out the clock and in part because it's just not that good.

22
by Or (not verified) :: Mon, 01/04/2010 - 11:56pm

Yes, but take into account the overall game... can anyone here honestly say they believe the Jets or Steelers are a more dangerous team than the Chargers? It's silly. The Chargers are phenomenal. Maybe It's buoyed by one particular strength, but why should that count against them if they can exploit it to the tune of convincing victory after convincing victory?

24
by Aaron Schatz :: Tue, 01/05/2010 - 12:08am

Try looking at weighted DVOA instead of total DVOA.

72
by dbostedo :: Tue, 01/05/2010 - 12:25pm

Well I'm sure you're talking about danger right now, but earlier in the year, the Steelers did beat the Chargers. And while their passing offense might do things like make the Chargers more "dangerous" and make other teams not want to play them, their lack of running offense and somewhat poor defense means that they are probably more likely to lose than you would think.

76
by R O (not verified) :: Tue, 01/05/2010 - 12:38pm

You need to look a little more closely at the stats.

The Chargers actually RARELY get stuffed in the run (6th best in the league). They also rarely give up huge runs (7th best at give up open field, 10 yard or more, gains). That is all you need with an offense as explosive as the Chargers.

The Chargers have been running out the clock fine. They make teams use huge chunks of time making comebacks and they run out of clock. Rivers almost always leads timely fourth quarter drives when necessary.

The Chargers do what they need to to win. They don't waste energy blowing out bad teams when they can just play "safe" and cruise to a win by minimizing mistakes.

I really think DVOA should account for how much time the Defense of a winning team forces the offense to chew up in the fourth quarter going on scoring drives. The Chargers have been masters at making teams take too much time to score in the second halves of games this year.

79
by C (not verified) :: Tue, 01/05/2010 - 12:44pm

I was watching the Chargers more as the year went on, and I think Steven Cooper and Shaun Phillips are better than the perception I had in my mind. Their LB core is pretty freaking legit. If Cesaire is healthy and they could put Willians in the middle that defense should be a lot better.

87
by R O (not verified) :: Tue, 01/05/2010 - 12:56pm

Brandon Siler may be even better than Cooper, who has been very solid over the years for the Chargers.

The Chargers also have a few good LB's on injured reserve. They are looking to be scary good at linebacker next year. If Jamal Williams does come back with a decent year, lookout.

93
by C (not verified) :: Tue, 01/05/2010 - 1:23pm

I agree, and if Larry English can continue to get better, Tim Dobbins is a nice backup etc... Nothing against Burnett but I like Cooper and Phillips better...

I don't think SD's linebackers get the credit they deserve here on the east coast. Who has the best 3-4 LB core? San Diego, Pittsburgh, Dallas, San Fran, or Green Bay?

96
by Rick A. (not verified) :: Tue, 01/05/2010 - 1:47pm

Exactly R O. It is amazing to see how many on these threads do not see what
S.D.'s success is about. Including the hosts who, amongst other minimizations, a few weeks ago proclaimed LT as "toast"...

100
by Joe Curwen (not verified) :: Tue, 01/05/2010 - 2:28pm

I think this site's statistics provides the best overall evaluation of the Chargers' play this season. They rank 6 in wDVOA, which is about right.

The large difference in wins and estimated wins seems a matter of luck to me and is very unlikely to be repeated. I've seen all the games and what strikes me is the utter sameness of their performance week to week. The Chargers are #1 in variance for a reason.

A good team with low variance can still lose if they are unlucky in a particular game. But this season, the Chargers were never unlucky - and that is pretty darn lucky.

102
by Eddo :: Tue, 01/05/2010 - 2:40pm

You're saying he's not toast? The guy averaged 3.3 yards per carry this season. And it's not like teams can stack the box against the Chargers.

Tomlinson had an excellent, Hall-of-Fame-level career, but at this point, he's done.

105
by Rick A. (not verified) :: Tue, 01/05/2010 - 3:16pm

Yes, that's right. I'm saying he's not "toast". What he's had was a serious, lingering, injury which at his age was tough to heal. He, being a top level individual, obviously did not resort to PED's to heal it quicker, unlike some others on his team. In the last half of the year he's finally coming around from that injury. He should have a good playoffs and improve on that 3.3...

107
by C (not verified) :: Tue, 01/05/2010 - 3:28pm

Not sure if he's injured ( but I'd believe it). LDT did look like about 75% of his former ( great) self though. He looked at least 10 pounds heavier (in the legs, glutes/lower body), and a linger injury wouldn't surprise me. I hope he's back to 100% as he seems like a great guy.

108
by Eddo :: Tue, 01/05/2010 - 3:39pm

So let me get this straight: an old running back suffers a tough-to-recover-from injury, has a rather poor season, and you expect him to bounce back? I'd like to see some reasoning for that besides "he used to be a great running back". Next season, when he's another year older, why should he be any less likely to be affected by injury? Why do you expect this specific lingering injury to heal completely when you freely admit it's "tough to heal" at his age (which is only increasing)?

As an aside, PEDs and character have nothing to do with this discussion.

EDIT: I wanted to check your claim that he did better as the season progressed, and, well, he didn't really.

Game-by-game yards-per-attempt figures:
4.23
2.14
3.89
3.09
3.11
1.83
4.00
3.65
3.00
3.20
2.38
3.69
3.69
(He also rushed for 1 yard on 2 attempts in his final game, but I'll leave that out because it was only over two carries.)

I don't see an upward trend. There's not really a downward trend, either. He's been pretty consistently mediocre-to-below-average this year. If he hadn't been getting carries at the goal line to push up his fantasy value, there would likely be a consensus that he had a rather poor season.

120
by Arkaein :: Tue, 01/05/2010 - 6:28pm

Mediocre to below average is putting it kindly based on those numbers.

More games below 3.0 YPC (3) than above or equal to 4.0 YPC (2). I was going to say no games above 5.0 YPC, but even that would be generous considering his best game was 4.23 YPC!

If his best game had actually been his season average he would still only be considered to have had a decent season.

125
by Rick A. (not verified) :: Tue, 01/05/2010 - 8:36pm

He has pretty well completely healed, closer to normal at around the midpoint of the season. You left out of your charts the teams he faced and the conditions. Watch him do well in the playoffs hater...

104
by LittleTrain (not verified) :: Tue, 01/05/2010 - 2:58pm

Although generally the Chargers can't seem to run or stop the run, from watching them it seems they have the ability to do both when it matters. The defense seems to get stops when the game is close, or at least hold teams to FGs in the red zone. And the offense has shown it can move the ball and chew the clock when necessary (e.g., the beautiful FG drives at the end of the Philly and Dallas games, as well as the entire 2nd half of the 2nd Denver game). Chargers have been able to get ahead, which negates somewhat the inability to stop the run. And the D has given up a lot of yards when they are willing to trade yards for time off the clock (again, the Philly game and Dallas's last drive are examples, as is Chris Johnson's TD run at the end of the TN game where it looked like the Chargers were purposely trying not to tackle him).

While the Chargers have obvious flaws, I just can't believe that this team is rated lower than last year's 8-8 team that could have easily lost 2 more times to the lowly Chiefs.

11
by Bobman :: Mon, 01/04/2010 - 11:15pm

Disappointed the Eagles aren't still #1.

I smell a conspiracy.

52
by ammek :: Tue, 01/05/2010 - 9:11am

Harbaugh is from the Andy Reid coaching tree. Just keeping it in the family.

64
by C (not verified) :: Tue, 01/05/2010 - 11:38am

Mike Tanier was playing with the DVOA last night and wanted to rig it for his favorite teams... Of course he had to throw the Patriots up there too.

You can no longer complain about the Patriots bias, it's the Tanier bias.

70
by jonnyblazin :: Tue, 01/05/2010 - 12:22pm

Harbaugh should really be considered to be from the Harbaugh coaching tree, all things considered. Since his offense and defense are completely different from Reid's (6 OL, unbalance line, power running and play action passing for offense, 3-4 hybrid, passive zone coverage on D) one wonders what he learned under big Andy.

73
by C (not verified) :: Tue, 01/05/2010 - 12:27pm

Correct me if I'm wrong, but that offense is more a function of Cam Cameron. Now he didn't do some of those things in San Diego ( he was good there though), but he's trying to maximize his offense with what he has. Bill Parcells could have a passing offense with Bledsoe/Simms, or he could have a run the ball and play defense team. I think the best coaches adapt to their personel and don't try and force feed their system to their talent ( Steve Spurrier fun & gun).

Cameron is one of the best offensive coordinators in my book ( but not a good head coach). Combine that with his arrogance and he sort of reminds you of Brian Billeck.

77
by R O (not verified) :: Tue, 01/05/2010 - 12:41pm

Where do you get the idea that Cameron is arrogant? I've never had that impression.

81
by C (not verified) :: Tue, 01/05/2010 - 12:47pm

It's perception and I don't see him as arrogant as Billeck, but the thing that sticks out in my mind was when he was trying to talk over a booing Dolphins crowd to justify the high 1st round selection of Ted Ginn in the draft a few years ago.

No, he's not Joe Buck level arrogance, or Brian Billeck, but he seemed pretty sure of himself.... more so than most people.

84
by R O (not verified) :: Tue, 01/05/2010 - 12:53pm

"but he seemed pretty sure of himself"

Well I would think that would be a requirement for a head coach. If you're not "sure of yourself" then you are...Jim Zorn. 'Nuff said.

12
by Paul R :: Mon, 01/04/2010 - 11:18pm

" To make the playoff odds simulation a little bit more accurate, we've adjusted things by using weighted DVOA through Week 16, not Week 17, for four teams whose games were affected by sitting starters: Arizona, Green Bay, New Orleans, and San Diego. "

Wouldn't this have an effect on the Jets, too?

18
by Key19 :: Mon, 01/04/2010 - 11:31pm

Not really... the Jets didn't sit any starters. :) lol

Seriously though, I don't think the Jets' performance after Peyton left the game was really enough to skew their numbers. And they just beat the crap out a team playing 90% of their starters last night. They dominated Palmer and Ochocinco, and they're both starters in my opinion. :)

21
by Billy the Kid (not verified) :: Mon, 01/04/2010 - 11:45pm

Hey you know it just gets to the point where stats lie. Ya know? Ya know?! Ravens #1 team? Know what I'm saying? Ya know?!?!

