2011 DVOA Projections

2011 DVOA Projections

by Aaron Schatz

The time has come for our annual preseason DVOA projections, updated from the projections that gave us the season forecasts in Football Outsiders Almanac 2011. There has been some movement up and down, although based on projected wins, our projected playoff teams remain the same as in the book. The Pittsburgh win projection is still way, way ahead of everyone else. That's what happens when you combine the best DVOA projection with the second-easiest projected schedule.

We must start with the requisite link to an explanation of DVOA. For anyone new to our site, DVOA stands for Defense-adjusted Value Over Average and measures a team's performance on every play of the season compared to league average in the same situation, adjusted for opponent. I know a lot of people are coming here from various message boards and this is just going to look like a jumble of pointless numbers. Trust me, there is a method to the madness, and over the past ten seasons past DVOA ratings -- as well as these multivariable-based DVOA projections -- have been a far more accurate predictor of future performance than wins or points.

Offense, defense, and special teams DVOA are all projected separately using a system based on looking at trends for teams over the past decade. The equations include a number of variables based on performance over the past two seasons in different splits (by down, passing vs. rushing, red zone vs. whole field) plus variables based on recent draft history, injury history, offensive and defensive pace, coaching experience, quarterback experience, and even weather. Strength of schedule was then figured based on the average projected total DVOA of all 16 opponents for 2011.

The projections here are updated from Football Outsiders Almanac 2011 based on changes in some of the variables, usually related to injuries, offensive line continuity, and quarterback experience. There are no manual adjustments after the fact, although there are manual adjustments because some of the personnel-related variables can be somewhat subjective. For example, there are variables based on mean or median age of projected starters, but we have to decide for some teams whether "projected starters" incorporates a fullback, two tight ends, or three wide receivers. Here's a look at the teams which have seen major movement since the book, and why:

  • A number of teams decline on offense due to upheaval on the offensive line, including Philadelphia, Dallas, and Cleveland.
  • Oakland, on the other hand, improves because they now plan on starting a more experienced offensive line than originally projected.
  • Washington improves on offense because Rex Grossman has more experience and a better "quarterback with no team variables" projection than John Beck.
  • General changes in projected starters move the Green Bay and St. Louis defensive projections up.

The numbers we are presenting here are exactly what the projection system spit out. As we say every year: "A few of them will look strange to you. A few of them look strange to us." Tomorrow on the site, we'll be running our usual staff subjective projections where we all talk about where we think the projections are wrong. (I'll give you one of mine right now: Dallas.)

You will notice that offensive projections come out in a wider range than defensive projections, primarily because offense performance tends to be easier to predict (and more consistent from year to year) than defensive performance. In addition, the average projection for the entire league is higher than the 0% DVOA "average" because the offensive environment in the league has been so strong in recent years when compared to the entire period used to construct the DVOA baselines.

(Speaking of which, for those of you who follow the intricate details of DVOA design: Behind the scenes this season I'll be working on a new normalized version of DVOA that's going to set things so every year averages 0% again. It turns out this is slightly better for projecting future performance, but we didn't want to start the gigantic project during the offseason with the knowledge that the lockout could end at any time and require us to stop what we were doing to finish the book.)

The first postseason odds report of the 2011 season is also online. The playoff odds report comes out with different win projections than the equation that we use based on the DVOA of each unit, the strength of schedule, and the balance between offense and defense. The method for the postseason odds report plays out the season game by game rather than looking at all 16 games together, and also includes a variable to account for road games with dome or warm-weather teams visiting cold-weather cities after November 1. I've printed both win totals on the table below. The postseason odds report generally projects win totals close to 8-8 for most teams. I suppose going through and playing each game individually, it seems greater chances for upsets. I'd like to do further research on why the equation and the simulation end up with different results, but to be honest, this is pretty much the worst year possible to try to do any kind of serious analysis in the preseason. The domino effect from the lockout still has me behind on a hundred different things, so I'm more focused on making sure we get all the articles, play-by-play breakdown, and game charting scheduled correctly for the first week. Just remember to take the two different win projections into account if you feel like putting down money on "Pittsburgh over 13."

Those of you who believe Hope Springs Eternal will note that every team has at least a six percent odds of making the playoffs, and no team in the NFC is below 10 percent.

Projected division champions are colored in light yellow. Projected wild card teams are colored in light blue.

