Writers of Pro Football Prospectus 2008

Most Recent FO Features


» 2018 Free Agency Cost-Benefit Analysis

Is Kirk Cousins the best free-agent quarterback in recent memory? Should Trumaine Johnson or Malcolm Butler have gotten the larger contract? And what makes a free-agent contract good or bad, anyway?

03 Sep 2014

2014 DVOA Projections

by Aaron Schatz

(Correction: Please note that the Super Bowl odds below mistakenly give the NFC team home-field advantage in the Super Bowl. Corrected odds are now listed on our playoff odds page.)

The time has come for our annual preseason DVOA projections, updated from the projections that gave us the season forecasts in Football Outsiders Almanac 2014

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 may be 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 dozen 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 and game location. (One of the hidden factors explaining why Buffalo's projection is better than conventional wisdom: The Bills have eight home games instead of seven this year.) Strength of schedule was then figured based on the average projected total DVOA of all 16 opponents for 2014, rather than based on last year's performance.

A number of teams have moved up or down since the projections we did for Football Outsiders Almanac 2014. Sometimes, the reasons for this will be obvious, such as a major injury or a quarterback change. Other times, the variable involved is less obvious, usually relating to the average age of certain units or experience along the offensive line. Six teams saw their projected DVOA improve by at least 1.0% since the numbers I ran for the book, and six saw their projection decline by at least 1.0%.

  • Starting with the biggest improvement: Miami, New Orleans, Jacksonville, Houston, Tennessee, and Washington.
  • Starting with the biggest decline: Buffalo, St. Louis, New York Giants, New York Jets, Arizona, and Cleveland.

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." As always, the offensive projections come out in a wider range than defensive projections because offense performance tends to be easier to predict (and more consistent from year to year) than defensive performance. If you are looking for subjective projections, Thursday we will be running our usual staff predictions article where we all talk about where we think the numbers are wrong.

The first playoff odds report of the 2014 season is also online, and I've added the playoff odds and Super Bowl championship odds to the table below. The simulation we do for the playoff odds report is much smaller than the simulation we do in Football Outsiders Almanac 2014, so the results are going to end up even more conservative. The season is simulated using only one set of mean projected DVOA ratings, not a spectrum of possible DVOA ratings. In addition, the playoff odds report does not adjust our win projections to better reflect the recent historical distribution of wins in the NFL (fewer teams between 7-9 and 9-7 than you would expect from a normal distribution, and more teams around 12-4 or 4-12). However, this kind of conservative forecast generally leads to smaller errors than a forecast that looks more like a real set of final standings with the best team around 14-2 and the worst team around 2-14. Also note that because this projection doesn't make the adjustment to better reflect the historical distribution of wins, the Super Bowl odds here will be different than those in the ESPN Insider piece I wrote a week ago.

Obviously, the best team in the league will likely have more than 10 wins, and the worst team will have more than 10 losses. We're also predicting 10 of 12 playoff teams to repeat from a year ago; that's extremely unlikely as a single event, but when we forecast each individual team on its own, that's what we end up with.

For a slightly broader viewpoint on season win totals, I have also included a mean wins total based not on the simulation but rather on a formula built from five variables: offensive DVOA, defensive DVOA, special teams DVOA, balance between offense and defense, and projected mean DVOA of this year's opponents (i.e. schedule strength).

It's also worth noting that the simulation uses the same projected DVOA for each team for the entire year, so teams penalized for missing players for half the season are essentially half-penalized for the entire season. That's important in the case of San Francisco; if we broke up the simulation into Weeks 1-9 and Weeks 10-17, teams that played San Francisco in the first half of the year would have slightly higher win totals, and teams that played San Francisco in the second half would have slightly lower win totals.

Because of an error in the playoff odds program, the initial playoff odds ran without the odds for getting the No. 1 pick or the odds for "special Super Bowls." We'll re-run and get those numbers up as soon as possible.