But hey you guys make money doing this worthless number crunching that have never correctly predicted anything --- so kudos to you.

Get high every day.

25
by anotherpatsfan :: Tue, 01/05/2010 - 12:21am

spam filter Haiku
more enjoyable by far
fewer troll retards

27
by Anonymousnonwewwwf (not verified) :: Tue, 01/05/2010 - 12:53am

If you were high, dude, you would think about it hard dude. Maybe too hard. Then your brain would hurt.

40
by Tundrapaddy (not verified) :: Tue, 01/05/2010 - 4:20am

Three readings, and I'm still not sure if you're trying to be funny or not.

I don't think that says anything about me, so much as...about someone else.

Captcha: 'noxious because'. I couldn't have said it better.

46
by Anonymously (not verified) :: Tue, 01/05/2010 - 8:21am

I seem to remember Football Outsiders having the Ravens, Steelers, and Eagles as three of the best teams in the league last year going into the playoffs. Not sure which jerk-off commentator you follow but I doubt he did as good of a job predicting last season's AFC/NFC championship games as these *lying* statistics did.

23
by Red (not verified) :: Mon, 01/04/2010 - 11:58pm

Aaron - I know that removing sit-the-starters games doesn't improve DVOA correlations, but just for the sake of common sense, is there any chance you could post unofficial DVOA ratings with those games tossed out?

28
by Anonymousnonwewwwf (not verified) :: Tue, 01/05/2010 - 12:57am

I have to say that the pats are extremely dangerous, more so than people think. They have destroyed the teams the should have and had close games against those they should have as well.

29
by dk240t :: Tue, 01/05/2010 - 1:08am

Owen Daniels, in 8 games, put up 134 DYAR. Double it to extend to a full season, and that would be just above the 2nd place Dallas Clark's 261 for the season. Damn shame. 4th in DVOA, too.

Houston's DVOA dropped 0.8 percentage points, while weighted DVOA went up 2.2. Definitely caused by opponent adjustments. Cincy dropped big, Indy dropped big (x2), Jax and Ten dropped (x2), Ari dropped, only NYJ and Buf went up substantially.

30
by alexbond :: Tue, 01/05/2010 - 1:11am

Rumor mill says Mike Leach to Raiders. Discussion?

39
by Jim (not verified) :: Tue, 01/05/2010 - 3:05am

New Raiders coach: Mike Leach

New Raiders QB: Donovan McNabb

55
by Kevin from Philly :: Tue, 01/05/2010 - 9:35am

And JaMarcus in an electrical closet.

71
by Will Allen (not verified) :: Tue, 01/05/2010 - 12:23pm

"Put him in the electrical closet" may have supplanted "Put him in the cornfield" for football fans, even for those of us old enough to remember the original "Twilight Zone".

41
by Tundrapaddy (not verified) :: Tue, 01/05/2010 - 4:25am

I doubt it. I think Al Davis is too busy listening to his fat little girlfriend, Cable.

31
by Anonymouse (not verified) :: Tue, 01/05/2010 - 1:16am

I take it that we won't be getting Quick Reads until tomorrow, though? They are still coming, right? I'm eager to see Curtis Painter's line.

32
by jonnyblazin :: Tue, 01/05/2010 - 1:34am

One of the interesting things about the Ravens schedule is that almost all the difficult games were away, and most the patsies were home (discounting division games, obviously).

Away: SD, NE, MIN, GB, OAK - Point differential = -8
Home: KC, DEN, IND, DET, CHI - Point diff. = +94

This might explain why they had such feast or famine results. They mostly played really good teams or really crappy teams - overall that yields an average opponent DVOA. But when most the good teams are away that results in close losses, and when the crappy teams are home they blow them out. They didn't play any average teams out of conference except Denver (who was really bipolar anyways). Cincy might be considered an average team but the Ravens matched up with them horribly, thats 2 loses. They swept CLE and split with PIT, unsurprisingly.

I would say their offensive cryptonite is great CB play. If the opposing D can take out Mason with 1 CB, the offense basically sucks (see CIN, GB, MIN before Winfield injury, OAK). If the opposing D needs 2 DBs to shut Mason down, Flacco can usually move the ball down the field easily, passing to Mason when he's single covered or finding the open man when he's doubled.

34
by RickD :: Tue, 01/05/2010 - 2:10am

Unfortunately for the Ravens, they are a 6 seed. If you thought their regular season road schedule was tough, right now it would be @NE, @IND, and then probably @SD.

Maybe if Ray Lewis were 10 years younger I could believe they could pull this off.

47
by Anonymously (not verified) :: Tue, 01/05/2010 - 8:38am

There were quite a few atrocious calls against the Ravens this season. Anyone else remember cartoons of Tom Brady wearing a skirt circulating the internet following Ravens@NE? Ravens probably lost four games that involved *questionable* calls, some which involved multiple bad calls. Ravens probably shouldn't have called out the refs after that NE game. Bad strategy.

Ravens also got penalized on a lot of good calls. A LOT. If the Ravens averaged 5 penalties a game they might be 11-5 or 12-4. And if the Ravens stay at 5 penalties/game in the post-season I see no reason why they should be considered an underdog. They do have the best all-purpose back (of post-season teams) heading into some cold weather games.

48
by Spoon :: Tue, 01/05/2010 - 9:03am

If Baltimore wins at New England, they would next travel to the dome in Indy. A win there, and in all likelihood the AFC Championship game would be held in San Diego (with a high today of 72­°). This weekend is likely to be the only time we'll see weather be an issue in the AFC Playoffs.

83
by 2468ben :: Tue, 01/05/2010 - 12:52pm

Same going for the NFC with their Domers.

56
by Theo :: Tue, 01/05/2010 - 9:51am

Bang Cartoon

129
by cfn_ms :: Tue, 01/05/2010 - 10:24pm

Aaron talked on the Simmons podcast about how his numbers have generally indicated that the type of penalties that Baltimore collects tend not to correlate with winning or losing, so they're not particularly penalized for them. However, Baltimore gets a LOT of those penalties, so he actually said that they probably ought to be getting docked for that. Since there's not much of a gap between them and the next two, I'd guess that if DVOA adjusted more for those penalties, they'd be 2nd or 3rd.

I still suspect that the Ravens are overrated even at 3rd, but I'm nowhere near expert enough to be able to back that up with actual facts; certainly their inordinate number of very close losses indicate that they're better than 9-7 would indicate, but I really couldn't say just how much better. DVOA provides one such guess, but you could easily justify something different instead.

50
by ammek :: Tue, 01/05/2010 - 9:06am

One of the interesting things about the Ravens schedule is that almost all the difficult games were away, and most the patsies were home

Which is the total opposite of the #2-ranked Packers. Take out the Vikings and Steelers, and their road games were as follows:

Rams, Browns, Bucs, Lions, Bears, Cardinals' scrubs.

This is basically the ideal schedule for an above average team. Patsies on the road (you should beat them anywhere — though the Bucs' game didn't exactly follow the plan); better teams at home, where you give yourself the best possible chance.

I do think DVOA needs a home/road variable. I know that Aaron has looked, and hasn't yet found one that makes a significant difference. But one day perhaps………

66
by C (not verified) :: Tue, 01/05/2010 - 11:42am

Agreed that home/road matters.

I'll vouch for Green Bay, they really should have beaten the Bucs if not for an awful DPI call on AJ Hawk on a clutch I want to say 3rd or 4th down at the end of the game.

Aaron Rodgers is good enough to win on the road and that schedule was a joke. Betting on the GB season win total futures of 8.5 was a joke. I thought they could have realistically started the year 9-2 with that schedule ( with a split with Minny and beating Tampa).

33
by The Powers That Be :: Tue, 01/05/2010 - 2:05am

Please feel free to delete this if I'm revealing too much pay-only information from the Premium Database, but a couple things really surprised me when I looked at tonight's update:

1. Every playoff team (and I'm guessing every team in the league) had at least 3 games with negative overall DVOA...except the Cowboys, who had only one. I would never have guessed that. To be fair, the Colts would've matched them if it weren't for the Painter follies.

2. The Cowboys' first win over the Eagles graded out higher than yesterday's shutout.

49
by Paul R :: Tue, 01/05/2010 - 9:05am

Re: #2 Wow. I didn't see that. The Premium Database has all the interesting stuff.

BTW, did you see Tony Romo changed his phone number again? The new one is (214) 692-8[Post deleted by moderator. This information is available to Premium customers only.]

53
by Temo :: Tue, 01/05/2010 - 9:31am

And yet the Eagles grade out far worse in the shutout than in the first game. It doesn't seem to make a lot of sense, unless you consider that Offensive DVOA seems to have a far greater impact on overall DVOA than defense.

35
by Will Allen (not verified) :: Tue, 01/05/2010 - 2:25am

What causes Sidney Rice to suddenly sit up in bed at 4 A.M., screaming, in a cold sweat, is the dream where The Zombie King actually does retire, and Rice-a-Roni has to start running routes again for Tavaris, before Sidney gets his second contract signed.

36
by Roger Cossack (not verified) :: Tue, 01/05/2010 - 2:49am

You say "actually does retire" like he hasn't actually retired in the past.

42
by Tundrapaddy (not verified) :: Tue, 01/05/2010 - 4:27am

Good point - I think Will meant to say 'the Zombie King stays dead'.

65
by MCS :: Tue, 01/05/2010 - 11:39am

The American Heritage dictionary defines retire as "to withdraw from office, business, or active life, usually because of age: to retire at the age of sixty." Since the old man has yet to withdraw from active life, he has yet to retire. Saying you will retire and actually retiring are two different things.

149
by tuluse :: Wed, 01/06/2010 - 5:45pm

I guess the word "or" now means "and"?

Favre has definitely withdrawn from the office and from business several times, ergo he has retired.

