Writers of Pro Football Prospectus 2008

Most Recent FO Features


» Defense and Rest Time

Do defenses really wear out over the course of a game? Do defenses benefit from long drives that give them more time to rest on the sideline? Guest columnist Ben Baldwin investigates.

30 Oct 2012

Week 8 DVOA Ratings

by Aaron Schatz

We have a surprising new number one in this week's DVOA ratings: Denver. The DVOA system believes that Denver's dominating victory over New Orleans (54.1% DVOA) was slightly dominating-er than San Francisco's dominating victory over Arizona (49.5% DVOA), and along with the continued expansion of our strength of schedule adjustments -- now at 80 percent -- that puts the Broncos slightly ahead of the 49ers. 

Actually, New England's victory over St. Louis was the dominating-est victory of the week (76.7% DVOA), but the Patriots had more ground to make up on the 49ers, so they're still behind Denver and San Francisco in third place.

Both Denver and San Francisco now rank in the top five for both offense and defense, with roughly average special teams. We all know the Broncos' defense has improved this year, but top five is a bit of a shock. However, this is where the strength of schedule issue really comes in. The Broncos have played the toughest set of opposing offenses so far, including five games against teams in the top 10 of offensive DVOA. And things are about to get much, much easier for the Broncos. Denver has the easiest remaining schedule in the league by a huge margin. Their average opponent remaining has a DVOA of -17.7%. Oakland is next at -8.8%. The average offense left on the Denver schedule has a DVOA of -11.3% (ranked 32nd) and the average defense left has a DVOA of 6.8% (ranked 30th, ahead of only Miami and Houston). Denver's road back to the playoffs is now so obvious that Von Miller could even see it without his ironic horn-rimmed glasses.

Although the AFC now has two teams in the top three, the big gap between the conferences didn't go anywhere. The bottom nine teams are still all AFC teams. The top nine teams in the league form a nice solid group that includes three AFC and six NFC teams. Then there is a gap, then Miami ranked 10th, then another gap, and then you have the beginning of (relative) mediocrity with Baltimore, Pittsburgh, and Minnesota. Those are three teams going in very different directions. Pittsburgh is on its way up, especially because like Denver it has a very easy schedule remaining (30th in the NFL). Baltimore is headed down because of injuries, and Minnesota because it has the league's toughest remaining schedule (and, to make matters worse, opponents may have figured out whatever Christian Ponder was doing right earlier this year, but I believe we're leaving disection of that to Andy Benoit in Film Room on Thursday).

Eleven of the 12 toughest remaining schedules belong to NFC teams. That's what happens when one conference is so clearly better than the other. The AFC's gap between the top three and everyone else is so large that the top three most likely teams to win the Super Bowl according to FO playoff odds are all AFC teams: Houston, Denver, and New England, in that order. Those teams just have very little competition except for each other.

Someone asked me on Twitter why the FO stats still have Atlanta just ninth despite its 7-0 record. It's actually summarized pretty easily. First, close wins: four by a touchdown or less. Second, its schedule so far ranks 29th in the NFL. Third, the Falcons have recovered 75 percent of fumbles. (On the other hand, Denver has recovered just 24 percent of fumbles including none of its own offensive fumbles; Green Bay has recovered just 14 percent of fumbles.)

Some years, it seems like FO is not applying opponent adjustments strongly enough because the top teams all seem to be listed with very easy schedules, but that's not the case in 2012. In that top group of nine main Super Bowl contenders, only two teams have played schedules ranked lower than 20th: Chicago (28th) and Atlanta (29th).

Scroll down a bit for some fun analysis of FO's new snap count data page.

* * * * *

All stats pages should now be updated (or, at least, will be in the next few minutes). FO Premium stats will be updated later this evening.

* * * * *

These are the Football Outsiders team efficiency ratings through eight weeks of 2012, 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 represents an attempt to figure out how a team is playing right now, as opposed to over the season as a whole, by making recent games more important than earlier games.

Note that as of this week, we are no longer using the DAVE ratings which incorporate preseason forecasts.

Because it is early in the season, opponent strength is at only 80 percent; it will increase 10 percent every week through Week 10. 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>

1 DEN 36.7% 5 36.7% 1 4-3 22.5% 2 -15.0% 5 -0.8% 20
2 SF 36.3% 1 36.1% 2 6-2 19.7% 5 -16.7% 4 -0.2% 18
3 NE 33.9% 6 34.2% 3 5-3 31.2% 1 0.5% 18 3.3% 11
4 NYG 32.0% 3 32.4% 4 6-2 22.0% 3 -5.2% 12 4.7% 7
5 CHI 28.7% 4 28.9% 5 6-1 -12.5% 25 -33.4% 1 7.8% 3
6 GB 28.7% 2 28.5% 6 5-3 20.0% 4 -7.6% 8 1.0% 15
7 HOU 25.1% 7 24.9% 7 6-1 10.7% 9 -24.1% 2 -9.6% 32
8 ATL 21.1% 9 21.0% 8 7-0 11.5% 7 -6.0% 10 3.6% 10
9 SEA 20.3% 8 20.3% 9 4-4 -1.4% 16 -19.5% 3 2.3% 14
10 MIA 12.3% 10 12.2% 10 4-3 -4.4% 20 -10.1% 7 6.6% 5
11 BAL 4.8% 12 4.3% 11 5-2 3.7% 12 7.4% 24 8.4% 1
12 PIT 3.7% 16 4.0% 12 4-3 8.9% 10 4.5% 20 -0.7% 19
13 MIN 3.6% 11 3.2% 13 5-3 -6.9% 21 -2.3% 15 8.2% 2
14 DAL 1.4% 18 1.6% 14 3-4 0.2% 15 -5.8% 11 -4.6% 27
15 DET -1.1% 15 -0.9% 15 3-4 13.8% 6 6.0% 23 -8.9% 31
16 TB -3.2% 20 -3.0% 16 3-4 1.3% 13 -2.6% 14 -7.1% 29
17 WAS -5.2% 13 -5.1% 17 3-5 7.6% 11 5.5% 22 -7.3% 30
18 NYJ -6.9% 17 -6.6% 18 3-5 -14.5% 27 -3.8% 13 3.8% 9
19 PHI -9.7% 22 -9.8% 19 3-4 -12.4% 24 -6.8% 9 -4.0% 25
20 ARI -10.4% 21 -10.8% 21 4-4 -22.0% 30 -12.4% 6 -0.8% 21
21 CAR -10.5% 23 -10.7% 20 1-6 -7.5% 22 -1.1% 16 -4.2% 26
22 NO -11.4% 19 -11.5% 22 2-5 11.2% 8 25.2% 32 2.6% 13
23 STL -11.6% 14 -11.8% 23 3-5 -10.2% 23 -0.6% 17 -2.0% 22
24 OAK -12.9% 27 -12.4% 24 3-4 -2.3% 18 4.9% 21 -5.7% 28
25 SD -13.3% 24 -13.4% 25 3-4 -13.9% 26 2.1% 19 2.7% 12
26 CIN -15.9% 25 -15.8% 26 3-4 -4.1% 19 16.7% 28 4.8% 6
27 BUF -16.7% 26 -16.6% 27 3-4 -2.0% 17 22.2% 30 7.6% 4
28 CLE -25.0% 29 -25.0% 28 2-6 -20.5% 29 8.5% 25 4.0% 8
29 IND -25.7% 28 -25.9% 29 4-3 0.3% 14 23.5% 31 -2.5% 23
30 JAC -35.6% 30 -35.6% 30 1-6 -22.7% 31 10.3% 26 -2.6% 24
31 TEN -35.7% 31 -35.9% 31 3-5 -17.7% 28 19.0% 29 0.9% 16
32 KC -41.7% 32 -41.4% 32 1-6 -29.1% 32 12.9% 27 0.3% 17
  • NON-ADJUSTED TOTAL DVOA does not include the adjustments for opponent strength or the adjustments for weather and altitude in special teams, and only penalizes offenses for lost fumbles rather than all fumbles.
  • 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. Teams that have had their bye week are projected as if they had played one game per week.
  • PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • 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).

1 DEN 36.7% 4-3 29.9% 8.0 1 6.6% 7 -17.7% 32 5.9% 9
2 SF 36.3% 6-2 35.0% 7.0 2 6.2% 8 6.3% 7 22.4% 30
3 NE 33.9% 5-3 37.5% 6.6 3 -2.4% 19 0.1% 14 10.9% 21
4 NYG 32.0% 6-2 31.1% 6.1 5 -1.8% 17 2.0% 11 14.6% 27
5 CHI 28.7% 6-1 32.9% 6.4 4 -7.8% 28 7.8% 4 12.8% 25
6 GB 28.7% 5-3 22.3% 5.9 6 3.3% 12 2.4% 10 13.5% 26
7 HOU 25.1% 6-1 24.8% 5.8 7 0.6% 15 -8.2% 28 15.7% 28
8 ATL 21.1% 7-0 29.7% 5.7 8 -8.1% 29 -2.0% 21 11.3% 22
9 SEA 20.3% 4-4 8.2% 5.7 9 8.3% 6 4.4% 8 3.9% 2
10 MIA 12.3% 4-3 12.5% 5.6 10 -5.7% 23 -0.6% 16 5.2% 5
11 BAL 4.8% 5-2 7.5% 4.3 12 -4.5% 21 0.4% 13 17.6% 29
12 PIT 3.7% 4-3 10.3% 4.0 14 -7.1% 26 -8.7% 30 9.5% 18
13 MIN 3.6% 5-3 11.1% 4.7 11 -10.1% 32 18.4% 1 7.0% 11
14 DAL 1.4% 3-4 -5.6% 4.1 13 14.9% 1 -6.4% 27 9.5% 19
15 DET -1.1% 3-4 -4.6% 3.8 15 4.6% 9 7.1% 6 4.4% 3
16 TB -3.2% 3-4 5.8% 3.4 18 -4.6% 22 1.1% 12 8.7% 14
17 WAS -5.2% 3-5 0.1% 3.4 19 2.3% 13 -1.9% 20 8.8% 15
18 NYJ -6.9% 3-5 -8.1% 3.4 17 10.2% 4 -8.6% 29 22.7% 31
19 PHI -9.7% 3-4 -12.0% 3.2 20 3.6% 11 -1.9% 19 3.0% 1
20 ARI -10.4% 4-4 -15.1% 2.6 27 8.6% 5 14.4% 2 5.7% 8
21 CAR -10.5% 1-6 -15.0% 3.1 21 12.7% 3 -4.4% 24 9.2% 17
22 NO -11.4% 2-5 -8.5% 2.9 24 -1.2% 16 8.4% 3 8.3% 13
23 STL -11.6% 3-5 -15.0% 3.0 23 13.4% 2 7.4% 5 8.9% 16
24 OAK -12.9% 3-4 -8.3% 3.1 22 -2.4% 18 -8.8% 31 10.3% 20
25 SD -13.3% 3-4 -4.2% 2.7 25 -9.8% 30 -5.1% 25 7.2% 12
26 CIN -15.9% 3-4 -4.0% 2.7 26 -10.0% 31 0.1% 15 12.4% 24
27 BUF -16.7% 3-4 -16.0% 3.5 16 -7.1% 25 2.7% 9 23.0% 32
28 CLE -25.0% 2-6 -7.4% 2.0 28 -7.6% 27 -1.2% 18 6.5% 10
29 IND -25.7% 4-3 -20.8% 2.0 29 -6.0% 24 -3.8% 23 5.5% 7
30 JAC -35.6% 1-6 -37.3% 0.8 31 4.5% 10 -5.6% 26 5.2% 4
31 TEN -35.7% 3-5 -32.6% 1.0 30 1.2% 14 -1.1% 17 5.5% 6
32 KC -41.7% 1-6 -44.7% 0.6 32 -4.5% 20 -2.9% 22 12.0% 23

* * * * *

So, this year, the NFL finally made snap count data publicly available, and we've been collecting it each week and displaying it for free on our new snap counts page. Here are some interesting tidbits from the data through the first eight weeks:

Many players have been in for 100 percent of their teams' snaps on offense, primarily offensive linemen and quarterbacks. The Patriots lead the league because of the super hurry-up they've been running, so Ryan Wendell is the league leader with 713 total snaps and 635 offensive snaps. Nate Solder has 686 total snaps and 634 offensive snaps.