PIT 31.7% 1 13.3 11.1 16.9% 5 -12.6% 2 2.2% 5 -3.4% 31
NE 28.8% 2 11.3 10.2 30.9% 2 3.7% 16 1.6% 7 4.4% 2
PHI 23.7% 3 11.6 10.3 12.3% 7 -8.5% 4 2.9% 4 -0.5% 18
SD 21.8% 4 10.8 10.2 32.6% 1 11.8% 31 1.1% 8 0.4% 14
NYJ 17.1% 5 9.5 9.3 -2.8% 19 -20.5% 1 -0.5% 21 4.0% 3
BAL 13.8% 6 9.6 9.5 11.4% 8 1.0% 9 3.4% 3 0.9% 12
GB 10.9% 7 10.2 9.4 9.9% 11 -2.2% 7 -1.1% 26 -1.9% 27
NO 9.3% 8 9.8 9.3 19.6% 4 11.2% 30 1.0% 9 -1.3% 24
ATL 8.4% 9 9.7 9.1 13.0% 6 3.6% 15 -1.0% 25 -1.1% 22
MIA 8.2% 10 8.5 8.5 10.6% 9 1.8% 12 -0.7% 23 3.9% 4
HOU 6.5% 11 8.8 8.8 20.0% 3 13.8% 32 0.3% 13 0.3% 15
NYG 6.4% 12 8.5 8.7 5.9% 12 0.1% 8 0.6% 11 1.8% 9
CHI 6.4% 13 8.9 8.8 -6.0% 26 -8.6% 3 3.7% 2 -0.7% 19
JAC 1.5% 14 7.7 7.9 4.6% 15 4.1% 17 1.0% 10 2.4% 5
IND 1.2% 15 7.9 8.0 10.4% 10 9.2% 26 0.0% 17 1.0% 11
CLE -2.3% 16 8.1 7.9 -3.0% 20 4.8% 18 5.5% 1 -2.7% 30
MIN -3.0% 17 8.3 7.8 5.7% 13 7.0% 22 -1.6% 28 -2.0% 28
DET -3.0% 18 8.3 7.7 5.3% 14 6.5% 20 -1.8% 29 -1.4% 26
BUF -3.9% 19 6.8 7.2 -2.7% 18 1.9% 13 0.6% 12 2.4% 6
SF -7.1% 20 7.7 7.7 -5.9% 25 1.4% 10 0.2% 15 -3.8% 32
DAL -7.4% 21 6.8 7.1 1.6% 17 8.9% 24 -0.1% 18 0.7% 13
TB -8.1% 22 7.5 7.1 2.8% 16 10.4% 28 -0.5% 22 -1.0% 20
WAS -10.9% 23 6.8 6.7 -5.6% 24 1.5% 11 -3.9% 32 -0.4% 17
CIN -12.2% 24 6.7 6.8 -6.3% 27 4.9% 19 -0.9% 24 -1.1% 23
KC -12.5% 25 5.4 6.2 -4.7% 22 7.3% 23 -0.5% 20 5.8% 1
STL -14.2% 26 6.6 6.3 -14.8% 31 -3.7% 5 -3.1% 31 -1.4% 25
CAR -14.8% 27 6.3 6.5 -14.1% 30 2.9% 14 2.2% 6 -1.1% 21
OAK -14.9% 28 5.4 6.1 -8.7% 28 6.5% 21 0.2% 14 2.0% 7
DEN -15.2% 29 5.5 6.4 -4.3% 21 10.6% 29 -0.4% 19 1.9% 8
ARI -16.5% 30 6.5 6.8 -4.8% 23 9.2% 25 -2.5% 30 -2.5% 29
TEN -20.1% 31 5.0 5.9 -9.9% 29 10.4% 27 0.1% 16 1.2% 10
SEA -22.5% 32 5.3 6.5 -23.7% 32 -2.4% 6 -1.2% 27 -0.1% 16

The Colts projection listed above assumes that Peyton Manning will come back and start most of the 2011 season. If we replace Manning with Kerry Collins as Colts quarterback, the Colts' offensive projection changes to -3.6%, which would rank 21st in the league. They would be projected for 5.9 wins, and the projections for Houston, Jacksonville, and Tennessee would all go up by about half a win.

UPDATE 3:30pm: OK, so what do we do now that Jacksonville has cut quarterback David Garrard? Actually, Luke McCown doesn't change the projection much. There's not much to go on, but the record he does have playing with the Jaguars offense is actually positive, so he comes out as basically average, not much worse than Garrard. So the Jacksonville projection with McCown drops only slightly, to 2.3% offensive DVOA. They drop from 7.7 mean wins (in the equation) to 7.4 mean wins. We're not going to bother re-running playoff odds with this change until after Week 1 games.


132 comments, Last at 20 Jul 2012, 4:38pm

1 NFC West Represent!

By that I mean a yellow bar in the second table! But hey, that's another year for us Seahawks fans to have nowhere to go but up, possibility-wise, and you have to admit that two years in a row of a sub-500 playoff team (Home game! Home game!) would be pretty awesome.

2 Re: 2011 DVOA Projections

So SD is projected to have an above-average special teams this year, huh?

17 Re: 2011 DVOA Projections

Special teams is the most volatile aspect of a team's performance, so it pretty much had to go up. The rules changes for kickoffs will also eliminate the Chargers' biggest weakness--kickoff distance and kickoff coverage.

122 Re: 2011 DVOA Projections

Certainly there is regression to the mean for coverage units. You can predict year-over-year kicking accuracy on under-40 yard FG attempts and punt distance -- and Kaeding and Scifres are among the very best in both. One is the most accurate kicker in history, and the other lead all Punters in net yardage betweeen 04 and 09 and was second in % downed inside the 20.

If coverage and returns regress to the mean, the overall result is an above average unit.

The Charger's roster turnover has been rather high for all special team non-specialists which suggest it would be very very hard to repeat last years disaster:


Almost full turnover in linebackers, with players responsible for the major ST breakdowns no longer on the team.
Extra WR's that are big ST contributors in coverage or returns. Younger/Faster in the secondary (other than Bob Sanders, who won't be playing on ST).

Linebackers are likely a defensive downgrade -- especially with the loss of Burnett -- but a big ST upgrade.

So I think the overall projection is close:

The ST will be a lot better
The Defense won't be as good*
The Offense will be top 3

* (I highly doubt it will be 31st in the league, but ST DVOA - Defensive DVOA may be ~= 0%). A team with 0% DVOA in ST and Defense combined and Offensive DVOA of near 25% is a very solid team.

27 Re: 2011 DVOA Projections

In reply to by Aaron Brooks' … (not verified)

I also couldn't help but notice that their defense fell from 7th last year to 31st this year. Does the loss of Ron Rivera really do that much damage? I know they had a variance ranked 30th last year, but that coupled with the sharp rise in ST makes it seem unlikely

57 Re: 2011 DVOA Projections

I'm also curious how the Chargers defensive DVOA drops from -6.4% last year to 11.8%? Are Bob Sanders and Takeo Spikes considered weaknesses due to their age? Is the change of system going to make that big of a difference?

82 Re: 2011 DVOA Projections

Change of system?

Its a very similar system. Both coordinators learned the 3-4 from Wade Phillips.

Rivera also ran the cover-2 with the bears, and learned blitz schemes in Philidelphia. So he has a more diverse background and is likely a better coordinator.

I'm not sure what makes the projection go so far down. 'defensive coordinator change reverts to the mean' makes sense, 'defensive coordinator drops team to basement of league' doesn't.