37
by morganja :: Tue, 01/05/2010 - 2:57am

I'm betting that if refereeing was weighed in accurately it would explain most of the discrepancy between the Ravens DVOA and their results. It seems that Baltimore is the most victimized team when it comes to refereeing. I haven't seen them much, but every time I do it seems like the refs despise them.

43
by Tundrapaddy (not verified) :: Tue, 01/05/2010 - 4:31am

I'm not going to argue that it's impossible that the refs might not have a built-in Ravens bias by now. Having said that, their own coach, Harbaugh, pretty much stated that they get flags because they play 'aggressively' on defense.

So I'm not ready to call out the reffing just yet. Now, let's see how many times they get flagged for brushing Brady this time around.

82
by R O (not verified) :: Tue, 01/05/2010 - 12:48pm

The Ravens "anomoly" is pefectly clear. Flacco feasts on bad teams and sucks it up generally against good teams or in crunch time.

He (and they) will be better with experience, and you know, halfway decent receivers.

51
by ammek :: Tue, 01/05/2010 - 9:09am

Well the Packers were #1 in penalties, and #2 in penalty yards behind Baltimore. DVOA obviously doesn't mind.

38
by Anonymous Jones :: Tue, 01/05/2010 - 3:01am

Thanks again for another great regular season of coverage. I still always take this site's brilliant manipulation of stats with a grain of salt (mostly as a result of the information loss inherent in the data collection (not that you haven't pushed data collection to new heights...you have)), but it's difficult not to appreciate analysis showing that SD has the greatest differential between pass and run offense DVOA since at least 1993 and that Gates had four of the top 9 season DYARs since you started keeping track. And bias be damned (especially if most of the bias is relatively consistent year to year), it is especially remarkable that six of the top 12 passing DYARs happened this year. Even if I have problems with the defense adjustment, the *change* from year to year has to be telling us something about the NFL in 2009-10. Anyway, great stuff.

44
by bubqr :: Tue, 01/05/2010 - 5:37am

Too many Ricks and Anonymous on this website.
DVOA gives credit to teams blowing out lesser teams, meanwhile "public opinion" doesn't. "Public opinion" gives a lot of credit to the "big games" winners, and not to the losers even if the game was close, meanwhile it's the opposite for DVOA. All in all, no surprise if most people disagree with DVOA rankings.
Can we find Adjusted Games Lost to injury anywhere, or is it reserved for premium members or/and 2010 Almanac ?

54
by Temo :: Tue, 01/05/2010 - 9:33am

AGL not even available to Premium.

97
by Bill Barnwell :: Tue, 01/05/2010 - 1:50pm

AGL Isn't calculated until after the season, since we have to go through and mark who was a starter and who was a reserve, who won a job, etc, etc. So it's not done until March or so.

58
by C (not verified) :: Tue, 01/05/2010 - 10:14am

Culpepper should have started the season over Stafford all year long, I'm not just saying that now either, I said that before the season started.

It was an interesting night yesterday as an NFL team more or less had the opportunity to pick who they wanted to play. In reality they couldn’t guarantee a victory over the Jets, but they sure could guarantee a loss.

I reviewed the tape on the Giants blow out loss to Carolina and it was easy to see why when studying the front 7. Stewart ran all over the Giants and Jonathan Goff in his second start at MLB was why. He is known as a "hitter", and that's great and all but he wasn't mentally there yet. He took poor angles, couldn't avoid blocks and once he was engaged he couldn't shed blocks. If he wants to be an NFL LB I suggest hitting the film room and hard. Jeff King was also a very good blocker in that game at TE.

151
by tuluse :: Wed, 01/06/2010 - 5:47pm

From what I saw of the Lions Stafford was outplaying Culpepper by a significant margin early in the year. Besides what would the Lions have gained by playing Culpepper? They weren't going to make the playoffs this year, they'll be lucky to next year. By the time they do Culpepper will be 35.

176
by Spielman :: Sun, 01/10/2010 - 11:30am

Exactly. Culpepper's corpse isn't the solution to anything for the Lions. Getting Matthew Stafford some experience may be.

59
by DeltaWhiskey :: Tue, 01/05/2010 - 10:23am

In Week 8, I explored the possibility that the regression equation associated with DVOA’s correlation to wins might be a tool for predicting final records.

The formula for these projections was the following formula: Projected Wins = (% of games left on schedule) X (projected season wins based on Team DVOA adjusted for strength of FUT Schedule) + current Win Total.

Below, are the final results as reported at the time with the addition of final wins and the difference between projected and actual. The first number is projected wins based on the above formula, the second is the actual wins, the third is the difference.

(Team, Proj Win, Actual Wins, Dif)
NO 15 13 -2
IND 14 14 0
MIN 14 12 -2
DEN 13 8 -5
NE 12 10 -2
PHI 12 11 -1
BAL 12 9 -3
GB 11 11 0
CIN 11 10 -1
ARI 11 10 -1
DAL 11 11 0
PIT 10 9 -1
ATL 10 9 -1
NYG 9 8 -1
HOU 9 9 0
MIA 9 9 0
SD 8 13 5
NYJ 8 9 1
SF 8 8 0
JAC 7 7 0
SEA 7 5 -2
CHI 7 7 0
BUF 5 6 1
CAR 5 8 3
WAS 4 4 0
KC 3 4 1
TEN 3 8 5
CLE 3 5 2
OAK 2 5 3
STL 2 1 -1
DET 1 2 1
TB 0 3 3

Observations and Comments:
- 62.5% of the predictions were within +/- 1 (one) game of the final record
- 28.1% were dead on
- The mean difference was .063 with a SD of 2.13. Three teams under/over performed their projection by a significant margin, SD, TEN, and DEN.

Utilizing the teams winning percentage in Week 8 as a predictor (i.e. multiplying the win% x 16) produced the following projections:

(Team, Proj Win, Actual Wins, Dif)
NO 13 16 -3
IND 14 16 -2
MIN 12 14 -2
DEN 8 14 -6
NE 10 11 -1
PHI 11 11 0
BAL 9 9 0
GB 11 9 2
CIN 10 11 -1
ARI 10 9 1
DAL 11 11 0
PIT 9 11 -2
ATL 9 9 0
NYG 8 10 -2
HOU 9 10 -1
MIA 9 7 2
SD 13 9 4
NYJ 9 8 1
SF 8 7 1
JAC 7 7 0
SEA 5 5 0
CHI 7 9 -2
BUF 6 6 0
CAR 8 7 1
WAS 4 5 -1
KC 4 2 2
TEN 8 2 6
CLE 5 2 3
OAK 5 4 1
STL 1 2 -1
DET 2 2 0
TB 3 0 3

Observations and Comments:
- 62.5% of the predictions were within +/- 1 (one) game of the final record
- 25% were dead on
- The mean difference was .027 with a SD of 2.22. Again, three teams under/over performed their projection by a significant margin, SD, TEN, and DEN.
- In Week 8, DVOA adjusted for future schedule DVOA predicts final season wins as well as the teams current winning percentage. I’ve conducted a similar study using past seasons that yields similar results.

80
by metro (not verified) :: Tue, 01/05/2010 - 12:44pm

Minnesota was 13-3 (losses to Carolina, Chicago and Pittsburgh), not 14-2.

86
by Eddo :: Tue, 01/05/2010 - 12:55pm

In fact, neither list has Minnesota's actual wins correct. The first has proj=14, actual=12, the second has proj=12, actual=14. So both lists are off by 1, not 2, on the Vikings.

133
by DeltaWhiskey :: Wed, 01/06/2010 - 3:08am

Minnesota was 12-4, projected wins via DVOA was 14, for a -2 difference.

For projections using win % I screwed up column order, my apologies to all. Should be (Team, Actual Wins, Proj Win, Dif).

99
by countertorque :: Tue, 01/05/2010 - 2:05pm

So, why bother with all the math necessary to do it with DVOA? I can just use current win percentage and do just as well.

134
by DeltaWhiskey :: Wed, 01/06/2010 - 7:03am

For trying to predict future performance, that is the conclusion I'm coming to also.

At least by Week 8, DVOA as a predictor for the Season is no more effective or useful than a team's win percentage. The corrolary to this finding is that the argument that Wins don't tell you how good a team is because they don't account for the strength of the opponent appears to be in jeopardy.

So now the question is, what is DVOA's value/utility/worth? What does DVOA tell us?

My thoughts:

- The correlation of DVOA and Wins is quite robust, and DVOA correlates more highly with Wins than conventional measures of performance (e.g. offensive yards/game). Therefore, DVOA more effectively and/or efficiently measures a team's performance on the field, and the components (OFF DVOA, DEF DVOA, and ST DVOA)may be a useful tool to describe what teams are doing/not doing to win/lose.
- DVOA is not an intuitive measure. The meaning of the difference between DAL 26.0% and NO 23.4% DVOA is not immediatlely clear. The regression equation for the correlation between DVOA and WINS suggests Dallas played like a team that typically wins 11.5 games and NO played like a team that typically wins 11.2 games. Roughly, a change of 7.25% DVOA represents one Win.
- DVOA accounts for about 73% of the variance in Wins. VOA (correlation with Wins - r = 0.883) accounts for 78% of the variance in wins. In 2004, following the update to VOA, the correlation of VOA to Wins was r = 0.861, meaning the 2004 VOA update accounted for 74% of the variance in Wins ( http://www.footballoutsiders.com/dvoa-ratings/2004/dvoa-gets-another-upg... ). A question that remains is have the changes to VOA resulted in statistically signficant chanbes to the R-square? Put another way, yes VOA (and DVOA) appear to be improving, but is this improvement a statistically significant improvement, or is it likely due to chance. Moreover, how much more room for improvement is there in VOA (DVOA)? Being able to explain almost 80% of the variance is impressive, nonetheless, it's use as a predictor appears limited as shown above. I would suggest that there are intangibles (random factors or luck, coaching factors, psychological factors, etc.) that VOA cannot capture and are difficult to quantify and capture in any case.