Only nine players have been in on 100 percent of defensive snaps: Morgan Burnett of Green Bay leads those players with 564 defensive snaps. The others are Ryan Kerrigan of Washington; James Laurinaitis of St. Louis; Kam Chancellor of Seattle; Curtis Lofton of New Orleans; Kurt Coleman of Philadelphia; Eric Berry of Kansas City; and both San Diego safeties, Eric Weddle and Atari Bigby.

None of those players actually lead the league in defensive snaps. That title goes to T.J. Ward, who has been in on 589 of Cleveland's 601 defensive snaps. No defensive lineman has been in on 100 percent of snaps, but Cameron Jordan of New Orleans is the closest at 94 percent, followed by San Francisco's Ray McDonald and Justin Smith along with Jacksonville's Jeremy Mincey at 93 percent. Rookie Chandler Jones has been on 92 percent of defensive snaps for New England.

No player has been in for 100 percent of his team's special teams snaps. Andy Studebaker of Kansas City is closest; he's played in 100 percent of Kansas City's special teams snaps in every week except Week 1 and Week 8, and he's played in 93 percent of special teams snaps overall. Jason Trusnik of Miami has also played 93 percent of special teams snaps. Studebaker also played 26 snaps on defense in Week 1, but has only been in on seven snaps since. Trusnik has played 15 snaps on defense. Both players are linebackers.

Which teams do the most to eschew a running back committee? Pretty much the teams you expect. Four running backs have been in on more than 75 percent of offensive snaps: Darren McFadden (85 percent), Ray Rice (83 percent), LeSean McCoy (81 percent), and Chris Johnson (79 percent). Larry Fitzgerald leads all wide receivers with 98 percent of team snaps. Calvin Johnson and Steve Johnson are at 97 percent. Donnie Avery (!!!) is fifth at 94.1 percent.

Tight end Heath Miller is the only "skill player" in the league, other than quarterbacks, who has been in on 100 percent of his team's offensive snaps.

The players with the most snaps without being in for special teams at least once are Tom Brady, Brandon Lloyd, and Jeff Saturday.

This year's biggest double threat (well, triple threat, including special teams) is San Francisco defensive end/fullback Will Tukuafu. In fact, you can't see all of his snaps in one place on the chart because San Francisco changed his number at some point from 92 to 48. (I'll get that fixed up by next week.) Tukuafu has been in for 56 snaps on offense, 20 on defense, and 102 on special teams. The only other player listed with at least five snaps on both offense and defense is Charles Tillman. Does anyone know what Tillman has been doing on six offensive snaps? I'm curious. Patrick Peterson, Antonio Cromartie, and Indianapolis linebacker Matt Harvey have been in for four offensive snaps apiece despite generally being defensive players.

Sometime in the next few days, I'll do some XP commentary looking at things like pass targets or tackles per snap.

(Note: I am not sure if fake field goals and punts count as offense/defense snaps or special teams snaps. I have an e-mail in to the league asking for clarification.)

Posted by: Aaron Schatz on 30 Oct 2012

202 comments, Last at 12 Apr 2013, 1:11pm by JoeyHarringtonsPiano


by Miggity Mike (not verified) :: Tue, 10/30/2012 - 4:08pm

Denver is clearly ranked too low because I am their fan. A baked potato is way better than this. Raider's be the worst team in the NFL!

by Insancipitory :: Wed, 10/31/2012 - 1:56am

The Seahawks are clearly ranked high because Russell Wilson hasn't yet completed his hero's journey. Dragons slewn and maidens rescued divided by descents into the underworld is way better than this. Kieth 'I will NOT wear those sunglasses. Ok, I will' David narrated his rise to prominance on NFL.com

by The Hypno-Toad :: Wed, 10/31/2012 - 2:04am

I happened to watch an old VHS of Nightmare Before Christmas tonight that had an ad for Gargolyes on it. That reminded me that I love Kieth David to an unreasonable degree.

by Dondraper78 (not verified) :: Tue, 10/30/2012 - 4:09pm

Why does Denver have 8 estimated wins with a 4-3 record? They should have won more games than they played?

by kbukie :: Tue, 10/30/2012 - 4:12pm

Estimated wins is on a per-week basis, and doesn't take byes into account.

by Nevic (not verified) :: Tue, 10/30/2012 - 4:23pm

8 wins still seems high given that they are only slightly ahead of SF in DVOA and have played a schedule of nearly identical toughness, yet SF only has 7.0 estimated wins. Is this due to the dispersion among opponent strenghts within the games they have played?

by kbukie :: Tue, 10/30/2012 - 4:30pm

It normalizes opponent schedules and fumble recovery rates (both of which will benefit Denver's estimate), and factors in a team's consistency (which also greatly favors Denver).

by Danish Denver-Fan :: Tue, 10/30/2012 - 4:36pm

Fumble recovery, perhaps?

by Jimmy :: Tue, 10/30/2012 - 4:12pm

Tillman is the deep back on kneel downs. Not entirely sure why they don't use an actual RB, maybe they want him to force a fumble if the offense does gove up possession with his patented ball punch.

by JoeyHarringtonsPiano :: Tue, 10/30/2012 - 4:34pm

Ah, the old anti-Greg Schiano strategy.

by Peregrine :: Tue, 10/30/2012 - 5:17pm

The deep back on kneel downs will only be required to do something if there's a fumble on the center-QB exchange. This might entail jumping on the ball or, more importantly, tackling someone. I can see why a DB might be well-suited for that role.

by Jerry :: Tue, 10/30/2012 - 6:20pm

The Steelers use Polamalu (when healthy) back there as well.

by BigCheese :: Wed, 10/31/2012 - 12:26am

Which brings up the point of why anyone would put an actual RB back there?

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by TomC :: Tue, 10/30/2012 - 6:44pm

As a Bears fan, I just celebrate that they've had that many kneel-down snaps already. Full seasons go by with fewer than six.

by kbukie :: Tue, 10/30/2012 - 4:13pm

Denver and Indianapolis, proof that all 4-3 records are not created equal.

Too bad Vegas oddsmakers are on to Indy a bit, otherwise this week's upset pick would have been the easiest of the year with Miami at Indy.

by Karl Cuba :: Tue, 10/30/2012 - 4:15pm

Maybe 'iconic' spectacles instead of ironic.

I also think I can preempt Mr Benoit and shed some light on Ponder's regression. Most young quarterbacks get worked out and then have to improve. Ponder is no different but he is also limited a bit by his lack of top level arm strength which means he needs to be in an offense that accommodates this such as a Bill Walsh descendant or maybe Bill Bellichick's modern spread. It also doesn't help that he doesn't have a target that can consistently win outside the numbers, even with the admirable Harvin and Rudolph.

by Brad M (not verified) :: Tue, 10/30/2012 - 6:42pm

Nope, ironic makes sense. They are hipster glasses.

by CeeBee (not verified) :: Tue, 10/30/2012 - 4:17pm

NYG DEF is 12th? They seem so much worse than that. Is it all the turnovers?

by Independent George :: Tue, 10/30/2012 - 4:49pm

Yeah, the computer doesn't differentiate between "the QB threw it right to the DB" versus "the defense pressured the QB into throwing into a tight window". I think the Giants have been lucky on turnovers in the first half-season, and will likely regress in the 2nd half. On the other hand, getting Canty, Williams, and Phillips back should also lead to an improvement.

I'm hoping that Stevie Brown can play opposite Kenny Phillips at safety, letting Antrel Rolle play close to the line in 'Big Nickel' packages instead of watching him blow all of his coverages. Jacquian Williams is their best cover LB, and putting him in instead of Blackburn should reduce the likelihood of further 18-catch days by TEs. Osi & Tuck have been underperforming thus far, but the interior line has been excellent.

I think the Giants match up well with Pittsburgh and Cincinnati; if they can reach the bye at 7-3 without further injuries, I like their chances even against that tough closing schedule.

by LionInAZ :: Tue, 10/30/2012 - 6:20pm

The Giants don't need luck to get turnovers -- they get four games against Romo and Vick every year.

by young curmudgeon :: Tue, 10/30/2012 - 9:14pm

oh, that's gonna leave a mark!

by Nevic (not verified) :: Tue, 10/30/2012 - 4:18pm

Wow, Chicago's defense is a healthy margin ahead of all the other top defenses. Also, it hasn't seemed to me like the GB defense is 8th best so far this year, but I guess they still have generated a decent number of turnovers to account for giving up yards.

by In_Belichick_We... :: Wed, 10/31/2012 - 8:59am

Not surprising when you look at their schedule:
3 weaker teams with brand new coaching staffs (Indy, St. Louis, Jacksonville).
A MNF game where Romo and his WRs were on two different planets.
An inconsistent-at-best Lions team.

They've been impressive with a pretty soft schedule. I don't think we'll hear any "greatest DDVOA ever" talk by Dec 1.

Entitled freeloader

by Danish Denver-Fan :: Wed, 10/31/2012 - 11:37am

Well the D in DVOA is supposed to account for that. So even considering their somewhat weak opponents they've still been flat-out dominant.

by tuluse :: Wed, 10/31/2012 - 11:41am

They also held GB in check, allowed 20 points with 1 TD coming on a fake FG, and Cutler melting down and throwing 4 picks.

by Mr Shush :: Thu, 11/01/2012 - 8:56pm

Supposed to is not the same as does. I firmly believe that some units are more affected by opponent quality than others, and I wouldn't be surprised if the Bears defense was one. If nothing else, I would fully expect Peyton to carve them up.

by Brent :: Fri, 11/02/2012 - 2:01pm

You are almost certainly correct that some teams are more affected by opponent quality than others. On the one hand, there are teams that certainly seem to play to the level of their opponents, seems like all their games are close games no matter the strength of their opponents. Then there are other teams that seem to smash poor teams but get run off the field against quality opponents. Variance should reflect that (though it won't distinguish among the different ways a team can be inconsistent), and a high variance tends to correlate with teams that are not great teams (but have some great games, maybe). I think that's really why they track variance. The really good teams tend to have a low variance.

by LionInAZ :: Fri, 11/02/2012 - 2:52pm

Really bad teams tend to have low variance, too.

by kbukie :: Tue, 10/30/2012 - 4:20pm

Another interesting thing is that Pittsburgh's surge this week makes 11 of the 12 playoff spots over 75% certain (with SEA at 60% for the last one). We could be looking at one of the more predictable postseasons in recent memory.