108 Re: 2011 DVOA Projections

Obvs. can't answer re: DVOA, but I expect their defense to be much worse than the last two seasons, esp. in the early games. Their horrible special teams (OLÉ!) meant that in key situations the defense was often not on the field. When it was (e.g. v. Oakland and the last game vs. Denver) they performed poorly. Their run defense, which hasn't been near average in years, has looked even worse in this preseason. The offense hasn't been able to eat clock with the run game; coupled with better special teams, this means the D will be on the field more, and perform much worse. Smith basically let the D-line collapse to attempt to address other areas in the draft, and his picks to rebuild the linebacking corps haven't really panned out (with MLBs Butler and Mouton you can add "yet").

130 Re: 2011 DVOA Projections

You had me until "let the D-line collapse": that's laughable. Antonio Garay, Corey Liuget, Vaughn Martin, Cam Thomas... these guys weren't on the team two years ago, and nowadays they'll combine for a majority of snaps. Garay was a stud last year, and Martin and Liuget are both physical freaks.

I agree the defense may decline. I think Norv sees his offense as being the team's main run defense. But if the offense is firing, opponents will have to pass, and the pressure will come from everywhere.

3 Manning

I would think that his injury should also increase the DVOA/playoff odds for the Browns, Steelers, Buccaneers, Chiefs, and Bengals, which the Colts face in weeks 2-6.

71 Re: Manning

In reply to by Adam B.

The difference between Manning and Collins is projected to be 2.5 wins.


93 Re: Manning

In reply to by Jim Glass (not verified)

A predictive system that swung wildly based on one player's absence would be seriously mis-calibrated. How could DVOA know that the Colts entire team is designed around the strengths of one of the most remarkable quarterbacks in NFL history?

I'd be astonished if the real difference were fewer than 6 wins.

121 Re: Manning

In reply to by John (not verified)

Does that mean you think they'll only win two games?

4 Re: 2011 DVOA Projections

Depressing: Redskins projection actually goes up because they are planning to start Rex Grossman.

5 Re: 2011 DVOA Projections

[Dallas] is clearly ranked [too low] because [Wade Phillips sucks]. [My hatred of Wade Phillips as expressed in predictive numerical format] is way better than this.

Then again, to paraphrase Bart Scott, the defense still can't stop a nosebleed. So maybe you're right.

8 Re: 2011 DVOA Projections

Seriously though, it looks like DVOA is unduly pessimistic on Dallas' offense, again. I actually agreed with it for the first time last preseason, because the O-line looked terrible in preseason and was old with no depth. The O-line seems improved this season and looks better than they did last year at this point.

However, you may actually be optimistic about the defense, which is hard when they're already predicted to be 24th. Your 2011 Dallas defense: Ware is good, Ratliff is decent, everyone else blows. yay!

50 Re: 2011 DVOA Projections

Completely agree about the offense.

I think the progress of the pass defense is the most important and uncertain facet to Dallas' improvement this season (aside from things like injuries and luck). If Jenkins plays like he did most of 2009 (no idea), Lee gives them an ILB that can actually cover (optimistic), and Butler provides a complimentary pass rush threat (somewhat optimistic), then the defense won't be too bad. There are enough average - mediocre players to where I don't feel too bad. Of course who knows what will happen.

6 Re: 2011 DVOA Projections

Wow, Ravens with a positive DVOA? That hasn't happened since 1996. But they're still projected as the 9th best defense, my subjective opinion is they're going to be pretty good. Deeper at D-line, better pass rushers w/ Kindle and Kruger, and much better secondary (their #1 CB last year is now 4th on the depth chart, Reed is fully healthy, no more god-awful coverage by Landry).

53 Re: 2011 DVOA Projections

What leads you to believe Kruger will improve or that Kindle is even close to ready to contribute anything meaningful? It's unlikely Suggs will improve on last year's career season.

60 Re: 2011 DVOA Projections

Well watching the preseason games, Kruger fared pretty well against 1st string tackles, and Kindle looked explosive off the edge. So, my 3rd rate scouting report leads me to believe that.

9 Re: 2011 DVOA Projections

Shocker to me:

1. ARI crazy low projection
2. SF being above teams like TB, WAS or DAL !!
3. Miami that high.

64 Re: 2011 DVOA Projections

SF had the best DVOA in the NFC West last year. They just won fewer games.

I'm surprised by Indy's low DVOA, although I guess I shouldn't be.

25 Re: 2011 DVOA Projections

I thought Miami are close to being ranked that high but I would have thought their strength would be on defense. I just don't see how they achieve a 9th ranked offense with Henne at quarterback. The line's good but Reggie and the rookie aren't great and neither are the recievers. How does that add up to 9th?

As for your second point, how are the niners a worse team that Washington? I can think of two players, Brian Orakpo and the OT Williams, who would clearly be a starter for the 49ers. It's even one of the few matchups where I'd rather have SF's quarterbacks. I'm not sold on Dallas either.

51 Re: 2011 DVOA Projections

re: Miami

"The line's good"??? Have you watched them lately? If anything the OL should be a reason to downgrade, I don't believe the absence of Jake Long somehow justifies whatever crap is going on in the right side.

They've shifted the (until last year) very good RT Vernon Carey to RG, and put Marc "was I supposed to block that guy?" Colombo at RT. It's not pretty, and it's not conducive to effective offense. I feel for Bush/Thomas/random RB.

The receivers are actually a strenght, and I do belive in the new Daboll offense Henne can blossom. I might be smoking weird things, but his physical makeup and the sudden wealth of receiving options make me think of Favre.

-- Go Phins!

58 Re: 2011 DVOA Projections

To be honest I couldn't precisely recall what the Dolphins had done on their line, I knew they'd moved Carey to guard but couldn't remember if it was because they'd signed a better player and upgraded two spots. I was trying to be nice in order to prevent a better informed Miami fan slapping me down for writing off their O-line with its shiny left tackle. Lesson learned, next time I'll bring some nasty.

70 Re: 2011 DVOA Projections

Well, I hope you did learn your lesson :P

In all seriousness, it is a mighty point of concern. If the OL is somehow magically improved from what we saw in the preseason, then I have no qualms about Miami as a postseason team.