145
by mrh :: Wed, 01/06/2010 - 2:39pm

Picking any particular week can be arbitrary. I arbitrarily picked week 7 the other day and calculated the correlations between the winning percentages of teams in weeks 8-17 (including the Jets-Colts etc. outcomes) and the following stats:

winning pct weeks 1-7: 0.40
dvoa thru week 7: 0.45
dave thru week 7: 0.50
off dvoa thru week 7: 0.49
def dvoa thru week 7: -0.27
st dvoa thru week 7: 0.00
voa throu week 7: 0.44
est wins thru week 7: 0.46

I'm not sure any of the differences are statistically significant. But when you look at an arbitrary week, it may be that dvoa looks worse than winning percentage or better. And while the pre-season predictions look bad in many instances, it is interesting that Week 7 DAVE did a little better than Week 7 DVOA in "predicting" future winning percentage.

The hardest thing, I think, is spotting major changes before they occur. There were five significant changes in winning percentage between week 7 and week 17:

TEN from 0-6 to 8-2 (FOA pre-season mean projection 9.3)
SD from 3-3 to 10-0 (12.5)
PIT from 5-2 to 4-5 (9.6)
NYG from 5-2 to 3-6 (10.0)
DEN from 6-0 to 2-8 (4.9)

You could include TB (0-7 to 3-6) and CLE (1-6 to 4-5) if you like but I'm arbitrarily excluding those as they did not move in or out of the playoff picture based on what they did weeks 8-17.

Looked at in week 7, the FOA pre-season projections for TEN, SD, and DEN seemed far off the mark, and the NYG and PIT projections seemed to underestimate those teams. But now, those projections sseem pretty good. The direction and magnitude of those major changes were all "predicted" by the preseason projection in the sense that the teams regressed to their projected mean; I doubt if many analysts got all five major changes right.

I agree with a lot of the criticisms of DVOA in this thread, and the pre-season predictions were off the mark in a lot of ways. Even giving them credit for the DEN prediction is iffy since they were 3+ wins off. But I think big picture the FOA predictions on those five teams look a lot better now than they did in Week 6.

159
by DeltaWhiskey :: Thu, 01/07/2010 - 6:11am

Agreed that picking the week is arbitrary. Weeks 7, 8, or 9 seem pretty good, b/c they are close to the the 1/2 way point of the season.

I appreciate your approach to examining the issue and agree that the most interesting part of the analysis is identifying what happened with the three to five teams that over/under performed. Equally remarkable, is the degree to which most teams finish out the way they've started.

I agree that the the most interesting results are those produced by DEN, SD and TEN as well, either because of there apparent overachieving/improving or there underachieving/declining in the latter part of the season. B/C my projections had PIT as 10 or 11 win team, their finishing with 9 wins isn't too far off, although, what is interesting is how PIT went about getting to only 9 wins. My projections had the NYG as either 9 or 10 wins, which again is pretty close.

Variance may describe the performances by NYG and TEN, but SD, PIT, and DEN were all average in the variance of their performance, so the question is what factors contributed to their final performance.

135
by greybeard :: Wed, 01/06/2010 - 7:25am

That would not be complicated, or "outsider" and people living by it would not feel any smarter than anybody else.
It would be all too banal and would not make people talking about DVOA or crafting templates about the ridiculousness of critiquing DVOA feel superior.

I believe FO's estimated wins in its best year has .31 correlation to the wins of the next year. Which is just slightly better than using previous years records or just guessing every team will go 8-8.
.31 is not significant correlation anyway.

Of course people will defend it as better than everything out there. But I don't know any other statistics that is predictive (or any other methodology of prediction) out there so I don't know what people are comparing DVOA against. BTW, I don't consider Vegas lines as a competitor as they are prediction which numbers will bring the largest amount of money to gambling. They are predicting the perception of gambling community.

136
by C (not verified) :: Wed, 01/06/2010 - 10:00am

It's not about bringing the most money to "gambling". It's about getting (close to) equal action on both sides. It's a market...

If I put out a line of New England +10 this week, I'd get a whole lot of action, but I'd probably lose a lot of money ( or have a lesser chance to win a lot of money).

Some people are good at computing the information in their heads and making correct predictions over and over and over again.

142
by Jeff Fogle :: Wed, 01/06/2010 - 2:06pm

Only thing I'd add about the markets is that there's a difference between the opening line (estimate of what the gambling world will think) and widely available lines after the money has hit (a reflection of what the gambling world actually thinks in composite).

If DVOA has quality predictive value, it would beat the opening lines like a drum (because they're estimates of people's thoughts rather than an attempt to reflect reality), and they would still win vs. the settled market prices because DVOA is better at capturing reality than gamblers. I think it's a fair way to evaluate ANY tool, including all the mainstream stuff. Don't want to suggest DVOA should be held to a higher standard. Everything should be held to the highest standard.

I guess an ultra-critic might say that beating the lines isn't good enough because gamblers aren't as smart as the best research tools. But, if the research tools aren't even measuring up to the current market...

60
by C (not verified) :: Tue, 01/05/2010 - 10:25am

I heard on the local radio that Washington is 99% likely to announce Mike Shanahan has the new coach and that it could happen as early as today.

I have NFL futures on DVOA rankins 2-4.

Ranked too low IMO ( the #1,#2 seeds in the AFC ranked 8 & 11?)
San Diego
Indy ( nearly unbeaten and ranked 8th in DVOA?)

Ranked too high
Baltimore
New England

Teams that didn't make the playoffs this year that could make noice next year, Houston, Atlanta, Titans, Miami, Carolina, Giants/Steelers to bounce back. I think when the season ends, I'll look at NFL futures for these teams and the Green Bay Packers at a reasonable price. If a couple games went the other way, this team would have a first round bye and be in the driver seat to go to the SB. I predicted a NE/GB SB at the start of the year and I'm inclined to write in GB for next year. Aaron Rodgers is in his 2nd year as a starter and is already the 2nd best QB in the NFC and has Drew Brees potential. Yeah, I'm usually bashing quarterbacks telling you their problems, but Aaron Rodgers is fricken awesome.

101
by Rick A. (not verified) :: Tue, 01/05/2010 - 2:37pm

You certainly are seeing it, in my opinion, but you'd better proceed with caution in all that "futures" betting. Next season the owners, and their employee the commish, have some serious surprises in store for the players and public. Everything that appears to be the wave of the future will likely take serious turns. But if all that wasn't on deck then what you say would likely be it--with the exception of a Giant bounce-back. They have deep problems that likely only free agency can fix, and like I just pointed out, free agency, as it has been, is about to be mutated.

As to what you are sort of forecasting for the playoffs-obviously S.D. and Indy are ranked too low in this DVOA system and Balt/N.E. too high. Personally, I think both Balt and N.E. will get knocked out although the QB in Balt is in his second appearance and may do well. And their RB's are serious. The Ravens are very, very dangerous. Somebody earlier in this thread gave a pretty good breakdown of Balt although they failed to mention what tremendous play they get from the running backs. On the NFC side I expect Minny to host the Championship game against ?? but I look for N.O., Philly and probably G.B. to get knocked out. To me, there are just too many flaws, and injuries, on G.B. and I expect Warner will exploit them Sunday. But you know Rodgers will be givings his guts, and what a tremendous player he is indeed...

109
by C (not verified) :: Tue, 01/05/2010 - 3:44pm

Yeah my friend's friend is an agent and he's talking about that dark cloud that hangs over the season as well... Last year a Cincy at 80/1 led by Carson Palmer and Marvin Lewis on the hot seat looked too good to be true, but I watched the making over the team over the summer and actually LOST confidence in the Bengals.

I disagree on the Giants. They should be strong on the offensive and defensive lines. That's with an expectation of letting Fred Robbins walk, and resigning Barry Coefield. The past two games they got horrible horrible play out of their 3rd string MLB Jonathan Goff and unless he improves, I don't see them getting on the field as a starter again. The obvious weak link is at safety, and Kenny Phillips should be back, and Reese will most likely bring somebody anybody better than Brown and Rouse back there.

The Giants need a 3rd RB, Brown could heal and be that guy or they can use a mid level pick or later for that guy. The WR's might have had decent stats this year, but they were very raw. Mario Manningham shows flashes but he does so many little things wrong it isn't even funny. I'd suggest jumping into Cris Carter's WR camp in the off season and learning the nuances of the game. If the Giants have a QB, the WR play should get better next year, they are strong on the lines... They made the playoffs the last 4 years in a row ( and won a SB) in a strong NFCE... Plus they have a GM that drafts well...

If there was ever a script in the NFL, Dallas beats the Eagles ( again), and Green Bay beats Arizona... That would put Dallas @ New Orleans rematch in round 2, and Aaron Rodgers vs Brett Favre ( yes it's a 1 on 1 game not 52 v 52).

Right now I'm leaning on all 3 road teams though with the exception of New England...
Jets, Pack, Eggos, and Patriots...

On a neutral field I think GB is the best team in the NFC, but the fact that they'd have to win 3 road games I'm not so sure they make it to the bowl.

I wouldn't count the Pats out of the AFC just yet... Brady, Bellicheck... and look at the other coaches in the AFC. The Patriots have a huge advantage in game planning/head coaching...

Caldwell = rookie head coach
Norv = He will be blamed if the team doesn't go to the SB (nobody likes him)
Harbaugh = Not sold on him at all yet
Ryan = Rookie head coach with a big mouth
Lewis = Not a good head coach
Bellicheck = Easily the best coach in the league.

119
by t.d. :: Tue, 01/05/2010 - 6:14pm

if Dallas wins, they play Minnesota. If Philadelphia wins, they play New Orleans. The opponent for the GB-Arizona winner is contingent on the result in Dallas.

126
by Rick A. (not verified) :: Tue, 01/05/2010 - 8:47pm

The G Men have yet to find an equal for Burris and I don't see them finding one to draft. Opposing coordinators do not respect their air game that much. They just haven't come up with a deep threat and so the box gets loaded. Their Bigs didn't do so well, and big Jacobs was pretty much a bust, as incredible as that was. The guy is a behemoth but appears quite reticent when he's not getting the holes opened nice and wide. But I like the Giants, their organization, their ownership. Pretty good people. I hope they bounce back. It's good for football when the G Men are winning...