Of course, one Jay Cutler broken thumb, and everything falls apart.

by Paddy Pat :: Tue, 10/30/2012 - 4:42pm

Adding to this point, we could well be looking at the most playoff repeats in a very long time with only 3 new comers to the party.

by snoopy369 :: Tue, 10/30/2012 - 4:32pm

Re: Tillman, I don't have all-22 film, but if anyone wants to look, there were 3 week 3, 2 week 4, 1 week 7 snaps. Weeks 3 and 4 were during and just after the Forte injury, so I could hypothesize he was brought in on some wildcat-like package as a decoy for Hester/Bell/Anybody. In the Week 3 box score, he has no mentions of pass targets or rushes, only tackles. Week 4, of course, he was an important part of the Bears' offense, but only while Dallas had the ball. :) Again, no mentions of pass targets or rushes.

Week 7, there is one interesting tidbit. He is called for offensive holding during a punt (that the Bears were receiving). I wonder if that might cause him to be listed as offense, even though it's clearly a special teams play?

by Podge (not verified) :: Wed, 10/31/2012 - 8:50am

Putting this together with the kneeldowns theory above doesn't work perfectly.

The 3 in week 3 make sense, as they ended the game with 3 kneeldowns against the Rams. They had 1 kneeldown against the Lions in week 7.

In week 4 against the Cowboys though they only had 1 kneeldown, but he's credited with 2 offensive plays. They also had 3 kneeldowns against the Colts in week 1, but Tillman wasn't (apparently) in on either.

by Travis :: Wed, 10/31/2012 - 10:06am

There was also a kneeldown at the end of the first half in Week 4.

Brandon Marshall was the last man back for the kneeldowns in Week 1.

by Jimmy :: Wed, 10/31/2012 - 10:08am

It isn't a theory, that is what Tillman has been doing on the field on offense. I knows it because I saws it. Just answering Aaron's question.

by Marko :: Wed, 10/31/2012 - 12:38pm

Jimmy and Travis are correct. I watch every Bears game and never miss a snap (well, unless the TV cameras miss it, which happened on one occasion this year when they came out of commercial late to join a play in progress because the replacement officials resumed play too soon; they also missed a snap on one of the kneeldowns in question, although you could see the players on the field immediately after the snap and identify who was on the field). Tillman has been on the field for numerous kneeldowns this year, including one at the end of the first half. He has not been on the field for any regular offensive plays.

As for players on one side of the ball having snaps on the other side, I know some teams have tall WRs on the field to defend against Hail Marys in obvious situations. I remember Randy Moss doing that occasionally earlier in his career. I wouldn't be surprised if some teams used tall and athletic tight ends for this as well, such as Tony Gonzalez and Jimmy Graham.

by Nathan :: Wed, 10/31/2012 - 12:48pm

Gronkowski has done this in NE.

by commissionerleaf :: Thu, 11/01/2012 - 12:55am

Randy Moss did it during his time in NE too.

by Travis :: Wed, 10/31/2012 - 2:17pm

The flip side - in their final play of the 2008 season, needing to go 96 yards for the tying touchdown, the Jets put Darrelle Revis, Ahmad Carroll, and Hank Poteat (all defensive backs) on the offensive line in order to get as many fast players on the field as possible for a multi-lateral play.

by Podge (not verified) :: Wed, 10/31/2012 - 2:19pm

Yeah, uh... can I say I was using the word "theory" in its scientific sense, meaning backed up with lots of evidence (even though I obviously wasn't)? I can? Brilliant.

by falcochicquera :: Tue, 10/30/2012 - 4:34pm

My gut feel (as an Atlanta fan) is that there are 8 good to great teams who can be ranked in any order, then a big gap to the field. Consequently, my biggest surprise is not that we are 8th but rather that Seattle is so close to our DVOA despite their horrific away performances, and that Miami is appreciably above Baltimore. If MIA and SEA were both around 7% DVOA, that would tally with what I've seen so far this year.

Of course, this is gut feel rather than stats, but I wouldn't be surprised if that's the way it trends over the next few weeks.

by JoeyHarringtonsPiano :: Tue, 10/30/2012 - 4:39pm

"Seattle is so close to our DVOA despite their horrific away performances"

What was so horrific about Seattle's away performances? They lost 4 games by a total of 17 points.

by Viliphied (not verified) :: Tue, 10/30/2012 - 7:42pm

21 points*

Also, pure margin of defeat isn't everything. Seattle's D is good, and their offense isn't, so their games tend to be low scoring. Seattle losing 4 games by 21 total points is worse than, say, New England losing 4 games by 21 total points. Their pythag for their 5 away games is only 2.2. Not necessarily "horrific" but well below average.

by LyleNM (not verified) :: Wed, 10/31/2012 - 11:04am

2.2 is well below 2.5?

Seriously, though, in all 5 of their away games, the game was decided on the final drive. If that's "horrific", what adjective can you possibly use for, say, Kansas City?

by Paddy Pat :: Tue, 10/30/2012 - 4:44pm

Miami has really been pretty good so far this year. The 2 overtime losses, etc. Baltimore has been getting thinner week by week, the offense not as effective as it was or seemed like it would be, and the defense slowing down. The difference in numbers between MIA and BAL also isn't great.

by FrontRunningPhinsFan :: Wed, 10/31/2012 - 8:39am

I'll add that in those two OT losses, they had chances to win each game on field goals (one game in regulation, the other in OT) and missed those FGs. They're probably getting a lot more publicity if their record is 6-1 with the only loss to Houston instead of 4-3.

Of course DVOA doesn't care too much about that I wouldn't think, but people's perception would definitely be different.

No matter what I'm thrilled to even be in any sort of playoff discussion. It's been a long, long time.

Fire Jeff Ireland.

by Sergio :: Wed, 10/31/2012 - 2:41pm


*You* are thrilled to *be* in any sort of playoff discussion? Which team are you following now?

-- Go Phins!

by FrontRunningPhinsFan :: Thu, 11/01/2012 - 8:45am

You got me. I'm back to being a fan.

Fire Jeff Ireland.

by jebmak :: Thu, 11/01/2012 - 4:26pm

I'm not.

by Danish Denver-Fan :: Tue, 10/30/2012 - 4:39pm

My sports fan defense mechanism is working overtime these days.

Must.. not.. get... hopes up...

by Danish Denver-Fan :: Tue, 10/30/2012 - 4:42pm

Also: How rare is it for a team to have n estimated wins in week n for, say, n > 5?

Can you, somehow, have estimated wins > n in week n? I mean have the Broncos maxed out the EM statistic?

by Aaron Schatz :: Tue, 10/30/2012 - 6:50pm

Denver has indeed maxed out the Estimated Wins stat. It is not possible to have more Estimated Wins than there are weeks so far (or, more accurately, games played so far by teams that have not yet had a bye).

Denver is top 10 in variance, top 5 in both offense and defense, and then also third in late and close offense and first in late and close offense by a HUGE amount (91.3% DVOA, no other team above 40%, SF and NE are closest). They max out on Estimated Wins despite being just 26th in Q1 offense.

by Podge (not verified) :: Wed, 10/31/2012 - 8:54am

But wait, we all know that Peyton Manning isn't clutch...

by Brent :: Thu, 11/01/2012 - 2:29pm

"third in late and close offense and first in late and close offense" doesn't make sense. Maybe they're third in late and close defense?

by kbukie :: Tue, 10/30/2012 - 4:43pm

I know the first quarter counts just as much as the fourth, but I'm curious what Chicago's offensive DVOA is broken down by quarter.

Mostly because that first quarter number is going to look ugly.

by Jimmy :: Tue, 10/30/2012 - 4:57pm

They are pushing the Chiefs for worst rating in the first quarter (which begs the question; How do you get a worse DVOA than a team that gives up three sacks and a pick every game?).

They are appalling in the first, poor in the second, very poor in the third and lights out in the fourth (3rd in league).

Thats all you are getting, ordinal numbers; to see more buy premium.

Buy premium.....buy premium...

(I feel if I give out their data I should at least give it a good plug, I think it is pretty good, you want to know how well your team runs on second down, it'll tell you)

by RoyFlip (not verified) :: Wed, 10/31/2012 - 10:28am

All due respect to premium and thanks for your offer, but all Giants fans know the answer: Bradshaw, inside, 1.5 yards. Leaving a third-and-2: rinse, repeat. Throw the ball Kevin, throw it. Yet, somehow, they are third in offense. First Place Problems.

by Paddy Pat :: Tue, 10/30/2012 - 4:45pm

What I'm impressed by is that with such huge disparity between the conferences, the AFC still looks to field a reasonable playoff field and no NFC team is likely to get horribly screwed. There were some years around 2004 when decent AFC teams didn't make it while horrendous NFC teams did.

by CraigoMc (not verified) :: Tue, 10/30/2012 - 4:54pm

Didn't the 2008 Pats go 11-5 and miss the playoffs in a year when 9-7 got an NFc wild-card?

by CraigoMc (not verified) :: Tue, 10/30/2012 - 4:56pm

Scratch that - Atlanta won their division at 9-7, Philly got in as a WC with 9-6-1.

by Danish Denver-Fan :: Tue, 10/30/2012 - 5:03pm
by Stats are for losers (not verified) :: Tue, 10/30/2012 - 5:33pm

... that horrendous 9-6-1 Philly team, of course, was #1 in DVOA and "only" managed to go 2-1 in the playoffs, losing the conference championship to the 9-7 Cardinals.

Also, the 8-8 Chargers were 2 spots higher than NE by DVOA, and 20 points higher on the same field. Wasn't that the year of the Ed Hochuli botched fumble call in Denver?

by Mr Shush :: Tue, 10/30/2012 - 10:58pm

A better example would have been 2004. The 9-7 but 31.3% DVOA Bills and the 9-7, 21.1% DVOA Ravens, not to mention the 7-9, 15.1% Chiefs and the 8-8, 13.2% Bengals missed out because they were in the AFC, while the NFC had spots for the 10-6, 2.8% Packers, the -4.8%, 11-5 Falcons, the 8-8, -8.1% Vikings and the 9-7, -2.9% Seahawks - who promptly lost at home to the 8-8, -27.2% Rams. The third best team by DVOA missed out, and the second worst got in. 10 AFC teams had better DVOA than the 2nd best NFC team. It doesn't show up so strongly in the W/L totals, because the AFC teams (especially in the East, North and West) beat up on each other so much, but that was a colossally mismatched season, conference-wise.

by jerry garcia (not verified) :: Tue, 10/30/2012 - 4:58pm

I can certainly understand the beauty of parody, but for God sakes man, most of these teams really stink nowadays. I don't miss the days of dynasties, but when teams like Denver and Miami are top 10... Yikes!

by RickD :: Tue, 10/30/2012 - 6:02pm

Parody ~= parity

by Stats are for losers (not verified) :: Tue, 10/30/2012 - 6:41pm

This is one of my favorite malapropisms, because it can almost always be read in a way that makes unintentional sense.

by TomC :: Tue, 10/30/2012 - 6:47pm

You gotta cut jerry some slack on the vocabulary what with all the drugs and being dead (literally).

by RickD :: Tue, 10/30/2012 - 8:14pm

as long as he's grateful

by LionInAZ :: Thu, 11/01/2012 - 5:12pm

But his fingers are still moving, all nine of them!

by theslothook :: Tue, 10/30/2012 - 5:26pm

I feel like Houston fans have the most reason to gripe with these ratings. That one horrible loss to GB at home and maybe a less than thorough drubbing of jets aside, this team has been pretty strong overall. There are still questions about the receiving core, but they feel just as strong as SF does honestly. Defensively they are better with a better qb, while sf has better o linemen and receivers. Kind of surprised Houston isn't higher.