Then again, it's a pretty big if.

-- Go Phins!

109 Re: 2011 DVOA Projections

Columbo is done. I'm not sure why the Dolphins picked him up aside from trying to be Cowboys East.

My bet is that by the halfway point, Carey will be back at RT....

119 Re: 2011 DVOA Projections

Totally, completely agreed. I cannot possibly fathom what Sparano sees in this guy. He is so slow now. It seems like everyone can see it, and Sparano HAS to, but he will never admit it. I just don't get it. The sooner Carey moves back to RT the better, in my opinion. Don't know who would be best at RG then, and I don't care really. Any of the Gs on the roster have to be better than Columbo at RT. Well, maybe not John Jerry.

As for the projection, it makes me excited. I definitely think Henne will be better this year, especially with the ability to audible! I also expect the defense to be completely dominant, other than FS. If Jones steps up there, they are top 5 easy.

All we need is a couple breaks to go our way and the playoffs could be in our grasp! And we deserve them after BS like the Steelers stealing the game from us last year.

80 Re: 2011 DVOA Projections

I agree with everything. The offense should be worse than projected on account of the line... well, at least the running game will be terrible. But I, too, believe Henne will have a good year.

And I think the defense will be outstanding.

When you can balance a tack hammer on your head, you will head off your foes with a balanced attack

10 Re: 2011 DVOA Projections

Ummm...question. San Diego's defense finished with the 7th best DVOA last season. This season, projected at 31??? That's a HUUUUUGE dropoff under any circumstances. Anyone got an easy answer where that comes from?

13 Re: 2011 DVOA Projections

No, but my eyes tell me that the numbers are likely pretty accurate. SD has looked terrible in the preseason, even when their first team defense was playing against SF's backups in the last game.

19 Re: 2011 DVOA Projections

That game was played without their starting nose tackle (one of the highest graded defensive linemen in football in 2010) and starting outside linebacker (11 sacks in 2010).

Health will play a big role in the Chargers' defense. If those two above-mentioned players are healthy, and they get any kind of production from Bob Sanders, they should be better than 31st.

28 Re: 2011 DVOA Projections


Their ones held Seattle scoreless for a half and the defense held Dallas to 7 points for an entire game.

They gave up one big play to Larry Fitzgerald and got beat on a reverse for another long run in that game.

They did all this without their best nose tackle and two of their starting linebackers.

11 Re: 2011 DVOA Projections

How much of Pittsburgh's amazing projected improvement in ST is due entirely to the new kickoff rule?

81 Re: 2011 DVOA Projections

Likely a healthy Sepulveda. Before injury, his punting alone was the chief improvement to their special teams ranking. Also, Tomlin and Colbert have been mixing savvy special teams vets like Battle with young linebackers. They've improved their coverage units with that mixture.

91 Re: 2011 DVOA Projections

Consider that Suisham isn't big-legged. Not many kicks of his have been out of the end-zone touchbacks, that means the rule mostly only really changes things by 5 yards for them.

92 Re: 2011 DVOA Projections

On second thought maybe that is the advantage. The Steelers get 5 yards of benefit on kickoffs whereas other kickers, who already got it to the end zone or inside the five see a smaller net gain.

15 Re: 2011 DVOA Projections

The Colts dropping from 15th (8-8) to 21st (6-10) with no Peyton Manning isn't really the Sky Is Falling scenario one would glean from Internet chatter and media bushwah these past weeks.

The DVOA numbers seem to argue that Indy will need to pull games out of their behinds like last year to be a playoff team even with Peyton up near 100% by, say, Week 3 or Week 4. Yet such is the level of mediocrity among the below-average teams that even no Peyton at all doesn't make the Colts "The Worst Team In The League" according to the projections.

Oh well, all speculative. DVOA is simply playing its cards close to its chest, so to speak, given how little data (i.e. zilch) there is to extrapolate from in terms of Colts-minus-Manning. If the team did believe these projections, it probably wouldn't make sense for Manning to rush back still injured and ineffective too early just to try and sneak back into near-contention...

59 Re: 2011 DVOA Projections

DVOA's projection system probably has a very wide confidence interval on the value of Peyton Manning to the Colts offense. The full sample of non-Peyton Manning QBs since he first suited up for the Colts:

121/209 (57.9%) for 1138 yards (5.44 yards per attempt), 6 TD, 5 INT. QB rating 72.6.

This isn't stellar. In fact, it's awful. And much of the good numbers come from Jim Sorgi (whom the Colts no longer have) against atrocious defenses. Sorgi had 3 good games, that account for basically all of the Colts' backups production. He put up 14 points in 2004 vs the Broncos, but his numbers looked good because both TDs were passing. Manning then torched the same defense for 49 points and five TDs the very next week. (It could have been 50; Manning slowed down the offense after they took a 35-3 lead in the first half.) The other two good games Sorgi had were against NFC West defenses in garbage games.

But that doesn't even get to the heart of the problem, which is that the Colts simply don't have a good supporting cast anymore, and they've been asking Manning to drop back nearly 700 times a season.

Since 2008, when the Colts' supporting cast began to wear down a bit, the backups have been 30/58 for 261 yards, 0 TD, 2 INT. Hovering between 4 and 5 yards per attempt. It's too small a sample size to draw good inferences, and obviously you can argue that Curtis Painter is simply the worst backup in the league.

We know that healthy Manning + the Colts' weak supporting cast is good for about 9-10 wins these days.

We don't know what they look like without Manning. They could be a 1-15 team, or a 6-10 team, or even a 7-9 team. It's a pretty wide interval. But they sure as hell won't be an 8-8 or 9-7 team.

87 Re: 2011 DVOA Projections

That's why I think if Peyton misses, say, six weeks and the Colts got 1-5 or 2-4 during that time there's no reason not to let Collins run out the string for another six weeks or for that matter the whole season. Peyton isn't going to come back from that kind of layoff and run off a streak of four or five wins to get them back into contention. Not when they're basically an 8-8 or 9-7 team with him playing the whole season.