62
by Led :: Tue, 01/05/2010 - 11:31am

Question for Aaron, et al.: How does DVOA measure fake punts and FGs? Are they counted as special teams plays or as pass/run plays? I'm guessing the latter.

67
by maestro876 :: Tue, 01/05/2010 - 11:53am

Given the split in the Patriots' record at home and away, I'd be very interested to see if there was a significant difference between their home DVOA and road DVOA. Anyone know?

68
by Boots Day (not verified) :: Tue, 01/05/2010 - 11:56am

I'd love to see a little more analysis of the Broncos' season. It's not just that they started 6-0 and finished 2-8 - they beat four playoff teams in that opening run, and lost to three teams with 11 or more losses in the closing run. Man, what a collapse.

114
by deflated (not verified) :: Tue, 01/05/2010 - 5:22pm

I'd be willing to go with game tape for the new Nolan defence on why it went kaboom.

Once teams had enough tape on Denver's tendencies they started picking it apart. KC, Oakland had excellent rushing numbers, Washington had good rushing numbers and very good passing yards. For me I can't consider a defence that gave up 241 yards and 317 yards on the ground to the Chiefs and Raiders to be a top 10 unit (or even close) - even a Rush D rank of 18 seems high.

If you had tracked Denver's rush Def. DVOA over the year it would be an ugly graph.

74
by LinksterAC (not verified) :: Tue, 01/05/2010 - 12:30pm

DVOA had a bad year.

75
by TomC :: Tue, 01/05/2010 - 12:36pm

Bears have fired Ron Turner. That pretty much clinches it that Lovie keeps his job another year.

edit: Marinelli expected to be named DC, also.

88
by dedkrikit (not verified) :: Tue, 01/05/2010 - 12:57pm

The '94 Oilers really did miss Moon.

90
by Aaron Schatz :: Tue, 01/05/2010 - 1:11pm

Quick note: Premium now has head-to-head matchup pages for all four wild card games.

103
by Jeff Fogle :: Tue, 01/05/2010 - 2:44pm

Aaron,

You said at the top it was a strange year for statistics. I have to ask, was it a strange year for total yardage, yardage differential, third down conversions, passer rating, TD-INT ratio, turnover differential, and any other standard stat? It doesn't seem like a strange year at all in statistics. As you've documented through DVOA, and which can be documented with the mainstream stats, the emphasis on passing continues to trend upward. Otherwise, WHAT'S STRANGE ABOUT NFL STATS? The same stuff is showing the same things.

I hope you'll consider the possibility that DVOA accurately captured the realities of football at the time you put the pieces together a few years ago...but that the game has evolved in a way that isn't as well captured now. The FO toolbox may have had "predictive" value compared to other stats back in 2005, which is a year I see referenced often. I think it's tough to make that case now. The Green and Yellow records from the Premium service posted over in the Discussion boards don't suggest predictive value. The differences between preseason win projections and the wagering markets in Las Vegas were stark. I ran that listing a couple of weeks ago in a comment. With the season over, the final mark ended up at 5-16.

(Listed by division because that's how I had them logged)
Miami (6.5) Under 7/7.5 (W)
NY Jets(6.5) Under 7/7.5 (L)
Buffalo (5.5) Under 7.5/8 (W)

Indy (11) Over 10/10.5 (W)
Jax (10) Over 8/8.5 (L)
Houston (6.5) Under 8.5 (L)

(No qualifiers in North)

San Diego (11.5) Over 9.5 (W)
Kansas City (7.5) Over 6/6.5 (L)
Oakland (6.5) Over 5.5/6 (L)
Denver (5) Under 6.5/7 (L)

Dallas (8) Under 9/9.5 (L)
Washington (7.5) Under 8/8.5 (W)

New Orleans (7.5) Under 8.5/9 (L)
Atlanta (6.5) Under 8.5 (L)

Chicago (10.5) Over 8.5 (L)
Green Bay (7) Under 8.5 (L)
Detroit (6.5) Over 4.5 (L)

Seattle (9.5) Over 7.5/8 (L)
St. Louis (8.5) Over 5 (L)
Arizona (5.5) Under 8.5 (L)
San Francisco (5.5) Under 7/7.5 (L)

Final Record: 5 wins-16 losses, even allowing for "shopping" for the best line when both 7.5 and 8 were available for example.

The college season isn't over yet, but the FO toolbox didn't seem to be having a banner year for the guys outlining expectations over there either. Even if you exclude Mr. Weintraub because he wasn't using an efficiency system, there wasn't much of a "there" there vs. market expectations. The FO toolbox is behind the market in both pro and college football by a good bit this year.

I appreciate that you guys have such an open forum so people can discuss the possible strengths and weaknesses of the approach. I hope you'll consider the possibility that pro football has evolved past what DVOA was apparently capturing at a team level when first devised (creating the impression of a "strange" year in stats when it's hard to find many strange things in the mainstream stats). Maybe it zoomed in too closely on the microscope...and backing up to bigger picture stuff (like TD-INT ratios and third down conversions) will help realign the lens. Just throwing out ideas. It just doesn't seem like a strange year in stats, or in terms of who made the playoffs and who struggled to win many games.

106
by Rick A. (not verified) :: Tue, 01/05/2010 - 3:24pm

Exactly. Although for someone like me, calling what the owners have orchestrated is not an "evolution" of the game. It's just business, as they see it. As the game grows closer to Arena Football the more ridiculous the attempt to quantify this and that element of it...

110
by theshadowj :: Tue, 01/05/2010 - 4:37pm

Exactly. Arena football is so evanescent it is completely unquantifiable.

115
by Aaron Schatz :: Tue, 01/05/2010 - 5:25pm

Re: comment 103.

Our preseason projections absolutely had a terrible year, no question about it. I do hope, however, that people understand the difference between "preseason projections" and "in-season statistics." Your post is entirely about the former. Also, I think the point about all the teams between 11 and 12 Pythagorean wins does a good job of demonstrating that this was a strange year using conventional stats as well. There is no more "conventional" stat than points scored and allowed.

116
by c (not verified) :: Tue, 01/05/2010 - 5:50pm

and why Las Vegas keeps building casinos in the desert

Conventional picks this year weren't that bad this year, especially in the AFC...
E. Philly
N. Minny
S. Atlanta
W. Arizona
WC. NYG/Car
WC2 Dal/GB/SF/Sea/NO

AFC
E. NE
N. Pitt
S. Indy
W. SD
safer WCs. Bal, Tenn
riskier WC calls on Mia, Hou, Jets

Predicting the future is hard. If you are going to hype up your winners, you are also going to have to talk about the picks that were horribly wrong as well. This year the news stand mags could have got by pretty easily since they tend to copy the same picks as the year before and the "conventional wisdom". FO will often jump out on a limb more so than most and this year wasn't a good year to do that. I wonder how much swagger they will write next years predictions with.

118
by Jeff Fogle :: Tue, 01/05/2010 - 6:13pm

Re: comment 115

I hope you'll reconsider the word "entirely" from the sentence "Your post is entirely about the former" given all the references to stuff that wasn't about preseason projections. The yellow and green rated games from the premium package were during the season. The college football stuff is during the season. I didn't post the records because I don't want it to seem like I'm piling on. Those records are posted elsewhere at the site.

And, if it wasn't clear, the mainstream statistical categories I listed meant YEAR END stats in those categories. Now that the year is over, it wasn't a strange year in those stats. This article is the Week 17 article. You said it was a strange year in stats. I listed a bunch of stat categories where not much strange happened.

Scoring differential, whether you use the basic totals listed in all the web standings, or you square the heck out of them to make them Pythagorean (which is known to statheads but far from conventional) shows a bunching of good teams in the upper half of the league, and a bunching of bad teams in the lower half of the league. What's strange about that? The fact that there's bunching rather than more of a classic bell curve may be "strange." The stats pointing it out aren't strange.

You said at the top it was a strange year in stats. Your post just above says "Also, I think the point about all the teams between 11 and 12 Pythagorean wins does a good job of demonstrating that this was a strange year using conventional stats as well."

That's a different thing. It was a strange year regarding the bell curve (though I haven't actually looked that up, I'll assume you're right). You can use conventional stats to show it was a strange year. What was strange about the stats? Everything correlated to success and failure the way it normally does in the mainstream world. But, DVOA has a 9-7 team listed as #1 in the league, and has Wildcard teams that went 0-6 against their division winners listed as the best three teams in the league. I think we all agree that's a bit strange at the 17-week mark.

If DVOA has predictive value, we would assume it's an important part of the green and yellow rated games on the premium service. Those didn't do well. That's a continuation of the preseason woes.

The Week 17 rankings may well foreshadow what's about to unfold. I think it's tough to make the case that "2009 was a very strange NFL season, at least when it comes to statistics."

122
by Aaron Schatz :: Tue, 01/05/2010 - 7:06pm

There's no point in getting in a huge back-and-forth. I happen to think that FO has generally been good about saying "yes, we're not perfect" when we get things wrong. Some readers disagree. That's life. To satisfy our more cantankerous readers, I'll say this as plainly as possible: There's no doubt 2009 was our worst year for preseason projections, and we owe better to the people who do spend money on the book or the Premium picks. That being said, it was just one year. That's not the biggest statistical sample.

Three notes on the picks against the spread, which will be my final comments in this discussion:

1) Yes, I'm frustrated with the fact that the "red" picks actually did better than the "green" and "yellow" picks. You better believe I'm going to spend some time in the offseason trying to improve the way I identify the best picks of the week.

2) The problem with picks was, in fact, the problem with preseason projections. In the first few weeks, when our judgment of teams was based in part on preseason projections, the picks had a losing record. From Week 10 onwards, when picks only took into account actual 2009 in-season performance, the picks were 66-54-7, a record commensurate with their record in 2008.

3) The college stats are a work in progress, and we've never said anything differently. Don't ascribe vanity to those writers when there has been none.

123
by C (not verified) :: Tue, 01/05/2010 - 7:59pm

Modest when you guys aren't right... but what about saying you guys "make the best predictions in the NFL" after a decent year?