by tuluse :: Tue, 10/30/2012 - 5:30pm

With apparently the league's worst special teams.

by Danish Denver-Fan :: Tue, 10/30/2012 - 5:34pm

Agreed - gun to my head I'd pick Houston as the best team. In any 25% seems way low, they feel like a +30% team at least.

by Alternator :: Tue, 10/30/2012 - 11:59pm

That one horrible loss is 14.7% of their season, and the other less-than-convincing win against the Jets is another 14.7%. Meanwhile, the top three teams don't really have a disaster, unless you want to count the New England-Arizona replacement ref "Holding is legal" game, and even so, that leaves the top two teams clean.

by MilkmanDanimal :: Tue, 10/30/2012 - 5:41pm

With Carl Nicks now gone for the year, can you just bump Tampa down about ten slots now? Saves time.

by commissionerleaf :: Tue, 10/30/2012 - 5:46pm

No Isaac Sopoaga on offense this year? I haven't watched all that much San Francisco, but I thought he was pretty much a staple of the goal-line packages.

by Karl Cuba :: Tue, 10/30/2012 - 5:57pm

He's been replaced by Will Tukuafu and Demarcus Dobbs.

by greybeard :: Tue, 10/30/2012 - 6:02pm

Will Tukuafu has taken over that role.

by jimbohead :: Tue, 10/30/2012 - 6:02pm

Will Tukuafu has taken his place as the DT that plays FB. No one in the organization has said anything about it that I know of (it's a pretty small sidebar), but I suspect it's to keep Sopaoaga fresh for playing NT.

by tuluse :: Tue, 10/30/2012 - 5:49pm

So Buffalo's run defense is on pace to be the worst in at least 10 years, and by a pretty good margin for run defense (which seems to have a much smaller std deviation than passing). They're 22.2%, with the next worst in the last 10 years being the 08 Rams with 18%. Those Rams went 2-14. The 0-16 2008 Lions also had terrible run defense, 17% for them, and the 2005 Falcons round out the worst of the worst run defenses with 17.8%.

by Jimmy :: Tue, 10/30/2012 - 5:59pm

They are the NFL's equal opportunity defense, allowing 22.3% against the pass and 22.2% against the run.

Run? Pass? Whatever your gameplan, Buffalo's defense is here to help you!

(apologies for misappropriating your joke)

by tuluse :: Tue, 10/30/2012 - 6:07pm

No apologies necessary.

I'm honestly fascinated by how badly the Bills are playing defense. I really wish someone would do an in depth breakdown of what's going on.

by Danish Denver-Fan :: Tue, 10/30/2012 - 6:05pm

And the D-line was supposed to be the STRENGTH of that team...

What happened to Kyle Williams - last year he was getting press as being an elite DT. Regressed? Never that good to begin with? I know it's a team game, but a great DT should be able to keep the run-d off worst ever pace...

by theslothook :: Tue, 10/30/2012 - 6:08pm

Can anyone explain why Buffalo's defense is this bad? I am absolutely not surprised that New Orleans' defense is this horrible- they were horrible last year and had no draft picks to bolster it. Buffalo was bad defensively last year, but they added two players in free agency-one them the best available defensive free agent squarely in his prime. They spent over 50 percent of their draft value on defense last year. And yet, they are one of the worst defenses in the league? Kinda remarkable

by Paddy Pat :: Tue, 10/30/2012 - 7:53pm

My sense is that it's heavily scheme. Wannstedt seems to be doing a godawful job of coaching, putting the players in very vanilla schemes that are easy to target, and failing to get proper coordination. If you look at the film carefully, there often seem to be several players unsure of their assignments on the running plays.

by bcsj (not verified) :: Tue, 10/30/2012 - 9:42pm

Not that I've seen a lot of Bills football (we get the Raiders here - that's enough masochism for any man), but when the Patriots played them, they just pounded it right up the middle. Buffalo's conditioning looked awful. And Mario Williams isn't anything more than a mediocrity.

by BigCheese :: Wed, 10/31/2012 - 12:50am

"We'll be fine."


- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by In_Belichick_We... :: Wed, 10/31/2012 - 9:14am

Since DW is a JJ disciple, he may be running JJ's "attack first, think later" defense which also includes some different gap assignments from traditional 4-3 schemes (like the Dungy/Kiffin scheme, is that what they were running previously?). If the problem is players not grasping new responsibilities, their D should improve soon (they'll learn someday, right?.
There is also the possibility that this scheme just can't stop an NFL offense anymore and DW is the only one that hasn't figured that out yet.

Entitled freeloader

by JoeyHarringtonsPiano :: Tue, 10/30/2012 - 7:54pm

I think Dave Wannestedt and his mustache have something to do with it.

by Ali (not verified) :: Tue, 10/30/2012 - 6:23pm

If we take out their historically bad 300-yard passing 300-yard rushing game against the 49ers, how bad would the Bills defense be?

by JonFrum :: Tue, 10/30/2012 - 6:39pm

With a DVOA defensive ranking of 30th, I doubt one game would turn it. In case you've missed it, people have been banging on that defense all season.

by Ali (not verified) :: Tue, 10/30/2012 - 7:50pm

That's my guess too--just wondering how much of an outlier that game was for the Bills.

by RickD :: Tue, 10/30/2012 - 8:15pm

They gave up 45 points to the Pats in the 2nd half of the prior week's game.

by Tim_quitusingmyname (not verified) :: Wed, 10/31/2012 - 6:41pm

Not much. They just gave up 200 yds rushing last week I believe.

by TomC :: Tue, 10/30/2012 - 6:53pm

I hadn't had a chance to see a full Denver game until Sunday night, and I came away very, very impressed. Their OL played more like a cohesive unit in both phases of the game than any I've seen (including New England). I know it was against a terrible defense, but the NO D-line is only mediocre, not historically awful. And I'm used to seeing NO wideouts running wide open all over the field, so I was also impressed by Denver's secondary, which I had thought was Champ Bailey and a bunch of schmoes.

(The QB is ok, too.)

by RickD :: Tue, 10/30/2012 - 8:17pm

The Saints miss Sean Payton badly.

by Ezzy Black (not verified) :: Tue, 10/30/2012 - 7:09pm

Wait, every sports writer, pundit, commentator, broadcaster, and wannabe in the country have come to a complete consensus that Atlanta sucks precisely because every team they've beaten sucks.

Now you tell me they beat the best team in the league? Is this statistical DVOA thing (come on, the V really stands for voodoo doesn't it?) like those beer commercials where you have to turn the label out? :p

Seriously though, it would seem advantageous to have above-average opponents with an easy schedule (Falcons) than top-notch opponents with a difficult schedule (Broncos). The former pads your strength of schedule while the latter weakens it.

Obviously not much can be done about the above unless you really want to dive into schedule strength way too deep. Perhaps, however, it takes up a bit too much of your calculations?

by Karl Cuba :: Tue, 10/30/2012 - 7:12pm

They're baaaack...

by theslothook :: Tue, 10/30/2012 - 7:34pm

You're ignoring a few facts. Firstly, yes they beat denver, but the score was close and denver had fumbles that went against them that were unlucky. This, actually, was mentioned several times how atlanta has recovered an unsustainably large amount of fumbles while denver has had the opposite effect-having not recovered all but one of theirs. On top of that, they've had some bizarrely close games at home against the raiders that they very easily could have lost and to the 1-6 panthers who they actually SHOULD have lost to if not for some stupid playcalling and a horrible lapse in the secondary.

Ill agree that the falcons might feel like they deserve to be number 1, but case can be made for any team. Everyone has had a bad loss or two and played poorly. The 49ers squarely lost to the vikings and were blown out by the giants at home. The giants have had two losses in the division and had several shaky moments. The Packers have lost some very bizarre games as well(that colts game still has me confused), and the Pats have interspersed dominance with mediocrity(see their games against the cardinals and the seahawks). The broncos have been well documented. So really, what do you expect? its only been 8 weeks.

by johnnyxel :: Tue, 10/30/2012 - 7:59pm

First: don't be the guy that resurrects the FOMBC with Atlanta.

The difference in record and DVOA is all about the Oakland game, and somewhat less to do with the Carolina game, two wins with negative DVOA. The other five wins--even the close ones, Aaron!--were with total DVOA over 30%, and if Atlanta were sitting at 5-2 with 21% total DVOA, and only a two game lead over Tampa Bay, we Falcons fans would think 8th in weighted DVOA would be just fine, even with a decisive (in DVOA terms) win over the #1 team. In sum: if don't you think this team is characterized by the one that threw three interceptions to Oakland, don't worry.

by Peregrine :: Wed, 10/31/2012 - 7:33am

What does FOMBC mean? Must have missed that one.

As a Falcons fan, the DVOA ranking doesn't worry me as much as the components that are adversely affecting the ranking, and that's rush offense and rush defense. The rush defense has been inconsistent but we've got a coaching staff that knows what it's doing in that regard, and they are working on it. The rush offense is another matter. Turner is a slug these days, not running with the same burst or power and it takes him 10 steps to get up to speed. And the OL is too inconsistent, although it was great to see Peter Konz start at RG over Reynolds and play pretty well in Philly. If we can run the ball even at league-average efficiency and protect Ryan (inability to do so was the main reason the Carolina and Oakland games were so close), our offense should improve. But a lot of that is on the OL, because I don't think there are any changes we can make at RB.

by Podge (not verified) :: Wed, 10/31/2012 - 9:01am

FOMBC = Football Outsiders' Message Board Curse - the tendency for a team to lose games the week after a lot of their fans complain about their low DVOA ranking.

by In_Belichick_We... :: Wed, 10/31/2012 - 9:19am

Do you have to be a "true" fan of the team for it to work.

Houston's DVOA is way too low!
Houston's DVOA is way too low!
Houston's DVOA is way too low!
Houston's DVOA is way too low!
Houston's DVOA is way too low!
Houston's DVOA is way too low!

Entitled freeloader

by In_Belichick_We... :: Wed, 10/31/2012 - 9:27am

does it work in reverse? Will the Broncos gain good luck because of all of the people complaining that Denver's DVOA is too high?

Personally, I can't believe the Broncos have lost a single game. Every time I see them play they look like the best team in the NFL. They must really suck when I'm not watching.

by RickD :: Wed, 10/31/2012 - 10:45am

Did you watch the first three quarters of their game in Foxborough?

by In_Belichick_We... :: Wed, 10/31/2012 - 11:26am

I suppose that wasn't very pretty. Denver did lose 3 fumbles though. If they keep the ball on a couple of those, the time of possession is much different. It wasn't like the Pat's D was shutting them down. This game was all about time of possession. Neither defense was very good.
They also seemed extremely ill prepared for the no huddle. I found this very surprising because it didn't look much different from the second half againts the Bills. Did the Denver staff not look at the Pats-Bills tape?

As a Pat fan, I'm not optimistic about a rematch.

Entitled freeloader

by Peregrine :: Wed, 10/31/2012 - 9:45am

Ah. Never have to worry about that with me. I'm very aware of the vagaries of the football gods. I keenly remember that infamous ESPN podcast with Schatz and Simmons in which they finely calibrated that the undefeated Patriots would beat the Giants by about 10 points. And I recall reading Paul M's long-winded diatribes during last season about how the Packers were historically great, etc., ... and then ker-plop. Optimism is fine, but it must be tempered with humility.