If he can't be playing like Peyton Manning by about Week 4 they might as well give him the season off to get healthy and at least they'll be drafting like a 25th place team and not like a playoff participant.

Of course that presume Collings stays healthy...

88 Re: 2011 DVOA Projections

Yeah, I can't think of a team deliberately tanking like that, nor do I think Goodell would allow it.

94 Re: 2011 DVOA Projections

When you're talking about nerve problems and the potential for career-threatening damage if he's rushed back, do you really think Goodell would blink an eye at the Colts IRing Manning, no matter what the team record may be at that point?

96 Re: 2011 DVOA Projections

If the damage was that bad, I think that'd probably be the best move for the Colts even without considering the draft. How much better than Kerry Collins is a significantly injured Manning? How much time do you shave off his career if you play him injured?

16 Re: 2011 DVOA Projections

The only team I watch consistently is Green Bay, and I've gotta say that with the talent they have, winding up the year with <+20% Total DVOA will be a disappointment. The predictions say an offense outside the top 10 and a defense outside the top 5 in DVOA - short of another injury plague, or long stretches without Rodgers or Matthews, I just don't see it. The squad is loaded and, with a few exceptions, quite young.

40 Re: 2011 DVOA Projections

I'll skip zlionsfan templating your response and just cut to the chase.

I think they're allowing for that youth to project some inconsistency. I seem to remember that same Super Bowl-winning squad scoring a whopping 6 points against Detroit in one of their games last year. They're still favorites to win the North and finally get Rodgers in Lambeau for January football.

84 Re: 2011 DVOA Projections

I'm a Packerfan as well but I don't forget that we were a shoestring tackle away from losing to PHI and 2 weeks later Todd Collins became the MVP of all time.

105 Re: 2011 DVOA Projections

The offensive projection likely foresees:

- no great improvement in the running game
- Rodgers throwing a few more interceptions than last year
- more fumbles (the Packers' RBs have to fumble at some point)
- a slight decline in the offensive line, especially in pass protection, given that Chad Clifton is 35 and has rarely made it through an entire season, and there's a new guard beside him
- continuing short-yardage travails, now that the team has gone from 3 FBs to one (the sometime ball-carrier).

124 Re: 2011 DVOA Projections

Ryan Grant has averaged ~2 fumbles a year (2010 excluded), and Rodgers was actually slightly above his career interception rate in 2010. In 2008-9 GB was 17th and 14th in the league in rushing yards per game (and 17th and 2nd in rushing DVOA), it's pretty likely that with Grant returning they'll be back to league-average in rushing. With Finley returning it's hard to comprehend a projection that has the Packers offense being worse than last year.

I can only assume that someone mistakenly plugged in TJ Lang's 2009 performance at LT and assumed he could allow just as many sacks while playing guard. Since I assume he'll be getting help from Wells on damn near every snap (because Sitton doesn't need any) that seems unlikely.

126 Re: 2011 DVOA Projections

I have two potential concerns with the running game. Clifton's decline has been most rapid in the running game, and while I think Lang is an upgrade in the run game over College, Clifton may be the worst run blocking LT in the game now. That being said I think all the TE's are better at run blocking than they were last year.

My other concern is 3rd downs and it may not be the running game per se, but the running backs on 3rd down. Brandon Jackson was good at that job, I believe that Green and even Starks can learn it, but Jackson still took about a 20 games to get where he was. That role is downgrade this year. That being said the 1st / 2nd down running game should be upgraded and the short yardage game might be a bit better too.

The other thing I think might actually turn around this year is special teams. If they don't, then Slocum better lose his job as this is by far the most talent he has had to work with. Cobb / Green are upgrades at kick return, even though there will be fewer of those in general. I also think Cobb is an upgrade of Tramon Williams on punt returns. The players on the coverage unit are better than what they had before too I feel. Masthay was a top 5 punter from week 8 onwards.

If you look at their expected points breakdown from last year, it's clear that kick offs and kick returns were by far their worst areas, followed by punt returns.

FG/XP...Kick...Kick R...Punt...Punt R

As mentioned rules changes and player upgrades should impact their lowest areas the most and that Punt number could be double digits this year. It was around -8 at one point least year (around week 6 I recall) and just kept climbing after Masthay finally put it together.

I actually think the Packers might end up with a positive ST DVOA, which last happened in 2007 (it was negative every other year from 2005 onward).

129 Re: 2011 DVOA Projections

Well I was right about Cobb, but the coverage teams I was not.

Masthay still punted well and I still think Bush downed that ball at the one, it looked like his knee was down before his feet crossed the goal line.

So I take back my thoughts of a positive DVOA on special teams.

Yes I have the right to over react to one game. :)

18 Re: 2011 DVOA Projections

The two teams I don't buy at the top are PHI and NYJ.

PHI I think will still put up points and are a "dangerous" team, but their O-Line is awful. Also, their DTs and LBs are suspect, and I imagine a good between the tackles run-based scheme would tear them apart (I think NYG and WAS will give them fits in this regard). Also, if Vick gets hurt, that team goes south quick.

NYJ with the #1 Defense? I'm not buying it. They still lack a consistent pass rush and with a (seemingly) hard schedule I see a lot of low-scoring affairs that leaves them at 9-7 when the dust settles.

49 Re: 2011 DVOA Projections

Gotta disagree. I think teams like Pittsburgh, which can mix it up and have patience on offense, will give Philly fits - as Pittsburgh did in the preseason.

But then again, Reid is excellent at making key adjustments from prior bad performance. The ability to beat Reid has always been "Can I outcoach him?" during the course of a game, and a few coaches are good at that.

I agree the O Line is questionable, but with either Young or Vick that could be a blessing, since it lets both of them do what they do well - move with the ball and make a decision to throw or run. Vick's rushed decision making in the Pittsburgh game was (to me) more the result of his new "run second, throw first" focus, but will change as the year progresses. In a real game, I can see him not having all those INTs and actually holding onto the ball and picking up yardage.

It will be intriguing to see how teams deal with the new look Philly defense. Something tells me "not well". But stranger things have happened.