There are people who make picks without DVOA and make money every year picking ATS... Capping the NFL isn't even always about picking, " the best team", sometimes teams are in hard situations that are predictable.

131
by Dales :: Tue, 01/05/2010 - 10:57pm

C,

You are busting on them for saying that they 'make the best predictions in the NFL' after years where they could make a very good argument that they did?

It would have been strange marketing to say "we've done good so far, knock on wood! Hope it lasts!"

I think the number of people who make money every year picking ATS is rather small. If, instead, it is the number of people who do well most every year, well, that would be a bigger number and I think FO would qualify.

But, honestly, is that really the point? Have you learned anything at all from their analysis? I have. I've learned quite a bit.

Do I agree with the general FO staff opinions at times? No. Do I think there can and should be improvement made to the system? Yes, even if I don't necessarily know how. I know the Ravens are not the best team this year, nor do I think they are particularly close to being the best team this year. Do I think they sometimes dig in when they should not? Sometimes.

I learn from these guys, I think they argue in good faith, and they are entertaining (at least most of them). If you don't have the same experience, then why do you bother coming back?

138
by C (not verified) :: Wed, 01/06/2010 - 10:15am

Yeah but at the time they were NOT the best Forecasters out there so it was rather annoying.

Yes, I like their stuff, they bring a different perspective and they don't fall victim to a lot of the football myths out there and there are some real good POSTERS on this site as well.

I like looking at the DVOA ( to see if I'm missing anything) but some of the Outsiders individual "scouting" type comments really aren't that impressive. I'll look at their numbers, read the posters, and enjoy it. My complaints is the occasional smugness to readers and hyping up winners and avoiding talking about losers can get a little annoying.

The hard part about using DVOA as predictive is that what a team DOES isn't neccesarily what a team could do... Sometimes teams under perform, sometimes they are playing about as good as they can, sometimes guys get injured and throw everything off, sometimes teams quit, and yes, some teams play a lot better at home than on the road, and some teams play about the same...

If you are asking, yes, I like this website, I've been a poster here for probably 5 or so years and I vote with my feet. If you want to compare CBS Sportsline to stat geeks, I pick the stat geeks but I do like quality gambling forumns as well.

128
by Jeff Fogle :: Tue, 01/05/2010 - 9:40pm

Understand completely why a big back-and-forth is inappropriate. You can't do that with everyone who disagrees about anything. I'll try not to exploit that in this discussion. If others have records they can post from past seasons so readers can have a bigger sample size to draw conclusions from, that would be greatly appreciated.

*It's true it's just one year. Does anyone have records from previous years against market prices week-to-week and against regular season win odds from the annuals? I've read enough to know some readers think past editions were poor, while others think they were excellent. Does anyone have the records against market prices?

*Looks like you're counting the red picks in that Week 10 onwards record. This is what ThePop said over in the betting discussion forum here at the site about what the guidelines are for the picks:

GREEN: High arbitrage. For the occasional bettor.
YELLOW: Reasonable arbitrage. For the avid bettor.
RED: Avoid unless you are in a picks pool that requires choosing a winner in all games.

I think it's borderline to include the "avoid" picks in a discussion about the predictive value of DVOA and the FO toolbox. Does anyone have the final records in Green and Yellow since Week 10? ThePop hasn't posted them yet.

PS: What was posted showed a 1-7 record in the 10th week of Green and Yellow selections. Am I right that this was at the 95% point of where this year's data stood in the mix? Odd to pick right AFTER this week as the best way to measure performance. Gradual difference between 95% and 100%.

*Did the selections show the same tendency last year? Where things really "blasted off" once that season was 100% represented in the data set? If so, that's something to learn from. If not, it's even dicier using what happened this year as an endorsement. Sample size works both ways.

*Brian Fremeau has been very open about discussing his ratings. Don't want to sound like I'm criticizing him. Rating teams accurately is a very tough nut to crack from whatever perspective you're using. That being said, it's hard not to ascribe vanity to a guy who named an index after himself.

Again, I hope you guys will consider the possibility that football is evolving away from what DVOA apparently captured to your satisfaction back in 2005 and earlier. It's true that we have a small sample size to judge from. The earliest stages after an evolutionary tick represent a small sample size...but then the changes continue as evolution continues.

I understand that no ratings system will please everyone. Using comparisons to market prices is a great way to get that out of the way and just focus on what the data is saying. Performance vs. the market tells the story. Then you have something to show the cantankerous people.

Aaron is done, as he mentioned. If anyone has data from past years that will increase the sample size to help readers draw meaningful conclusions, I hope they'll post them.

127
by Rick A. (not verified) :: Tue, 01/05/2010 - 9:04pm

You have said it all. There is something blatantly wrong with a system that purports a 9-7 team as best in the league. No matter how dangerous they may be in the coming playoffs the fact is they compiled their wins against some weak competition. This system has also proclaimed teams that went a combined 0-6 vs their division winners as the best teams. My point as an observer this year has been very simple and I've been roundly attacked for it--this system tries to measure competitions that were just BS, non-games, like the Tenn/NE debacle where one team had the "blue flu", blue as in Titan blue. And so many of these teams, at this point, have been put together by ownership groups that couldn't care less about fielding a winning team that it makes the whole process of trying to quantify results of the games involving those teams as ridiculous as it can be...

124
by Dan :: Tue, 01/05/2010 - 8:33pm

The problem with the preseason projections was in the defensive projections. Projected offensive DVOA correlated with actual offensive DVOA with r = .55 (.58 if you exclude wk17), special teams correlated at r = .42 (or .44 excluding wk17), but defense didn't correlate at all (r = -.02), so overall DVOA correlated at only r = .16 (.20 excluding wk17).

139
by Bowl Game Anomaly :: Wed, 01/06/2010 - 1:46pm

That is an interesting and potentially instructive comment. Maybe the key to improving the projections is to focus on defense. (Or, maybe there were just a lot of teams with fluky and unpredictable changes to their defensive performance from last year to this year.)

144
by Jeff Fogle :: Wed, 01/06/2010 - 2:12pm

How did "actual" defensive DVOA from 2008 correlate with "actual" defensive DVOA from 2009? using those measures?

146
by Bowl Game Anomaly :: Wed, 01/06/2010 - 3:30pm

You can calculate that just as easily as I can, so I'll let you do it. I'm not sure if it's going to teach us anything though. FO has demonstrated pretty definitively that defense is a lot more variable year-to-year than offense is.

150
by Jeff Fogle :: Wed, 01/06/2010 - 5:45pm

I promise...if I tried to do that it wouldn't come easily (lol). Just wondering if there was a lot of dramatic turnarounds on that side of the ball from last year to this or not. Will try to eyeball it later today to see what it shows...then just plain old total yardage defense too. Would help add context to the discussion if it was just a bizarre year in defensive turnarounds...

155
by randplaty (not verified) :: Wed, 01/06/2010 - 10:10pm

Great discussion. Appreciate FO allowing an open and frank discussion about DVOA here.

156
by Bowl Game Anomaly :: Thu, 01/07/2010 - 12:52am

Ugh. Please don't look at yardage. If you must compare DVOA to a traditional stat, at least use points allowed. Or, if you must use yards, try yards per play, not yards per game. Anything but yards per game.

157
by DeltaWhiskey :: Thu, 01/07/2010 - 3:47am

Correlation of DVOA with:

PTS 0.776
YDS 0.800
Y/P 0.828
1stD 0.741

DVOA is most similar to YDS/Play.

158
by tuluse :: Thu, 01/07/2010 - 5:16am

Probably because DVOA is very similar to yards per play, just adjusted for circumstance.

PTS/drive would be interesting to look at.

160
by Jeff Fogle :: Thu, 01/07/2010 - 3:50pm

Wasn't looking to compare DVOA to anything. Was looking to see what evidence there was that an abnormal year in "defensive turnarounds" or whatever you want to call it could have been the monkey wrench that was wreaking havoc on projections...and wanted to use a variety of tools to see if that was true.

Definitely looks to be out of character from at least the prior season. That can be expressed in a variety of ways. Using something simple like average and median changes in defensive DVOA ranking shows an average change of 6.4 spots going from '07 to '08, with a median of 4 spots. Going from '08 to '09 showed an average of 7.7 spots and a median of 7.

Most extreme examples from this year:
Denver advancing 24 spots from 31st to 7th
Tennessee dropping 22 spots from 5th to 27th
Tampa Bay dropping 19 spots from 6th to 25th
Chicago dropping 15 spots from 7th to 22nd
San Francisco advancing 14 spots from 18th to 4th
Buffalo advancing 14 spots from 22nd to 8th
NY Jets advancing 13 spots from 14th to 1st
Green Bay advancing 10 spots from 12th to 2nd

Complicating matters is something that the FO guys have mentioned a few times this year if I recall. Defenses are contracting toward the middle with fewer extremes, at the same time they're getting worse compared to past performances (offense is trending up, defense is down...which I'd submit is likely tied to relaxing the holding rules awhile back--and more offense equals bigger fantasy stats(!), lol)

If things are contracting, it makes it easier for extremes to happen. You're leapfrogging a tighter pack (or plummetting off the cliff)

Quick evidence:
*Only three teams were outside 4.8 to 5.8 in defensive yards-per-play this year. There were eight last year.

*DVOA shows worse rankings this year in spots 1-26 on the ladder, out of 32 teams (26 of 32 spots are worse), but spots 27-32 were actually worse last year (meaning the worst defenses are condensing toward the middle over time).

A bit confusing, but both of those things are happening at once. Defense is getting worse, and bunching together.

Bottom line in terms of the rankings changes, if you improve, you can make a very big jump in the rankings. If you fall, apart, you sink a long way. And, projection systems that don't properly anticipate changes could get out of alignment quickly.

Noticed something interesting with the teams making changes. Could easily be "effect" rather than cause. But, defensive time of possession changed with most of them.