The Falcons are a pretty good team that has some obvious strengths but there are some weaknesses, and any given Sunday and all that. Me screaming at Tony Romo on Sunday - which I'll be doing in person - will have more impact on whether we win or lose than how FO ranks us.

by RickD :: Tue, 10/30/2012 - 8:23pm

If I were a Falcons fan, I'd be pleased as punch to be rooting for a team that's 7-0 and almost certainly going to lock down the #1 seed. I wouldn't care about the DVOA rating unless I thought it indicated that the team's record was a bit of an illusion.

If you want to see the Falcons get a bump, root for them to beat a good team convincingly, or to blow out anybody.

by commissionerleaf :: Thu, 11/01/2012 - 1:04am

The Falcons' record is a bit of an illusion. They took three picks from Manning in one quarter, against a Denver team that then stomped them the rest of the game. They beat a terrible Carolina team by two points despite rushing for something close to ten yards a carry. And they almost lost to the (admittedly underrated) Raiders at home in a game where they were outgained by almost 200 yards.

Atlanta does have a cupcake schedule. Apart from two games at New Orleans, which presumably will wake up and start playing 45-40 shootouts every week like the Saints we know and despise, rather than allowing 45 points and not scoring, and the Giants (at home, although against the Giants I'd rather play outdoors), they're pretty much smooth sailing. The Cowboys may be their most difficult game other than that!

by Jobu Nation (not verified) :: Tue, 10/30/2012 - 7:09pm

When Saints opponents drop back to pass, they basically turn into the 2007 Patriots offense.

by Bobby Womack (not verified) :: Tue, 10/30/2012 - 7:53pm

Denver has been down by 20+ points in four games. And has beaten 1 team with a winning record, the good to average at best, Steelers. But are #1?

by RickD :: Tue, 10/30/2012 - 8:31pm

But...they've lost to good teams!

I know, it's kind of irritating that they've jumped up so much over the last two weeks just for beating the collapsing Chargers and the collapsed Saints.

DVOA isn't a perfect system. Small differences in DVOA ratings shouldn't be taken too seriously. But don't take the Broncos lightly. Peyton has been regaining his arm strength and that defense isn't bad at all.

by Purds :: Tue, 10/30/2012 - 10:09pm

To balance it out, Manning's old team, the Colts, beat a bad team and went down in DVOA.

by Mr Shush :: Tue, 10/30/2012 - 11:05pm

As a Texans fan, I am very, very glad that there is not much chance of having to go back to Denver in the playoffs. I'd even possibly rather Foxborough.

by merlinofchaos :: Wed, 10/31/2012 - 1:40am

In the Falcons game, they went down that far due to 3 bad picks thrown by Manning. Since that game he has only thrown one pick (and that pick since is squarely the fault of communication with the fifth receiver on the team and isn't likely to happen again). They were one third down stop (which, by the way, the Falcons got away with a pick on) from having a chance to come back and win. This is also well recognized as being early in the season while Manning was still getting a feel for the game.

In the Texans game, they got behind that far because J.J. Watt and Brian Cushing are GOOD and it took the Denver offense a full half of play to figure out ways of coping with them. They were, again, one third down stop away from having a chance at winning that game. The Texans played a better game than the Falcons did, honestly.

In the Patriots game, a freak fumble by Thomas (seriously, go watch that play. That fumble should have never happened) shifted momentum, and the faster than the guy who talks really fast for commercials hurry up offense gassed the Broncos defense. In fact, that particular game was the worst performance by the defense ever, and it's mostly because the team was so completely surprised by the hurry up that they simply couldn't cope. This was the only game that Denver hasn't actually been within one play of winning, and there was a giant gimmick involved that we can't tell if it would work again. If they played that game a second time, I think it would be much more competetive. And the way the AFC is shaping up, they just might play it again.

In the Chargers game, they went down that far due to two, TWO muffed punts and it just so happens that's the same week they signed a new punt returner, so there was some butterflies going on on special teams. The punt return unit still plays a little nervous, so to be fair that's something we could see happen again. And let's not forget the blade of grass that tackled Decker and the ensuing pick-6 that someone must've mind controlled Manning into throwing. And then, of course, Rivers collapsed like some politically correct army when some other politically correct army marched into their ranks. (I was going to name specific ones but then realized any one I picked would probably be offensive.)

So you have a 4-3 record, but in all the 4 wins against the Steelers, Raiders, Chargers and Saints, the Broncos simply dismantled the other team (and in one game overcoming tremendously poor special teams play, and a pick-6). In 3 losses, 2 of them were 1 play away from a potential comeback win and it's very believable that the Manning-led offense could've done it, and one loss due to a gimmick offense that might work again. (I'd love to see it if it does!)

DVOA likes teams that can come back from big deficits, and DVOA really likes teams that dismantle the other team in wins. When those 3 losses are against other teams that are also in the top 10 in DVOA (currently #3, #7 and #8) and opponent adjustments make those losses look not so bad.

Cap that list off with the Manning offense very clearly needing a few games to get a feel for how it's going to play and get the timing and communication together with the receivers, and you have the makings of a team that is much, much scarier than anyone predicted in August. (Even the homers here were only predicting 10-6, 11-5 for the crazy ones). Looking at the Denver schedule, they *could* sweep their next 9 games (though it seems a little unlikely) and 11-5 (2 more losses) seems like it would almost be disappointing.

by TimK :: Wed, 10/31/2012 - 2:09am

Particularly in view of weighted DVOA Denver seem to have been getting better throughout the season so far. Not only Manning getting more in sync with his teammates on the Offense, but also the Defense seems to have improved communication/organisation as well. Getting Kuper back is also a big boost for the OLine (note that his replacement Manirez was on the most sacks allowed list a week or two ago...).

Number one is a surprise (even, perhaps especially, as a fan), and given the negative opponent adjustments coming up they could well lose that whilst continuing to post good results, but they do look a team with some considerable ability to play both run & pass on both sides of the football.

by rageon :: Wed, 10/31/2012 - 11:21am

Excellent summary.

I think it's in line with my own thoughts on the Broncos: they lost some games largely because of events not likely happen again, and they won the games they should have in convincing fashion. The chargers game was really a combination of both -- falling behind because of odd events and then dramatically outplaying them thereafter.

by Bobby Wommack (not verified) :: Wed, 10/31/2012 - 5:57pm

The fact Denver was getting BLOWN OUT by 20+ in 4 games says just as much about this team as their DVOA or blow out wins against bad opponents.

They came back and beat the terrible Chargers, still a good win given the situation, and scored a bunch of garbage time TDs to make the other 3 games relatively close in the final minute.

I just think everyone is ignoring the fact this team hasn't put together an entire game against a good opponent (unless we think the Steelers are good) and was on the brink of getting embarrassed on four separate occasions.

by baubo (not verified) :: Thu, 11/01/2012 - 12:11am

As a Texans fan who watched the Denver-Houston game, I have to seriously challenge the misleading story regarding that game. Especially when it comes to DVOA.

1. The Texans with their 20-point lead was marching down the field with ease, I think at the Broncos 50-yard line. By all account that was at least a FG or a punt deep into Denver territory. Then, Ben Tate fumbled, and it was recovered(!) by the Broncos. Despite what stats say about Broncos lack of fumble luck, that was a lucky bounce that changed the entire complexity of the game. And I would imagine everyone would say converting a 3rd and 5(what Texans did to put the game away) was much less lucky than recovering a fumble from a team with a 20-pt lead in the 4th quarter.

2. The lone Denver TO in that game was a fumble on the last play of the game, where they had to return the kick for a TD. Seriously, THAT is going to count against their "lack of luck"? And on the flip side, should the Texans feel "luckier than normal" for recovering a fumble on a return where the other team is just wildly lateralling the ball around out of desperation? I know stats can't differentiate between Ben Tate's almost game changing fumble and a worthless one on a desperation play. But the people watching the game should be able to differentiate the two. And this is the sort of thing that should be accounted. There are INTs and fumbles on desperation plays all the time. And these are the sort of things that advanced stats should mitigate these things.

by Danish Denver-Fan :: Thu, 11/01/2012 - 5:25am

Oh come on. If a lost fumble at midfield up 20 with 10:28 to go changes the complexion of a game, you aren't dominating anybody.

BTW: I know hail mary picks are thrown away for the purposes of DVOA, so I'd guess a hook-and-lateral-o-rama fumble is as well.

by The Hypno-Toad :: Thu, 11/01/2012 - 11:48am

This jibes much more strongly with my perception of that game. But I didn't want to say too much because I had to miss the first part of it because some friends of mine decided to get married in the mid-afternoon on a Sunday in football season.

by dmstorm22 :: Thu, 11/01/2012 - 5:51pm

I do think that the Texans are the one team to really outplay Denver. Manning was pressured all game, forced into low-percentage, contested throws (26/52). New England was a bigger win, but that had real fumble luck. The Texans was just an old-fashioned butt-whupping most of the game.

That said, just because the Texans outplayed Denver in their head-to-head doesn't make them automatically better. I would pick the Texans if they play, but I do think Denver is the 3rd or 4th best team in the NFL behind Houston and San Fran.

by Mr Shush :: Thu, 11/01/2012 - 9:13pm

I'd pick the Texans over Denver or San Fran on a neutral field, but I think both those teams would have a much better chance than the Texans against the Patriots or Packers. It's all a bit rock-paper-scissors.

by zenbitz :: Tue, 10/30/2012 - 9:44pm

How odd is it that the 49ers have the best DVOA against the Run, but only 15th in DL stats? Where is the difference made up?

by Nathan :: Tue, 10/30/2012 - 10:26pm

Awesome linebackers?

by jimbohead :: Tue, 10/30/2012 - 10:43pm

Seriously, this is one impressive LB corp. I've heard Willis and Bowman referred to as the best and second best 3-4 MLBs in the league, in no particular order. And Aldon Smith has become very solid in run support on the weak side. Between him and Brooks, they set one helluva edge.

by coboney :: Tue, 10/30/2012 - 10:53pm

Patrick Willis and Navarro Bowman in particular make up any issues there in the middle. Donte Whitner charging around like a maniac helps too.

by tuluse :: Tue, 10/30/2012 - 11:33pm

DL stats aren't adjusted for down and distance or opponent.

by zenbitz :: Wed, 10/31/2012 - 3:42pm

oh, that makes sense. It had to be something like that for the YPA to not be pefectly correlated with Rush DVOA.

by Nick Bradley (not verified) :: Wed, 10/31/2012 - 1:15am

two all-pro inside linebackers?

by zenbitz :: Wed, 10/31/2012 - 3:41pm

Oh, I am being unclear. To me, 2nd level yards (they are 4th) and Open Field Yards (they are 10th) represent LB and Secondary support on the run plays.

In previous years - the Niners have ranked 1/2 in those numbers and their line numbers were better too.

by theslothook :: Tue, 10/30/2012 - 11:59pm

I would like to say, the 49ers are widely cited as the most talented defense in the league. Consider the sheer volume of great players they have: Willis, Bowman, Justin Smith, Aldon Smith, amhad Brooks(whos actually been better than Aldon), Culiver, Rogers, Goldson. I've named nearly their entire defense. And yet, between last year and this year, they aren't an all time great defense. Why?

WHy are the bears playing at a higher level than the 49ers and who exactly do they have that makes them better? So far, just from what I've seen, the headlines are tilman, jennings, peppers, and melton. Now i know they are deep, but does that explain why the bears are FAR AND AWAY the best defense in the league?