22 Re: 2011 DVOA Projections

Granted anything could happen, but seeing Miami with a Higher Ofensive DVOA than GB is pretty stunning. I have a pretty difficult time envisioning many scenarios where that happens that do not inlvolve kidnappings or plane crashes

23 Re: 2011 DVOA Projections

Just look at how quickly the defensive projections fall off. The Jets at -20% at first, and within 5 ordinal rankings we're down to -3.7% with the Rams.

24 Re: 2011 DVOA Projections

I haven't bought the Almanac yet, but from reading the Viking's chapter and the hints given out, I find Chicago's projection to be surprisingly optimistic.

Although, I think the offense is going to be better than 26th, while the defense will probably be worse than 3rd.

26 Re: 2011 DVOA Projections

Six defenses with negative DVOA? Out of 32. Check the spreadsheet, somebody's range got screwed up.

30 Re: 2011 DVOA Projections

There actually are seven (the Packers project in seventh place, at -2.2% defensive DVOA).

Remember, too, that the baseline for DVOA takes many years into account, so the projections might be saying that this year will skew towards offense in a relatively extreme way.

31 Re: 2011 DVOA Projections

Not to worry. DVOA is known to have an approximately 3% bias towards offenses, mostly (but not entirely) because the game is evolving to favor offense. It's normal to have most teams with positive Def DVOA. Remember: DVOA is based on historical averages - I forget how many years are averaged. Anyone?

If you look at the actual values in the lists, you'll see that Defensive DVOA is a fairly straight-line distribution with a median of about 4% - but with four teams well off the straight line: the Jets, Steelers, Bears, and Eagles are projected to have unusually good defenses. The Jets especially are expected to be great on defense.

33 Re: 2011 DVOA Projections

From your mouth to Gods ear, Aaron. Eagles as NFCE champs, with the Giants and Cowboys out of the playoffs? And Dallas with only 7 wins? Man, I'm no statistics expert, but I'll take those results! Especially after watching the O-line flounder so much in pre-season. If only I were blissfully unaware of the birds not living up to previous years' DVOA expectations, I'd be really happy.

34 Re: 2011 DVOA Projections

Who needs defense? The 3 worst defenses are supposed to win 3 of the 8 divisions!

35 Re: 2011 DVOA Projections

Miami is projected to have a Top 10 offense! So I guess Chad Henne is going to figure it out this year, or is Reggie Bush going to get 2000 yards from scrimmage?

83 Re: 2011 DVOA Projections

Henne has been a changed man this preseason.

When you can balance a tack hammer on your head, you will head off your foes with a balanced attack

36 Re: 2011 DVOA Projections

And Jacksonville screw the predictions up after only a few hours by releasing their starting QB. It is like they were trying to screw with you.

41 Re: 2011 DVOA Projections

The Vikings win projection seems high to me, possibly because my guess is that McNabb only starts 8 games at most. One reason I might be too pessimistic is because there might be a huge upgrade in offensive line coaching; I think the worst aspect of the Childress area, outside of the failure to find a longer term solution at qb, was the coaching of the offensive line. It's possible that better preperation for that unit will keep McNabb healthy, and that will make eight or even nine wins a non-miraculous event. 4 to 6 wins would not shock me in the least, however.

45 Re: 2011 DVOA Projections

I tend to agree with you, Will. I guess it's good to be optimisitc (in your position) regarding the Vikings' offensive line, but from where I sit in Chicago, I feel like the Viking fans are about to experience the same frustration Bears fans have over the past three years.

I could definitely see things shaking out well for the Vikings, but I can't say I find it likely at all.

48 Re: 2011 DVOA Projections

Oh, I'm not optimistic at all; it's just that I think the Vikings offensive line coaching in the Childress era was horrible, and there is a chance that a change in coaching will give an opportunity for McNabb to stay on the field.

42 Re: 2011 DVOA Projections

With a lot of wishful thinking, a playoff road consisting in home games versus Jets (via Jets over Texans) and Chargers (winning over Ravens, then Pats), I REALLY love the chances for another shot at the Lombardi for the Steelers. Can we forfeit the season and go right into January?

On the NFC, a GB @ PHI for the championship game would be great, and the ensuir SB could either be a chance for some revenge or the PA Bowl.

I'll but it!!!

43 Re: 2011 DVOA Projections

Paragraph now added above to account for the David Garrard release.

46 Re: 2011 DVOA Projections

For all the Hate that has been coming out of Football Outsiders and the Statheads towards the Bears...you have them as a playoff team?

Did I miss something?

54 Re: 2011 DVOA Projections

Can anyone explain to me Tennessee precipitous drop? They lost Babin on D and dropped from 8th to 27th. 11th to 31st overall. With an expected improvement at QB. Tulloch to Ruud is lateral, at worst. A lot of these projections defy reason. Ie. Steelers special teams. It's not like they improved anywhere. They're pretty bad everywhere on ST. The new KO rules will help them be less bad, but I'm dumbfounded as to how they could be predicted to be good.

55 Re: 2011 DVOA Projections

I wonder how much some specific predictions are just the result of applying a formula that has a lot of assumptions and uncertainty built-in? Remember, it's designed to be as accurate as possible across all teams, but doesn't have any actual "intelligence" built-in. For instance, if the projections nailed 30 teams perfectly, but was wildly wrong on the other 2 for seemingly no good reason, it would still be a fantastic system. The system wouldn't be able to look at the numbers and see that it was out to lunch on 2 teams. It's just, on average across all teams, going to be closer than other possible predictors like previous won-loss or points scored/allowed.

That means when you looks at the tables, there may be things that seem wrong, but on average across all teams these projections will come closer to correct than other things. Aaron acknowledges that in the introduction, as some of the numbers don't look "right" to him either. Steeler's special teams and SD defense are likely two of these.

86 Re: 2011 DVOA Projections

My question is what input could possibly spit out the Steelers as a top 5 ST? You have to seriously question the model if many of the prediction results are completely incompatible with unit history and roster moves.