Denver's defense was on the field 1:10 less this year (30:06 to 31:16), corresponding with its improvement. Chicago's defense was on the field 42 seconds more per game (32:26 to 31:45), corresponding with its demise. Is that tied to Cutler's inability to run clock (he succeeds too quickly, fails too often, and it wears down his own defense)?. Just theorizing. Obviously it could easily be related to other things completely. I'm trying to keep an eye on time of possession/possession management stuff this year. I think that may be at the heart of "cantankerous" discussions at this website once the puzzle is solved. Some teams are better at getting that extra opportunity than others, and both DVOA and traditional stats are more focused on production than possession/time squeezing, if that makes sense. Blogged about the possibility awhile back. Others have mentioned it in comments regarding Peyton Manning I think. Glad to see a big article at ESPN (linked to by your site today) on the topic.

Other examples:
Tampa Bay was on the field 47 extra seconds per game this year
NYJ were on the field 1:31 less per game this year
Green Bay was on the field about 2 minutes less per game this year

Also looked up strength of schedules to see if that might be an influence. Maybe DVOA isn't fully accounting for that. Using Sagarin's strength of schedule stuff to keep it all outside for this particular category (if DVOA is polluted because it's not capturing strength of schedule properly, you don't want to use its SOS).

*Green Bay jumped from 12th to 2nd while playing the 32nd ranked schedule instead of the 21st.

*Buffalo jumped from 22nd to 8th while playing the 17th ranked schedule instead of the 8th.

*Tampa Bay fell from 6th to 25th while playing the 1st ranked schedule instead of the 27th.

Those are the most extreme examples. We do have counter-examples...with the Jets making a big leap forward while playing a much tougher schedule...and Denver doing the same thing. Both of those teams had coaching changes, and moved toward much more conservative offenses that may have better protected the defenses (Orton and a rookie compared to Cutler and a sore-armed Favre).

Trying not to bog down in data. But, you do a few hours of research you want to get something down on paper!

Summing up:
*Defenses did have dramatic turnarounds in a way that's likely uncharacteristic from past seasons. That could reasonably be at the heart of mis-assessments this year.

*Defenses are trending worse while bunching up, which isn't something people would automatically assume would happen. This might have been the year they figured out a way to counter-act "legal" holding...there may have been a big difference between have's and have not's (as we see offensively with QB's). Instead, things got worse and bunched up.

*Analysts might consider placing more weight on the QB/offense's ability to protect a defense when evaluating how a defense will perform (fresh defenses are better than tired defenses, so there is some synergy there).

*Coaching changes can lead to more extremes in performance than might normally be anticipated, with Denver and NYJ making big leaps forward and Tampa Bay falling off the map. A combination of a coaching change with a QB change could magnify the issue.

Going back to the theme from other posts. Might be better to consider it a "strange" year in defensive transitions rather than a strange year in stats. The fact that stats aren't strange helped us see that. We'll have to see through the playoffs if what seems strange to many in the overall DVOA rankings ends up making more sense (or less) once the top teams are playing each other.

162
by nat :: Thu, 01/07/2010 - 4:50pm

That's a little long. Could you put in Haiku form?

163
by Jeff Fogle :: Thu, 01/07/2010 - 5:03pm

An explanation?
Defenses in transition
Sure to cause headaches

164
by Jerry :: Thu, 01/07/2010 - 7:14pm

Nice work, Jeff. The questions are interesting.

165
by DeltaWhiskey :: Fri, 01/08/2010 - 6:31am

Excellent commentary and analysis of the state of defenses.

If I were going to try and improve DVOA, I would look to improve the measurement of defense. Below are the correlations of DVOA, OFF DVOA, DEF DVOA and S.T. DVOA with wins.

DVOA 0.854
OFF DVOA 0.658
DEF DVOA -0.512
S.T. DVOA 0.239

The correlation of OFF DVOA and DEF DVOA with wins appear to be fairly similar; however, the R-square for each, reveals a picture that is vastly different. The R-square for OFF DVOA (0.433) suggests that about 43% of the variance in WINS is explained by OFF DVOA. That’s a fairly robust finding, but it makes sense – roughly, ½ of winning is due to offensive production.

The R-square for DEF DVOA (0.262) suggests that about 26% of the Variance in WINS is explained by DEF DVOA, and the R-Square for S.T. DVOA (0.057) suggests 6% is explained by Special Teams play.

My suspicion is that OFF DVOA is pretty much maxed out at this point, but there may be room to improve DEF DVOA. Improving the correlation of DVOA with wins to -0.58 would improve the R-square to around 0.34. Essentially this would mean that instead of DEF DVOA explaining only ¼ what goes into WINS, it would now explain 1/3. Moreover, if I’m calculating correctly, overall DVOA’s correlation to WINS would improve to around 0.895 with an R-square of 0.801.

My impression is that DEF DVOA is calculated in a manner that is essentially the inverse of OFF DVOA. That is, if a team gets +5 offensive success points on a particular play, then the DEF also receives +5 success points (which since DEF DVOA is calculated in the negative direction is a bad thing). However, a different metric may be required. Perhaps a play that receives +5 offensive success points should only penalize the defense +3 points, or a play that is a -3 for the Offense, should perhaps be a -5 for defensive calculations. Essentially, the scale of offensive success may not be the appropriate scale to measure defensive success.

This may be a fundamental error for DVOA, which I suspect is related to how the measures were developed initially. I don’t get the sense that the concept of success points was empirically derived. That is, rigorous statistical techniques were not used to derive either the appropriate definition of success or the proper score for success. Nonetheless, it appears that good definitions and scores were developed for offense, but perhaps on the defensive side of the ball, the measures are lacking – especially if defense was derived based on assumption that it is simply the opposite of offense.

166
by Bowl Game Anomaly :: Fri, 01/08/2010 - 11:36am

There is also the possibility that since the league seemingly keeps changing the rules to help offense and hurt defense, offense actually does matter more than defense and DVOA is accurately reflecting that.

Basically, a team with good offense and poor defense will win more games than a team with good defense and poor offense, because "good defense" isn't as impactful as "good offense." More precisely, "good defense" is not as "good," in absolute terms, as "good offense."

Or maybe I'm getting this completely backwards and the correct fix would be to manually adjust the baseline so that offense and defense are always equal in a given year.

167
by Rick A. (not verified) :: Fri, 01/08/2010 - 1:05pm

Try as the owners may, and try as they might, truly good defense will still win in December/January pro football. That remains my personal, long honed view. The entire "school" of thought that minimizes defense in today's NFL, and certainly DVOA and it's readers tend toward that, just do not see truly good defense very often. Just on the fundamentals level we have, today, even the top teams unable, or unwilling, to tackle properly and execute the way they are supposed to. Prior to the injury to the middle linebacker the Viking front 7 was perhaps the best in all football, yet even they would undergo periods of horrible tackling and undisciplined gap maintenance. During this past decade all of this sloppiness increased in the NFL. People are drafted based on their physical potential, not on their attainments of honing their game skills, with the idea that they can be coached into what it takes. Yeah right.

When a team does get tight on defense, and has the talent, even with today's never ending rule changes in the direction of offense, they can still stop juggernauts. (NYG over N.E.) Right now we have the Jets making it in--while alot of people, including many of the posters/readers of this site, mock Rex Ryan and the things he says. There is an entire article and thread here devoted to ridiculing an interview with the guy. Don't be surprised if the Jets create havoc in the coming playoffs. At this point I'm not so sure they won't get all the way to the AFC Championship even though I still expect Indy to be there in the end. To me, any team in there will have their hands full with them...

168
by Aaron Schatz :: Fri, 01/08/2010 - 1:31pm

There was a significant problem with our 2009 defensive projections that was very different from other years. I'll address this in full in next week's DVOA ratings article.

169
by nat :: Fri, 01/08/2010 - 2:39pm

...the league seemingly keeps changing the rules to help offense and hurt defense...

People keep saying this, and in some senses it is true, but the league has been remarkably good at keeping scores similar from era to era. Looking at median points allowed in 2009, 2005, and every fifth year before that back to 1960, I see values ranging from 19 pts/game (1970) to 21.6 pts/game (1965). Very high scoring years might go a bit higher, like 21.9 pts/game in 2007 and 2008.

I think the league is fiddling the rules to keep the right range of scoring. They seem to want the scores high enough that the game is interesting, and low enough that underdogs have a reasonable chance of an upset, again to keep games interesting.

170
by Jeff Fogle :: Fri, 01/08/2010 - 2:56pm

Nat, do you have the median values just for this decade? If it's a hassle to come up with, never mind. But, if its handy, it might shed some light on how things may be changing during the window in time DVOA was developed. Or, how they're not changing much if that's the case. You said 2007 and 2008 were "very high scoring years." That's immediately after the league liberalized holding rules to basically only call the obvious takedowns.

Conceivably, the big picture efforts of the league to keep scoring in a certain range could wreak havoc with ratings systems designed during a relative "down" phase. And, if the league is shooting for the high end of the range because it maximizes entertainment value...systems designed before those adjustments might need additional tweaking...

171
by nat :: Fri, 01/08/2010 - 3:56pm

I did this "by hand" so there may be errors in it, but here goes:

2009 20.6
2008 21.9
2007 21.9
2006 20.6
2005 20.8
2000 20.4

Those are the ones I did in the "aughts" decade. It's 23 points extra per team in 2008 vs 2000. Think of it as an extra scoring drive every third game.

One caveat: I was use using points allowed. That includes defensive scores allowed, so it is possible that the scoring is stable while the offensive scoring varies more. Someone should check that.

172
by Jeff Fogle :: Fri, 01/08/2010 - 4:36pm

This article just dropped off the front page of the site, so few may read this, lol.

Maybe I spoke too soon about defenses trending worse. Nat shows a decrease in scoring. Even though the first 26 spots (going from memory now because the comment box only shows the thing you're responding too) on the DVOA defensive ladder were worse this year, and only the bottom few spots better...the SUM of the defensive DVOA is actually better this year than last. I get about +69 this year, compared to about +108 last year (by hand because I was short on time). The best defensive numbers are negative, so going down means improving.

Quickly eyeballed "16th place" in some defensive stats (shortest cut I know for estimating the middle of the range). Yardage allowed went from 329 last year down to 327 this year, ppg went from 21.9 allowed to 20.8 for the 16th place team.