Anyone else find it weird that the 49ers are so far behind the bears, by dvoa standards, the bears are nearly twice as dominant!

by Independent George :: Wed, 10/31/2012 - 12:14am

My guess is the sacks, interceptions, and fumbles forced - 23, 16, and 11 for the Bears, 15, 6, and 5 for the Niners.

by BigCheese :: Wed, 10/31/2012 - 1:08am

While that is certainly a big part of it, the Bears defense also has not had anything resembling the shelacking the Giants Offense put on the Niners. Nor did they allow 24 points to the Christian Ponder purple gang.

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by theslothook :: Wed, 10/31/2012 - 1:17am

No I get why the bears are ranked higher, my point is, why, with all the talent the 49ers have, are they not putting up a better dvoa? Again, what big name stars do the bears have that you might have said coming into this year?

This sorta reminds(though really, its not meant to be an insult at any team), but Joey Porter back in 2004 screamed at the media going, "the ravens D is a hollywood defense, while we are the real number 1 defense"

by BigCheese :: Wed, 10/31/2012 - 2:52am

Comming into this year? Well, the four that went to the pro-bowl last year for starters: Urlacher, Briggs, Peppers and Tillman (who might not have been a big-name but was definitely long-deserving of the selection). Melton and Idonije were playing well last season too. And they added a first-round pick to the DL.

I mean, it's not like this is a defense populated by no-names. I find the question rather odd, in fact. The only real surprise is the amazing leap in play that Jennings made this year.

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by theslothook :: Wed, 10/31/2012 - 3:34am

What i meant to really intimate was...is it so much raw talent that makes a great defense, or allocation of talent and the right scheme? On face value, the 49ers FEEL like they have an edge talent wise and yet the results on the field are drastically in the favor of the "lesser talented" defense.

The ultimate point I was making was, to have a great defense, its not just about having better players, but maybe having a better composition, ie- maybe depth at corner and dline outweighs star power at linebacker and d tackle. Maybe a strong rotation on the d line is more valuable than star power in general. And maybe, once you have a strong foundation of pass rushers, the cover 2(which i know the bears don't exactly play) caters best to having a great defense.

These are really just open ended questions, but given the startling difference in on the field play and perceived talent, it does make me wonder about it.

by Independent George :: Wed, 10/31/2012 - 10:00am

It's always iffy to talk about perceptions of talent, but I just don't agree with the idea that the Bears being viewed as the 'lesser talented' defense. Peppers, Briggs, and Urlacher are three of the biggest defenseive names in the league. Maybe it just comes from living in Chicago, but I don't know anyone who doesn't think this is a supremely talented defense; the question marks going into the season was (1) whether they could stay healthy, as their big names are all in their 30s, and (2) how many turnovers Jay Cutler would have. The talent of the defense was never in question.

Personally, I think a lot of their performance thus far is the soft schedule, but they're still an excellent defense regardless, and that really isn't likely going to change much going forward. Houston and Green Bay are the only powerhouse offenses remaining in the schedule, and I think the Bears scheme matches up extremely well with Houston.

by Tim_quitusingmyname (not verified) :: Wed, 10/31/2012 - 6:52pm

As a Packer fan I'd say the fact they haven't beaten the Packers since early 2010 leads me to believe they are not a "supremely talented defense".

I mean, "supreme talent" should reign supreme right?

by Steve in WI :: Wed, 10/31/2012 - 9:47pm

Well, the fact that the Bears' offense has been pretty much godawful against the Packers since 2010 might have something to do with that, too.

The fact that the Packers beat the Bears in week 2 this year by "only" 23-10 in a game where Cutler threw 4 interceptions and the Packers scored a touchdown on an admittedly brilliant fake punt leads me to think that the Bears defense does okay against the Packers.

by BigCheese :: Mon, 11/05/2012 - 1:54am

The fact that the Bears defense held the vaunted Packers offense to 16 points ON THE ROAD ON THURSDAY NIGHT when Cutler threw 4 picks also leads me to think that the statement was completely off the mark.

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by tuluse :: Wed, 10/31/2012 - 11:38am

I think a couple 49ers have regressed from last year. Justin Smith in particular does not seem to be having the same effect.

I think the Bears secondary is significantly better than the 9ers. Goldson and Rodgers are good players, but I don't think any other member of the 49ers secondary would start for the Bears. Maybe not even get play time, as the coaches love DJ Moore as the nickel back.

Charles Tillman is playing the best football of his career at 31. Last year he made the probowl mostly as a lifetime achievement award, but he might deserve a 2nd one this year. Tim Jennings has turned from a washed up journeyman into a guy on pace for 15 picks this year. Chris Conte and Major Wright are both better than average safeties (though mistake prone in Conte's case).

Depth on the defensive line is probably helping too. Through 7 games, the Bears have 7 players with at least 149 snaps, which is over 20 per game. The 49ers have only 2 players with that number of snaps, though Sopoaga comes close with 136. I know the 49ers run a 3-4, but none of their backup linebackers have more than 11 snaps.

by BigCheese :: Mon, 11/05/2012 - 1:52am

OK then, let me rephrase then: You are the only person I have ever heard say that the 49ers have an edge on talent on the 49ers. Where on Earth are you getting that idea from?

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by theslothook :: Wed, 10/31/2012 - 3:39am

Btw, just out of curiosity, how does a guy from mexico become a bears fan? I guess I could ask myself how a guy from the bay area became a colts fan though.

by BigCheese :: Mon, 11/05/2012 - 2:01am

One gets a computer football game which featured the real NFL teams and players (literrally as X's and O's, for defense and offense respectively) where you and oyur opponent (either a human or the computer) selected a play from a VERY extensive play-book (around 50 plays on each side I think), and the computer ran the play. Kind of like NFL Head-Coach for the PS2. And one enjoys this game (which the more I think about it was amazing for its time. I recall you had full rosters and could mess around with depth charts). And then one happens to notice that the absolute best play in the game is the Chicago Bears throwing a screen pass to some guy named Walter Payton.

And so one starts to follow the real NFL, paying attention to this Payton guy. And the year is 1984. So, turns out they have a pretty good defense as well. And the rest is history....

Now, why I also root for the Bulls (Jordan), Cubs (liked the uniforms and name) and Blackhawks (had never seen a game of Hockey, took a list of the teams, covered up the cities, and picked a name I liked, at which point I decided to give in and just root for any and all Chicago teams except the Black Sox), well, coincidence I guess :P

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by bravehoptoad :: Wed, 10/31/2012 - 11:30am

Neither Justin nor Aldon Smith are playing as well this year as last year. Justin may finally be starting to age, and Aldon is playing every down for the first time. He's still not very good in coverage; there are quite a few situations where the 49ers would much rather have had Parys Haralson in the game, had that been an option. Carlos Rogers is also not playing so well this year. Though this is probably Dashon Goldson's best year, he has never played particularly well, though he has played in spectacular fashion, which is probably why he has a reputation. Even Patrick Willis is having an off year, by his standards.

In other words, in a lot of cases those players' reputation is greater than their actual performance this year.

Now i know they are deep, but does that explain why the bears are FAR AND AWAY the best defense in the league?

Depth certainly has something to do with it. The 49ers can wear down late in games because they don't rotate much. The Bears, as I understand it, have a solid rotation all along their D-Line.

I'm wondering how much synergy between Special Teams DVOA and Defensive DVOA there is. The 49ers defense sure looked a lot better last year when they had the #1 special teams in the league than this year, when they have the #18. Chicago's special teams are ranked #3....

by tuluse :: Wed, 10/31/2012 - 11:42am

It has been theorized that special teams performance is indicative of depth. I don't know if there is solid evidence either way, but it's interesting to think about.

Also, check out the snap count data for the Chicago front 7 vs the 49ers, it's a striking difference.

by commissionerleaf :: Thu, 11/01/2012 - 1:09am

Culliver, Rogers and Goldson are not actually that good?

The 49ers have a great front seven that masks the league-average secondary. The Bears have a good-to-great front seven and Tillman and Jennings on the outside.

by Karl Cuba :: Thu, 11/01/2012 - 12:44pm

I have to disagree with your assessment of the 49er defensive backfield. Culliver has been very impressive this year, he's big, fast, physical, has long arms and has been playing the ball much better this year. Brown has continued his high level of play from the end of last year. Rogers has regressed a little from last year (though we wouldn't be saying that if he'd have hung on to that Eli Manning throw) and the safeties have been very good, Goldson might be having his most solid year.

If you'd have been watching the niners you would have seen that the vanilla pass rush scheme hasn't been as impressive as last season and quarterbacks are left scanning the field trying to find receivers in the impressive coverage as often as they are getting harried by the four man rush.

by tuluse :: Thu, 11/01/2012 - 1:40pm

These assessments are too black and white. Yes, the 49er's secondary is good. However, that does not mean it's as good as the Bear's. Tim Jennings has 6 picks in 7 games. That is ridiculous production. Charles Tillman is very possibly having the best year of his career, and held Calvin Johnson to 34 yards.

Check out defense vs types of receivers,

Here are the Bear's ranks: 3, 3, 5, 3, 5.

Here are the 49ers: 5, 12, 8, 17, 19.

From left to right, that's #1 receivers, #2, other, TEs and RBs.

This matches my thoughts from the beginning of the year that there are simply no weaknesses in the Bears defense. I don't think individually, the members are great like a Revis or Reed, but there is just no one to pick on. This of course assumes Tim Jennings is not a mirage, which I'm still not sure of.

The numbers say that perhaps the 49er's linebackers are not as good as their reputation in coverage, or perhaps the scheme is just designed to let TEs and RBs get yards, while shutting down receivers.

by Karl Cuba :: Thu, 11/01/2012 - 2:24pm

I wasn't talking about the Bears, I was talking about the 49ers. At no stage was I making any comparison with the Chicago defense.

As to the rankings you have put up, they're not in a vacuum, the niner pass rush hasn't been that good this year. The front seven is still pretty damn good against the run and they get decent pressure but they are too over reliant on the Smiths, McDonald and Brooks to play the entire game and they get worn down as a result. I hope they'll be a bit fresher after the bye but they might need a bye in the playoffs to have a long run there (which will rely a great deal on managing to beat the Bears). I'm also getting a little exasperated at the niners' unwillingness to tune their game plan for each opponent, they run the same stuff against everyone. I couldn't care less what the numbers say about Willis and Bowman, you just have to watch them to see how good they are. Where are you finding those numbers? I can't find them in premium.

The Bears are able to rotate fresh linemen in all game, it helps, and that defensive scheme has always generated turnovers whenever the line generates a rush.

by tuluse :: Thu, 11/01/2012 - 2:40pm

This is free stuff on the team defense page.

Also, I know you didn't talk about the Bears, but commissionerleaf was responding to a post directly comparing them.

I've always been impressed by Willis when I watch him, but it's clear the 49ers are better at defending receivers than TEs and backs. It could definitely be schematic.

It could also just be small sample size, when Kyle Rudolf makes 3 amazing catches that win the game for the Vikings it can skew things.

by Karl Cuba :: Thu, 11/01/2012 - 3:06pm

How long has that stuff been hiding down there?!?!?