90 Re: 2011 DVOA Projections

Suisham was pretty darn accurate last season, and Brown appears to be a solidly above average if not spectacular returner.

56 Re: 2011 DVOA Projections

The projections for the Jets offense and defense strike me as being in the right ballpark, if somewhat pessimistic for the offense and optimistic for the defense. I'd be surprised if their ST DVOA is negative, however. The Jets haven't had a negative ST since 2000 (the year before Westoff became the coordinator) and have been in the top ten the last 5 years and 8 of the 10 years that Westoff has been in charge. ST seems like an area that's influenced by the quality of the coaching more than the players and the Jets have been lucky to have one of the better coaches for a decade.

65 Re: 2011 DVOA Projections

I find the Indy projection interesting (7.9 wins to 5.9 without Manning). I debated here and on other football sites that a star QB was probably worth 1-2 wins tops to a team over an average QB. The guy who runs advanced NFL Stats suggested it was something like 5-6 games if I recall correctly (Romo vs Kitna). Most of the people I debated seem to fall into the 3-4 games camp.

I've always felt individual players were grossly overrated by the media, fans and NFL ownership. I think the true stars of the game are the managers. If I ran the Vikings I'd offer Indy just about anything for Polian; give them Peterson, Harvin, and toss in a first. Guys like him seem so undervalued it's crazy. One can go on and on about Peyton Manning, but if SD took Manning I'm confident Polian would still have found a way to be a top team every year. He'd just do it a different way, he'd find a way to be a top team.

79 Re: 2011 DVOA Projections

Individual players do tend to be overrated. Quarterbacks aren't, and if you look at how Polian has divided up his salary cap over the years, he sure doesn't think the difference between Manning and an average qb is two games.

110 Re: 2011 DVOA Projections

I agree with the broader point about the value of a GM, but the Colts' personnel acquisition in recent years has not been too hot. It may be that Polian is suffering from age-related decline . . .

67 Re: 2011 DVOA Projections

So the Saints were 9th in defensive DVOA last year, have what is arguably an easier schedule, lost no key contributors and added a bunch of talent... and they'll be 30th? Mmmmkay.

89 Re: 2011 DVOA Projections

Also curious how Seattle is going to end up with the 6th-best defense, seeing as how their non-NFC West schedule isn't made up of college teams.

68 Cowboys Over Philly

The Dallas Cowboys will win the NFC East and pluck the Eagles on their way to top on the NFC East.

73 Re: 2011 DVOA Projections

You say your subjective analysis will disagree with how Dallas was placed. Subjectively, I have no idea whether you're going to talk about the projection being too low or too high, which I guess means it looks pretty good to me.

97 Re: 2011 DVOA Projections

it's obviously low. two things: it's usually low (last year was the first time in half a decade they didn't drastically underestimate the cowboys' final win total in the preseason projections), and, they almost won seven games last year without their starting quarterback for 3/4 of the season. i think they'll be good (they have an easy schedule), but not elite

76 Re: 2011 DVOA Projections

I thought Denver would be just improved enough to not be in the bottom 5, but these projections give me hope that if a couple of bounces go their way (and by that, I mean recovered by the opponent), the Broncos could commandeer the Suck for Luck bandwagon.

78 Re: 2011 DVOA Projections

#1 Defense? I knew DVOA would come around and agree with Rex

95 Re: 2011 DVOA Projections

Overall rankings seem good. The breakdowns seem a little off.

Offense seems mostly solid, apart from a few flukes: GB looks low, Chicago looks way low, Miami looks ridiculously high.

Defense though looks kind of weird across the board. The standout example of weirdness is of course SD's defense. While not as weird, I expect NO to be way better than #30. Houston's an odd one but they can't possibly be last this year – that defies all logic. TB also seems to be rated much lower than I'd expect. And Seattle seems really high.

Overall I wish we'd got a table published comparing last year's final ranks with this year's predictions, so it would be easy to tell which of these things that seem odd are actually odd and which are just knee-jerk reactions. Too tired to look it up and compare manually now.

"When you absolutely don't know what to do any more, then it's time to panic." - Johann van der Wiel

111 Re: 2011 DVOA Projections

The Texans defense has defied all logic for half a decade. You think they can't keep doing it? I hope you're right, but I'm not going to feel confident about it until I see it on the field.

98 Re: 2011 DVOA Projections

Pittsburgh and Baltimore are both really old on defense, and one of these years, one or both of them will crater. Given their opponents, this probably won't be the year that happens to either of them

100 Re: 2011 DVOA Projections

Ravens are old on D?

Ray is 36
Reed is 33
J. Johnson, C. Redding are 30 but have a stable of talented young players behind them. Probably won't be starting by the end of the year once their understudies get their feet wet.
Ayanbadejo is the only other defender over 30, and he's mostly a special teamer.

Seriously, people have been saying the Ravens D has been getting old since 2004, when will this myth die?

101 Re: 2011 DVOA Projections

I don't think the old defense will fall off a cliff theory holds water when the old players are still playing because they win the positional battle against competition. I know the Steelers' players so I'll run through them:

DE Aaron Smith is 35, but Ziggy Hood, 24, has been very good in backup and would start on just about any other 3-4 line.

DE Brett Keisel is 32, which probably gives him a year or two before declining. And Cameron Heyward was just taken in round 1 to replace him when the time comes. Early to tell, but he'll likely work out.

NT Casey Hampton is 34, and backup Chris Hoke is 35. They definitely need a backup plan there, but NT being just about entirely about strength on the strength vs. speed spectrum, is a position where you should expect older guys to remain competitive longer.

OLB James Harrison is 33, but the other OLB, LaMarr Woodley, is only 26. Jason Worilds looks like he made a lot of progress from last year to this year, but the jury's still out on him. But any replacement for Harrison no matter the age is a downgrade.

ILB James Farrior is 36, but the other ILB, Lawrence Timmons is 25, and Stevenson Sylvester has looked pretty good this preseason.