Those aren't in line with "defenses trending worse," so maybe we're just dealing with a bunching effect only...even with the majority of the spots on the ladder being worse off this year than those spots were last year. Always something new to think about...

Wish I "spoke" DVOA better. It's like trying to learn Klingon when English will do...

173
by Rick A. (not verified) :: Fri, 01/08/2010 - 5:10pm

All of this that you guys are speaking of may be true in the aggregate--but there is no question that the League has wanted to help the offense. My thought, without checking the entire history of it is that, in spite of the rule changes, there are more teams futile on offense today while the good teams are getting better at scoring. This has to do with the free agency era and the simple fact that certain teams are not fielded in good faith by ownership. (of course that's another premise I've been attacked for putting forward on this site but I stand by it) So the average scoring per team remains pretty much the same. But you have alot of shutouts these days--I mean, how's this for futility?, the Rams were shutout twice in the opening 4 weeks of the season. How often has ineptitude like that happened in the past ? Even with expansion teams. At the same time there are more teams lighting up the boards. You know, the League, I assure you, does not approach things with the kind of science you seem to think they do when you say they want to keep scoring high enough to not be boring but low enough to allow for upsets. What they want is scoring and they want to protect their multi-million dollar horses which are primarily the QB's. They don't care about all the things of Rozelle's yesteryear--i.e. "parity". Not enough to do anything meaningful about the teams that are simply not competitive. They can't anyway. Earlier this year we've went around and around on these threads about the supposedly enforced, good faith, cap and bottom. I assure you the teams that want to get around that...

174
by tuluse :: Fri, 01/08/2010 - 9:00pm

Scoring does not necessarily equal offense, it's possible special teams and defensive scoring have increased or decreased. Also, teams might be putting together longer more consuming drives that end up lowering score by shorting the game, but is actually a byproduct of it being easy to play offense.

I would like to see median yards as wells.

Offensive DVOA has definitely increased.

175
by Rick A. (not verified) :: Fri, 01/08/2010 - 9:38pm

Yes, bullseye. All very good points. All things are relative inside the game. Obviously, the essence of football is to try and get two or three scores up on the other guy and then keep it away from him, shorten the game, and so on. The really proficient teams do that, of course, and thus your average score in a ballgame remains pretty much the same as it's ever been--saying nothing whatsoever about how much difficulty, or not, there was in the superior team's road to that position...

98
by Johnny (not verified) :: Tue, 01/05/2010 - 2:02pm

Not that I think Ron Turner is a great OC, but they have bigger problems than him. I think most OCs would have had the same or at least similiar results at Turner this year: terrible o-line, young receivers, forte looked 2 steps slower this year, and oh yeah....the Cutler.

Good luck next OC.

111
by panthersnbraves :: Tue, 01/05/2010 - 4:52pm

I think it would be telling (given the small sample) to look at the Panthers D-line problems (first three games) and the Jake melt-down in the middle, and see how the weekly DVOA chart tracks.

I think what I am saying is "How well would the Panthers have done if the players in week 16 had been the players all year?" Might they have gotten two more games and a Wildcard?

112
by Eddo :: Tue, 01/05/2010 - 4:59pm

I see your point, but two games would have put the Panthers at 10-6, and both NFC Wildcard teams had 11-5 records.

I think that, while the Panthers did improve quite a bit, they still would have fallen a little bit short; with their best players in, I'd say subjectively they were an above-average team, but the NFC was rather strong at the top this year. My guess is the Panthers, playing Moore all year among other things, would have finished as the seventh-best NFC team (or maybe the sixth-best, ahead of Arizona, but that doesn't help them make the playoffs).

I don't intend this to be a knock on the Panthers, just explaining my thoughts.

113
by panthersnbraves :: Tue, 01/05/2010 - 5:08pm

True, I realized that one of the wins would have to be against the Eagles, dropping them to 10-6 as well, and then giving the Panthers the tie-breaker....

I realize that revisionist history is a tail-chasing exercise - mostly just trying to put a happy finish onto an odd and frustrating year. I'm hoping that playing the "injury card" is at least defensible.

121
by morganja :: Tue, 01/05/2010 - 6:37pm

I give Fox a lot more credit than most. I suspect that Moore just wasn't ready early in the year. Remember that McCown was suppose to be the number 2 guy but was horribly mangled by the Eagles and put on IR. Moore was most likely getting zero reps before then in practice. I think that if Moore had started in week 4 or 5 he would have had his confidence destroyed and the Panthers would have had no one to QB the last half of the season. Fox salvaged a terrible situation with all the injuries, a new defensive coach, and the hardest schedule in the league and turned it into an 8-8 season. That's a job well done.

130
by cfn_ms :: Tue, 01/05/2010 - 10:35pm

Can someone go a bit into exactly why the Colts and Saints are rated so poorly? They both have mediocre schedules, but it doesn't seem substantial enough to matter. Green Bay, on the other hand, had the 31st ranked schedule, was only 11-5, and seems intuitively like they should be below those two. Ditto for Minnesota, who had the 29th schedule and was rated right between Indy and New Orleans.

137
by jedmarshall :: Wed, 01/06/2010 - 10:10am

I think the Colts have always been tricky for DVOA because their style of play differs quite a bit from the norm. They let their foot off the gas a lot more than most teams do, which I believe hurts their ratings. They win a lot of games 24-17 after the other team takes too much time getting a garbage TD or two late, even though it feels like the game was never in question.

Also, no stats to back this up, but it seems that DVOA underrates Tampa-2 style defenses. The success rate against the Colts is probably pretty good. They allow a lot of 4 yard 1st down runs and 6 yard passes which DVOA likes at the benefit of rarely ever giving up a big play. They'll allow slow methodical drives, then hold them to a field goal or have a big drive killing sack. Plus it always helps to know even if the other team scores, you have Peyton Manning coming out on the field. Colts fans will trade field goals for touchdown (see BAL game) any day.

140
by Bowl Game Anomaly :: Wed, 01/06/2010 - 1:50pm

DVOA might underrate Tampa-2 defenses, but then you'd have to explain how the best defense since 1993 according to DVOA was a Tampa-2 (the 2002 Bucs).

132
by Dales :: Tue, 01/05/2010 - 11:08pm

Aaron or any of the staff that would know this-- I've been curious about one thing. I think I recall reading that DVOA is better correlated with a teams' strength the next season than late-season performance. Yet it seems like FO has been putting more of an emphasis on weighted DVOA as being the thing to look at when it comes to teams.

Which has correlated better with winning in the post-season? Season's end DVOA measures, or season's end weighted DVOA? Which has correlated better with the subsequent year's performance?

141
by Bowl Game Anomaly :: Wed, 01/06/2010 - 1:56pm

I'm almost positive that regular DVOA predicts next year's DVOA better than wDVOA does. I think wDVOA is better at predicting the playoffs but I don't know for sure if there has been a formal study comparing DVOA, wDVOA, and other measures for ability to predict the playoffs. Probably no measure is great, because when you're talking about single elimination over a short season (1-4 games), random and unpredictable factors can have a huge impact. Plus, DVOA doesn't consider home/away or bye weeks, which have a large impact in the playoffs.

143
by Anonymous420 (not verified) :: Wed, 01/06/2010 - 2:09pm

it's odd how little the Vikings get talked about on this site. /just sayin'

153
by Rick A. (not verified) :: Wed, 01/06/2010 - 8:46pm

That's only very recently. Earlier in the year there was much back and forth about Minny. There were, seemingly, several people from the Twin Cities area who were all emotional about the big start. There were others who were seemingly very motivated to post threads voicing their dislike and resentment of No. 4. There was me, someone with a longtime involvement in pro football proclaiming them the best in the NFC in my judgement. I had seen them live along with several others, including N.O., to contrast them with. I'd also seen the team pre Favre in the last couple of years, along with several of the other main candidates for this year's King. Then the slump came and a lot of people stopped commenting. Nonetheless, they earned more Pro Bowl starters than anyone else just as I anticipated. And, as for my outlook since the slump--the world knew what their weaknesses were to begin with and they were certainly exploited by the Cards and Bears, but my forecast remains that they will host the Championship game. I further see them winning that contest. Obviously, any of the teams with a focused pass rush or a heavy air artillery can cause them problems--and that would include any of the other 5 teams. Still, they have the most talent and the least problems in my view. Cream most often does rise to the surface...

147
by Bright Blue Shorts :: Wed, 01/06/2010 - 4:19pm

I think Darius Heyward-Bey just had an unlucky set of circumstances ... has any WR, who drops that many balls, ever got to play in that many games with a QB who has such poor decision-making skills and such poor accuracy?

152
by tuluse :: Wed, 01/06/2010 - 5:51pm

I'm sure it's happened but it's not a big story because the WR wasn't the 7th overall pick, and the QB wasn't the 1st overall pick.

154
by Al Davis (not verified) :: Wed, 01/06/2010 - 9:25pm

I agree with you! I've been saying that he's the next Jerry Rice since April! Finally another SANE person!

Whooops, I pooped. Gotta Go!

177
by anton (not verified) :: Thu, 11/11/2010 - 8:32am

The first priority of the site administrator for your comments right know, thank you. Really very good information-sharing gives you.
Thanks in knowledge

178
by burbul (not verified) :: Fri, 11/12/2010 - 12:19am

of course you are right here...

179
by porno (not verified) :: Fri, 11/12/2010 - 4:05pm

the anime was probably Gundam Unicorn

180
by NYC Beer Gardens (not verified) :: Sat, 08/06/2011 - 1:43pm

I'm glad the new season is finally coming!

181
by chirs (not verified) :: Thu, 09/15/2011 - 6:17pm

Thanks nice post

182
by hanna (not verified) :: Thu, 09/15/2011 - 6:18pm

The first priority of the site administrator for your comments right know, thank you. Really very good information-sharing gives you.
Thanks in knowledge

183
by shankar (not verified) :: Fri, 02/15/2013 - 6:57am

thanks for this very nice http://www.packersmoversinmumbai.com