Interestingly while the niners are ranked lower by DVOA against TE's and backs, they are giving up fewer yards to both of them. That pattern could suggest that when the backs are up against the Willis and Bowman they aren't targeted but there are plays to be made when you can isolate a TE or RB against the bigger, slower outside linebackers.

by theslothook :: Thu, 11/01/2012 - 2:49pm

This is to both you and Tuluse,

Its not at all clear to me that Pass dvoa breakdown by receiver type tells you which positions on your defense are playing well and which aren't. Do we assume not being able to guard tight ends is a problem with your linebackers? If that were the case, the 2nd best team is ATL and the best team is BUFFALO! Thats the problem with pass defense, its so damn hard to disentangle which positions are doing well and covering for others because everything is so intertwined. Its probably why Bill Walsh felt like you could play call your way on offense but you could not have sub par players on defense( i bet he was specifically referring to Pass D).

To Karl,

I don't think either smith have been particularly good this year in pass rush(aldon in particular-who's game is built on violent hands and power-but he feels like he gets caught in the wash of blockers). The best pass rusher on the 49ers this season feels like its been brooks by a mile.

The reason I started this thread was precisely something Karl mentioned. Its not just the 49er talent, its partly the scheme. Vic Fangio runs a very basic concept of play assignments and we'll beat you with better athletes. It works, but how much better would it be if it used rex ryan, dom capers, or dare i say mike nolan, or someone more innovative. These teams all exploit wandering style presnap movements by linebackers and safeties-executing different blitz concepts. I feel like the 49ers have a once in 5 years type of defensive talent that isn't being fully exploited.

by Karl Cuba :: Thu, 11/01/2012 - 3:09pm

While I would quite like to see how a Ryan would run the niner defense, I'm not ready to junk Fangio just yet, his approach certainly minimises errors and really helped last year with minimal time to install this scheme.

I'd also like to point out that with 7.5 sacks in eight games the rumours of Aldon Smith's demise have been greatly exaggerated.

by theslothook :: Thu, 11/01/2012 - 3:30pm

Well, i guess I was wrong about Aldon. Strange that the 49ers are so bad in FO's sack rating.

by theslothook :: Thu, 11/01/2012 - 3:30pm

Well, i guess I was wrong about Aldon. Strange that the 49ers are so bad in FO's sack rating.

by tuluse :: Thu, 11/01/2012 - 3:36pm

It's hard to be very critical of a guy who's coaching the #4 defense in the league.

by theslothook :: Thu, 11/01/2012 - 3:53pm

But this gets into the heart of trying to isolate coaching from talent. I like to do this all the time. Pass defense fascinates me because I feel like its so vital to success and yet so difficult to actually understand:

Consider my favorite example, the jets. I feel like the jets have less defensive talent than maybe even New England, but they somehow make do. The fact that this defense is even average is amazing to me and its a testament to Rex Ryan.

The bears- i don't know if its scheme, but my starting hypothesis is- to be a great pass d, you need good pass rush and good corners, if you have that combination, you can defeat most any pass offenses that are receiver reliant(don't know how they'd do against ne, probably would come down to pass rush).

The 49ers i feel have the most talent of any defense maybe in the last 3 years. But has their defense ever been exceptional? The bears are currently twice as good as the 49ers by dvoa. Why? Thats why im subtly or maybe not so, hinting at coaching being possibly the reason.

by Will Allen :: Wed, 10/31/2012 - 12:50am

John Fox is such a good coach that he was able to make Jake Delhomme look like a good qb, albeit intermittanty. He now has one of the best qbs ever, perhaps the smartest qb ever. The Broncos defense has talent. They are really going to be hard to beat in January, especially if they run the table against a soft remaining schedule, and the Texans screw up and drop 3 of their last nine against their soft schedule. Stranger things have happened.

Every time I watch the Lions, they don't look well-coached.

The Vikings are done, absent a playoff intensity performance in a tough Seattle environment, and even that may not be enough. I may have been premature in saying my 3-6 win projection on them was well off the mark. It wouldn't surprise me to see them win one more game, and only mildly if they were to lose them all.

by Nick Bradley (not verified) :: Wed, 10/31/2012 - 1:03am

Denver is #1 because they suck at recovering fumbles. Got it.

by BigCheese :: Wed, 10/31/2012 - 1:16am

Ahhh... the complete missing of the point of the fly-bys. How I've missed you.

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by DragonPie (not verified) :: Wed, 10/31/2012 - 1:20am

Denver is number one because extensive research has shown that the supposed skill of recovering fumbles is based almost entirely on chance. In their writing, sometimes they assume that you've read this research which is probably a false assumption for people new to the site, but it's understandable considering how tedious it gets to repeat it in every article where it's relevant.

by Nick Bradley (not verified) :: Wed, 10/31/2012 - 10:30am

No, I've read the argument; I just think its an invalid assumption. It's as random as a tipped pass that's intercepted, yet that's considered predictive and fumbles aren't.

Chicago has intercepted multiple tipped passes this year, and have returned multiple tipped passes for TDs already this year. They beat the Panthers because the turf under Steve smith gave out. According to FO, that's predictive.

speaking of "non-predictive events", FO believes that a return team does not influence opp. gross punting...so nobody has kicked Away from Devin Hester? The panthers didn't botch a directional kick and end up w a 6 yard punt?

Show me the evidence That fumble recoveries are random and all the stuff the Bears do are "skill", and ill change my mind.

all I've seen are unbacked assertions from the FO staff

by snoopy369 :: Wed, 10/31/2012 - 10:56am

Anybody who's watched Bears games knows that the Bears actually do practice recovering tipped passes, and are actually very good at it.

by Nick Bradley (not verified) :: Wed, 10/31/2012 - 12:48pm

nobody practices recovering fumbles. nonody

by Eddo :: Wed, 10/31/2012 - 2:46pm

That just seems incorrect. Are you saying that in practice, when there are loose balls, player actively avoid falling on them?

I'm not saying that high/low recovery rates are sustainable, just that I'd guess teams *do* practice recovering fumbles.

by Nick Bradley (not verified) :: Wed, 10/31/2012 - 3:58pm

No; I'm being incredibly sarcastic

by RickD :: Wed, 10/31/2012 - 11:02am

The "fumble recovery is random" argument has been done. I've argued extensively that falling on a loose ball is an act of physical skill, not chance.
People who have studied this question more thoroughly than FO have concluded that there are different recovery rates for fumbles based on where the ball is fumbled. Also, there's this abuse of the word "random" which in popular parlance has come to mean "with equal probability for both outcomes." A roulette wheel is random but that doesn't mean that green comes up with equal probability as black and red. Also, if you want to get really abstruse, the space of random distributions includes constant distributions. So, in that sense, saying a physical phenomenon is "random" doesn't say much at all.
What can be said is that teams with fumble rates that deviate significantly from 50% are displaying behavior that it unlikely to persist. The way I'm stating this is a statistically testable thesis (as opposed to the metaphysical questions about what it means for something to be random). And FO has shown that this thesis is supported by the evidence. So, to the extent that you believe this thesis, it's reasonable to think that a team that has done a poor job of recovering fumbles is likely to regress toward the mean in the future.

by tuluse :: Wed, 10/31/2012 - 11:18am

FO also adjusts fumble recovery rate for situation.

by Eddo :: Wed, 10/31/2012 - 11:46am

"People who have studied this question more thoroughly than FO have concluded that there are different recovery rates for fumbles based on where the ball is fumbled. Also, there's this abuse of the word 'random' which in popular parlance has come to mean 'with equal probability for both outcomes.'"

Actually, FO is the only organization I've seen that has published anything regarding fumble rates by type and location.

From this article (http://www.footballoutsiders.com/stat-analysis/2010/fo-basics-regression...): "In football, [regression to the mean] is more complex. Very few things are a 50-50 shot, the way that flipping a coin is. Fumble recoveries are a good example. While what the league refers to as aborted snaps are recovered by the offensive team about 80 percent of the time, the vast majority of fumbles have been proven to be about a 50-50 proposition. Take sacks: In 2009, 94 of the fumbles caused by sacks were recovered by the offense and 93 were recovered by the defense. On running plays, the offense recovered 73 fumbles, while the defense got to 108. That seems like the defense is recovering a larger portion of the fumbles than expected, but the year before, the offense recovered 88 fumbles on running plays and the defense picked up 91."

Your points about the usage of terms like "random" and "luck" are very good, though. That's something that's lost on non-regular FO readers.

by RickD :: Wed, 10/31/2012 - 3:25pm

The article you link to is a bit chatty and less than comprehensive.

This is a bit more of what I'm looking for

I thought I'd read something Brian Burke had done but I cannot find it at the moment.

As for how FO treats fumbles, I knew that they actually did so, but since DVOA is proprietary, it's hard to know exactly how they do so. It has the unfortunate phrase "forcing fumbles was almost entirely a function of luck and not skill" but if you ignore the metaphysics of a statement like that the details are good.

by Independent George :: Wed, 10/31/2012 - 3:36pm

See also this post from Chase Stewart, which includes a comment from a familiar figure:

By the way, this is exactly how we treat fumbles in our stats. Fumbles are not all penalized the same; the penalty is different based on how often we would expect the offense to retain the ball given normal luck.

by Eddo :: Wed, 10/31/2012 - 4:14pm

Thanks to you both for the link. That's a very cool article.

by rageon :: Wed, 10/31/2012 - 11:16am

I just sort of assume that the Bears broke the non-predictive events issue long ago, particularly with respect to special teams play and defensive touchdowns.

by Eddo :: Wed, 10/31/2012 - 11:42am

"Chicago has intercepted multiple tipped passes this year, and have returned multiple tipped passes for TDs already this year. They beat the Panthers because the turf under Steve smith gave out. According to FO, that's predictive."

Aaron has stated many, many times - specifically in regards to this year's Bears team - that defensive touchdowns are absolutely NOT predictive.

In the past, Aaron has also done research on Hester's effect on opposing teams' punts and kickoffs, and has adjusted the Bears' special teams rating accordingly.

by Tim_quitusingmyname (not verified) :: Wed, 10/31/2012 - 6:57pm

Doesn't seem like it would depend "entirely on chance" to me. If you put two guys on the field, one with arms and one without, I would guess that the guy with arms would recover the fumbled ball nearly 100% of the time.

by In_Belichick_We... :: Thu, 11/01/2012 - 7:11am

This is the exact reason that teams almost always draft players with two arms.

Entitled freeloader

by The Hypno-Toad :: Wed, 10/31/2012 - 1:16am

Eurgh. This is very alarming. I don't trust this at all. It seems silly that a team who has gotten three quarters of their wins against teams with double digit negative DVOA could be #1. This feels like a prank on me and other Denver fans.

by nibiyabi :: Wed, 10/31/2012 - 2:31am

1 = v PIT: W 31-19 (solid W against average team))
2 = @ ATL: L 21-27 (close L to good team)
3 = v HOU: L 25-31 (close L to good team)
4 = v OAK: W 37-06 (STOMPED poor team)
5 = @ NEP: L 21-31 (somewhat close L to excellent team)
6 = @ SDC: W 35-24 (solid W against poor team)
7 = BYE
8 = v NOS: W 34-14 (STOMPED below average team)

All 4 of their wins have been by at least 11 points (18.5 on average) against opponents that are -12.6% DVOA on average (controlling for home vs. away).

2 of their losses were by 6 points each against opponents that are 23.0% DVOA on average (1 on the road), and their other loss was by just 10 points on the road against the Patriots, who should play roughly at a 42.7% level at home.