CB Ike Taylor is 31, but he's still one of the fastest players in the NFL, and he's a conditioning freak likely to last another 3 or 4 years. Other than Taylor, CB is a weakness for the Steelers; losing any of the other vets to age is relatively easy to replace. Hopefully at least one of Keenan Lewis (25), Curtis Brown (22), or Cortez Allen (22) becomes a starter.

FS Ryan Clark is 31. But he's not that great to begin with, and Ryan Mundy, 26, has been a serviceable backup. I'm not saying he's the solution long term, but the dropoff isn't insurmountable.

SS Troy Polamalu is 30. He doesn't seem to be slowing down, but he does seem to be fighting mounting injury pressure. He's the greatest strong safety of all time and clearly irreplaceable.

The hidden benefit of old players is that they play a long time, which means you can cherry pick and develop replacements, who will be more likely to be good players who play a long time in a virtuous cycle.

Timmons/Sylvester is likely going to end up a better combo than Farrior/Foote. And there's plenty of room for the CBs to be an improvement. Harrison and Polamalu are the only two to worry about.

102 Re: 2011 DVOA Projections

I understand that they have players groomed to replace their aging players, but, much like with Kolb behind McNabb, sometimes the backup projected to step in isn't up to snuff. Further, they were average without Polamalu the past two years, so it wouldn't necessarily take much for them to slip

107 Re: 2011 DVOA Projections

Isn't the fact that the old guys are still on the roster taking up cap room evidence that the team does not deem the younger backups to be sufficient replacements at this point?

117 Re: 2011 DVOA Projections

At this point, yes. In a year or two, no. Most Steeler defensive players spend a year or two red-shirting. Notable exceptions are Hampton (who plays a fairly uncompleted position), Clark (who was a FA signing), and Farrier (another FA). Smith, Keisel, Harrison, Woodley, Timmons, Polamalu, Taylor all started with a red-shirt season or two on the bench.

115 Re: 2011 DVOA Projections

33 is ancient for a safety? Tell that to Rod Woodson, Brian Dawkins, Rodney Harrison, etc. There's plenty of comps suggesting safeties can play into their mid to late 30's and remain excellent/good.

And Ray isn't really a "difference maker" - not sure what that means - who relies on athleticism to make plays, he's more of an on the field coach who's value lies in his ability to diagnose plays and call audibles. And Seau played until he was 40, so there is a precedent for old ILB remaining effective into their late 30s.

106 Re: strength of schedule

What am I missing? The difference between the Ravens' and Browns' strength of schedules is more than three percentage points of DVOA. Yet I can't get that figure from the table.

Outside of the 12 games against common opponents, Baltimore plays Cleveland twice (a cumulative -4.6% points of DVOA) plus the Jets and Chargers (+38.9%), total +34.3%.

Cleveland, meanwhile, plays Baltimore twice (+27.6%), plus Miami and Oakland (-6.7%), total +20.9%.

That's a cumulative difference of 13.4% DVOA, or less than 1 percentage point per game.

112 Re: 2011 DVOA Projections

Two teams really jump out at me as WTF??? That's probably two more than usual. The Bears and The Rams for opposite reasons. I'm a big believer in looking at the offensive line. Good teams tend to have good lines and bad teams tend to have bad lines (standard caveats apply, but will likley be ignored by those who disagree)

The Bears OL is in shambles. The Rams OL should be above average and has a chance to be very good. In both cases, I think that's also a statement you could make about the entire team.

113 Re: 2011 DVOA Projections

I really think you're underselling the Bears' OL. Last year, it was in shambles (and even so, they did win the division and host the NFC Championship); two years ago, it could barely be called an offensive line.

This year? Webb looks like a passable LT. Carimi looks like he'll be at least an average RT. Losing Olin Kreutz, in my opinion, is addition by subtraction (though the interior line is still below average).

The line won't be a strength, but it shouldn't submarine the team like it had a tendency to do the last two years.

(And even so, the Bears are projected for a -6.0% offensive DVOA, 26th in the league. And that's with a competent (at least) quarterback.)

116 Re: 2011 DVOA Projections

I didn't like Carimi coming out of college. I don't think he gets his hips low enough, and I think he plays stiff. Yes, these are things that can be corrected with coaching, but that takes time and results aren't guaranteed.

I don't see the Bears being cover-your-eyes awful, but I do think they'll be .500 or slightly worse when it's all said and done - and that might put them in last place in the Central.

120 Re: 2011 DVOA Projections

The Bears offense (terrible line and all) ranked just above league average post bye week last year. I think the line will be at least a little better this year; I would expect fewer mental mistakes and the line as a whole is bigger and more athletic. I would rate the depth quite a bit better than last year too (actually I would say the Bears have pretty good depth on the line this year).

123 Re: 2011 DVOA Projections

Can someone say a couple words about why the SD defense is projected 31st in DVOA? I thought they were pretty dang good last year and didn't lose any major contributors?

125 Re: 2011 DVOA Projections

In reply to by Anonymous Coward (not verified)

Because the Chargers had a cupcake schedule last year, one of the easiest in the league (played NFC West), which made their offensive and defensive numbers look better than what they really were (No. 1 in both categories the media continually reminds us!). They only faced 3 playoff teams last year (Pats, Colts, Chiefs).

Its not about the numbers, its about who they played.

127 Re: 2011 DVOA Projections

That doesn't affect DVOA, which is already adjusted to account for quality of opposition. Even though the "D" stands for "Defense", it applies just as much to opposing offenses as it does to opposing defenses.

131 Re: 2011 DVOA Projections

Actually, the "D" stands for "Defense-adjusted". I guess it's because VOA was originally an offensive stat, and then was adjusted for the quality of the the defenses faced.

Techincally, defensive DVOA should be called OVOA. But then DVOA would apply to offenses and OVOA would apply to defenses, and you'd need an STVOA and it would be more confusing.

132 Re: 2011 DVOA Projections

These DVOA rating were completely on the money! Pittsburgh and Philadelphia? Like QBR ratings the other hack site does. Green Bay will win it all in 2011. New England will win it all in 2011, and this year. New England will go undefeated in 2012? Really Pete Prisco/CBS?