So, they've lost close games against good to excellent teams (2 out of 3 on the road), won solidly against an average team, and stomped every poor to below average team they've come across (albeit mostly at home).

by The Hypno-Toad :: Wed, 10/31/2012 - 2:58am

I understand the basic reasons for why Denver is ranked number 1. But it still seems unusual for a team that just won consecutive games for the first time all season, and hasn't beaten a team that currently has a better than 4% total DVOA to be where they are.
I know that the actual W-L outcome of the game doesn't really bear on discussion of DVOA, and I know that playing a harder schedule increases DVOA (particularly if a team shows reasonably well against that schedule)... But really, this one seems out of place to me. And I'm saying this as someone who wants Denver to be the best team in the league, but I really, really don't see them that way at this point.
Hopefully I'm wrong and Denver's well set up for a deep run in the playoffs, but something about this really feels hinky to me.

by BigCheese :: Wed, 10/31/2012 - 2:58am

And remember, stomping average teams is one of the best indications of a good team according to FO. A stat that I cringed to hear mangled by Matt Williamson on the Football Today podcast as "According to FO and their advanced metrics, historically great teams they.. they... absolutely destroy other teams. That's just what they do."

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by Will Allen :: Wed, 10/31/2012 - 10:11am

I want to look into the question of historically great teams again. Off the top of my head, I just looked at the 1989 Niners. After week 5, they went 13-1, with only one victory by fewer than 10 points. They won their playoff games by 28, 27, and then 45 points. That's pretty much crushing everyone from early October on.

by BigCheese :: Mon, 11/05/2012 - 2:09am

But I wasn't saying that historically great teams don't regularly stomp opponents. They do. And we all know that. We don't need advanced metrics to tell us that! What I'm saying is that he took the notion that consistently stomping bad opponents is far more predictive of a good team than narrowly beating good opponents, and mangled into that statement and then attributed said statement as the result of FO's advanced metrics.

You do agree that the two are VERY different things, right?

- Alvaro

Phil Simms is to analysts what Ryan Leaf is to NFL QBs

by RickD :: Wed, 10/31/2012 - 11:14am

I wouldn't take it too seriously.
Basically what happened is Houston shat the bed against Green Bay. That could be viewed as an unreliable outlier, but DVOA doesn't do that.

Denver doesn't have a bad week. And they have a couple stomps against weak teams, which DVOA also doesn't treat as outliers as much as it could.

by Alex (not verified) :: Wed, 10/31/2012 - 10:22am

What are the chances of getting the "Revenge on Peyton" CHI-DEN Super Bowl listed on the "special Super Bowl matchups" list? Not special enough? It's special to me. :(

by Marko :: Wed, 10/31/2012 - 12:26pm

Many people asked about that a few weeks ago, although usually referring to it as something like "Cutler's Revenge" or "Cutler's and Marshall's Revenge." I think the Cutler and Marshall angle is a better angle than "Revenge on Peyton." They obviously didn't respond to that request.

by Jimmy :: Wed, 10/31/2012 - 12:56pm

If the Bears get to the Superbowl this year and have to play Peyton Manning again I am going to cry.

by bengt (not verified) :: Thu, 11/08/2012 - 8:25am

You can actually easily calculate that yourself by multiplying the 'Conf Win' probabilities. 4.7% puts it right behind the 'Randy Moss reunion special'.

by Eddo :: Wed, 10/31/2012 - 11:40am

I think breaking down Denver's season by games isn't really helping understand their DVOA. Remember, DVOA is a per-play statistic, not a per-game one.

In the case of the Broncos, it could be that the reasons for their losses/early deficits come down to a few scattered big plays, while on the majority of plays, they outclass their opponent. I haven't seen all their games in enough detail, but I would say the Chargers game fits this mold quite well. San Diego scored twice early due to a muffed punt and a fumbled kickoff return by the Broncos.

by Brent :: Thu, 11/01/2012 - 12:46pm

Houston as well. The Texans got up big early, mostly due to 2 big plays. After that, Denver out-played them, but didn't quite have enough to come back and win.

by nat :: Wed, 10/31/2012 - 12:04pm

Denver is ranked just right because they've had a lot of good plays against both run-of-the-mill and good teams, even in their three losses. Obsessing about rankings rather than looking at exact DVOA values is much worse than this. Still, the Patriots did beat them by two scores, and are only slightly behind in DVOA, and that because of opponents adjustments. Go Pats!

Did I do it right?

by merlinofchaos :: Wed, 10/31/2012 - 2:26pm

Er no, it's entirely too reasonable and moderate to have been done right. Shame on you. Or something.

by jimm (not verified) :: Wed, 10/31/2012 - 1:52pm

I saw Tillman's name come up in another context but it brought to mind something I've seen him and others do that I think should be called a penalty and not a good play...that is when they try to punch out the ball with a series of punches...I'm not talking about the one punch at it, I'm talking about when a guy is held up and the defender just starts raining punches in the general vicinity of the ball. How is that not at least 15 yards for unsportsmanlike conduct?

by DisplacedPackerFan :: Wed, 10/31/2012 - 2:34pm

As long as they hit the ball and not just the player I don't see why it should be. Even when it's in someones possession the ball is still live until the whistle blows. Holding up the ball handler and trying to strip the ball comes with the risk of them breaking the tackle or just having it blown dead for end of forward progress. I just don't see the issue.

by akn :: Wed, 10/31/2012 - 5:48pm

Defenders already have to consult a dictionary to figure out when and how they can and cannot tackle various players. You want to add a punch count to that list? Keep the punches away from the head and stop when the whistle blows the play dead. Other than that, anything else seems fine with me.

by DisplacedPackerFan :: Wed, 10/31/2012 - 2:27pm

So I expected the Packers DVOA to drop, though the passing DVOA performance wasn't as poor as I thought it might be. This is a team that is without it's #1 and #2 WR currently, the #1 TE is playing with a shoulder injury, though I'm starting to think that is a blessing in disguise, and the #2 receiving TE (Quarless) has been on IR since training camp. They also didn't have their FB (who really plays like a #2 TE does on a lot of other teams; 40 or so receiving DYAR the last couple years and on pace for that again this year) or their #1 RB, and yes this has affected their running game.

I did some poking. Nelson was on pace to have 441 or so DYAR by the end of the year. He had 517 last year, so he actually wasn't on pace for as big of a drop off as I might have thought, being on pace to be in the top 3 of receiving DYAR most years isn't too bad a drop. Jennings nagging injuries this year limited him in the games he was in, but he has consistently been 230 - 320 range the last 6 years or so, even if he was going to slip some, it's still a loss.

It's good they got to play Jacksonville with all those missing pieces (and they had 3 starters out on the defensive side of the ball too).

Then you add that they still managed a -1% passing DVOA with all those injuries and the running game. -27.2% (against New York) was the worst the run game was last year. In 2010 -32.3% in the Detroit game that Rodgers was knocked out of was the worst they did. They were -57.3% this year vs St. Louis, and -25.8% vs Jacksonville. I'm used to a bad running game, but Nelson still hasn't practiced this week, Jennings is not going to play this week. Yeah Arizona has no offense, but they have a defense (#5 overall, #6 passing DVOA and #12 rushing though there are quite a few good rush D's this year as they are at #12 with a -14.7% DVOA). So I think they can still win and then they get the bye. So I'm not too worried, but I'm feeling this year the injuries are worse than 2010.

I'm so disappointed in Green. I really thought he was going to be a good pick, and I know the line is an issue, but this is worse than Brandon Jackson/DeShawn Wynn as featured backs in early 2007, and while that team had an o-line, Favre wasn't the runner at that point that Rodgers is. But that team only went as low as -35.6% rushing DVOA before Ryan Grant got going. I just don't see the back to back awfulness in the run game that this team has produced.

by Eddy (not verified) :: Wed, 10/31/2012 - 7:14pm

I don't follow other teams as closely, but I've never heard of a teams 1st and 2nd receivers(Jennings, Nelson), 1st RB (Benson), 1st FB (Kuhn), and 2nd TE (Quarless) all being out for the same game. Two guys are still out from horrific leg injuries suffered last season! (Quarless, Sherrod)

The same game was played without their starting S (Woodson), CB (Shields), and LBs (DJ Smith, Bishop).

by ammek :: Thu, 11/01/2012 - 6:56am

But it's the play of the line — the healthiest part of the offense — that is giving most cause for concern.

The Packers are carrying six receivers on their roster. The drop-off in talent and reps limits them in terms of versatility, but they still have enough bodies at WR to be more productive than they were against Jacksonville. What perplexed me was why McCarthy called 23 RB carries with only one healthy back, who was playing poorly behind an offensive line that did him no favors.

by dmstorm22 :: Wed, 10/31/2012 - 8:54pm

Just got power, so here are my (admittedly biased because of my Peyton-love) thought's on Denver being #1.

I can see it. I think Houston has been more impressive, but I can see that their loss was just flat worse than any of Denver's three, and same with San Fran's losses (who I think have a case at #1 as well).

Denver's offense has been great except for a propensity to fumble, and more importantly, not recover those fumbles. Other than that, they are incredibly efficient. Manning has been playing at a level that I haven't seen from him since late 2008. That year, the Colts started 3-4 and rattled off 9 straight. I can see something similar here.

As for the Defense, Miller and Dumervil have continued to be good, but I'm so impressed by their rotational lineman like Wolfe, Vickerson and the rest of the DTs. They have been good against the run mostly, but their pass defense has been really impressive. Only one team had over 220 net passing yards (the Texans). Their defense has contested a ton of passes, and guys like Tony Carter and Mike Adams have been playing great.

I like a couple of conventional stats to give me help in finding teams better or worse than their record. One is passer rating differential, where the Broncos have a 24.3 differential. The Packers have a 29, same with the 49ers. Everyone else, I believe, is below. The other is Yards-per-Play differential, which the Broncos, I believe, are the #2 team gaining 6.1 yards per play and allowing 4.7 (the 49ers are 6.1-4.6). They are so balanced right now, and that schedule is only getting (way) easier.

They have a shot at a bye, and I think if they can avoid New England, a great shot at the Super Bowl.

by commissionerleaf :: Thu, 11/01/2012 - 2:43am

Denver seems likely to notch the #2 or #3 seed in the AFC.

With the #3 seed, they get someone like Miami or Pittsburgh, which I would not anticipate a problem with, in the WC round, and then probably NE in the second bracket. With the #2 seed, they take the bye and get NE after NE clocks same.

The road runs through New England, unless Baltimore wakes up and grabs a seed (kicking NE or DEN to #4, with the game against presumably easier-to-beat Houston (???).

That said, I agree that Houston isn't a terribly impressive #1 seed; their dominant defense may not be what it was without Cushing.

by dmstorm22 :: Thu, 11/01/2012 - 5:52pm

Baltimore is looming as a potential #3 if they can get their act together, with a one game lead and a h2h win over the Pats.

Pittsburgh does have three AFC Losses, so it will be tough for them, but I guess they could too potentiall get the #3 seed if they win that division.

New England's games against Houston and San Fran really loom large.

by Mr Shush :: Sat, 11/10/2012 - 9:07pm

Right. I mean, if the Pats beat the Texans they could still very well take the #1 seed. They could even lose to SF and have a decent chance (maybe the Texans also lose to the Bears and one of the Vikings, Lions and Colts twice)

by JoeyHarringtonsPiano :: Fri, 04/12/2013 - 1:11pm

I followed the link and read that paragraph on Virgil Livers "The Joe Theismann of testicle trauma". I still can't believe the dude went out to play in the next defensive series AFTER his testicle was shattered. I'm cringing in pain just thinking about it.