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

DEN 23.7% 1 11.7 22.9% 1 -2.6% 8 -1.7% 29 -0.1% 20 10.2 80.4% 13.9%
SEA 16.1% 2 9.9 3.4% 10 -10.7% 1 2.0% 3 1.2% 13 9.6 66.6% 10.2%
GB 15.1% 3 9.8 18.6% 2 2.2% 23 -1.3% 28 2.1% 9 9.4 62.5% 10.6%
NE 13.8% 4 9.7 10.3% 5 0.0% 17 3.5% 1 0.4% 17 9.4 66.2% 6.8%
NO 12.2% 5 10.0 8.2% 7 -4.3% 4 -0.3% 18 -0.2% 21 9.3 61.9% 9.1%
SD 10.4% 6 9.5 16.2% 3 5.7% 30 -0.1% 16 0.0% 18 9.2 61.1% 4.7%
SF 10.2% 7 8.9 5.4% 9 -3.8% 5 1.1% 7 2.2% 6 8.9 52.4% 5.6%
CHI 8.8% 8 8.9 8.7% 6 0.1% 18 0.1% 15 1.6% 11 8.8 49.1% 6.3%
PHI 6.5% 9 9.7 10.9% 4 1.2% 21 -3.2% 32 -2.1% 25 9.0 57.1% 6.0%
CIN 4.1% 10 9.0 0.9% 13 -3.1% 6 0.2% 12 -1.0% 22 8.6 50.1% 3.3%
PIT 3.0% 11 9.3 2.8% 11 -1.0% 13 -0.8% 25 -3.8% 30 8.8 53.1% 2.9%
CAR 1.3% 12 7.5 -4.9% 22 -6.0% 2 0.2% 13 2.4% 4 7.9 32.3% 2.6%
IND 0.2% 13 8.9 5.8% 8 5.2% 29 -0.4% 19 -2.9% 28 8.4 48.6% 2.4%
TB -1.1% 14 7.5 -1.9% 17 -1.6% 12 -0.8% 26 0.7% 16 7.8 31.1% 1.5%
MIN -1.4% 15 7.2 -0.6% 14 2.5% 24 1.7% 4 2.1% 7 7.6 26.6% 1.7%
BUF -2.2% 16 7.3 -4.9% 23 -2.3% 10 0.4% 10 0.8% 15 7.7 30.6% 1.2%
STL -2.8% 17 6.9 -6.9% 27 -2.5% 9 1.5% 5 2.4% 5 7.4 23.7% 1.1%
NYJ -3.3% 18 6.9 -6.7% 26 -2.9% 7 0.6% 9 2.1% 8 7.4 25.6% 0.8%
ATL -4.2% 19 7.2 1.6% 12 6.1% 31 0.3% 11 1.0% 14 7.4 22.8% 1.2%
ARI -4.4% 20 6.8 -4.6% 21 -0.8% 14 -0.7% 23 2.0% 10 7.3 23.1% 1.3%
KC -5.6% 21 6.5 -5.9% 24 -0.1% 16 0.2% 14 3.1% 2 7.0 20.4% 0.5%
NYG -5.8% 22 7.7 -11.1% 31 -5.6% 3 -0.4% 22 -2.3% 26 7.6 27.6% 1.1%
TEN -6.0% 23 8.1 -1.3% 15 4.2% 26 -0.4% 21 -5.0% 31 7.8 36.0% 0.7%
MIA -6.0% 24 6.7 -4.3% 19 0.9% 20 -0.7% 24 1.4% 12 7.2 22.7% 0.5%
BAL -6.0% 25 7.5 -10.4% 29 -1.9% 11 2.5% 2 -1.8% 24 7.5 28.3% 0.9%
WAS -6.4% 26 7.6 -4.5% 20 -0.4% 15 -2.2% 31 -1.6% 23 7.5 25.6% 0.8%
DET -7.3% 27 6.4 -2.2% 18 3.0% 25 -2.1% 30 2.8% 3 6.9 16.0% 0.6%
DAL -7.9% 28 6.9 -1.5% 16 6.3% 32 -0.1% 17 0.0% 19 7.1 21.4% 0.8%
HOU -8.4% 29 7.7 -6.6% 25 0.9% 19 -0.9% 27 -5.4% 32 7.6 32.9% 0.7%
CLE -13.8% 30 6.7 -10.3% 28 4.6% 27 1.0% 8 -2.4% 27 6.8 17.8% 0.3%
JAC -14.1% 31 6.8 -10.8% 30 4.6% 28 1.3% 6 -3.1% 29 6.9 19.0% 0.2%
OAK -18.5% 32 4.9 -16.2% 32 1.9% 22 -0.4% 20 3.5% 1 5.8 7.2% 0.0%

Posted by: Aaron Schatz on 03 Sep 2014

76 comments, Last at 08 Sep 2014, 6:04pm by BengalFaninIN


by PaddyPat :: Wed, 09/03/2014 - 5:31pm

Is there any kind of standard deviation for these projections? ie. +/- 1.0 or 0.5 DVOA within a confidence interval of 80 percent or 65 percent or something of the like?

by lroach5 :: Wed, 09/03/2014 - 5:35pm

Nice. As a Giants fan I came here to make sure that FO projected the Giants to have an elite defense because I'm convinced that it's going to be one of the best units in the league. Nice work, FO. Unfortunately, I didn't have time to look at any other columns besides Defense DVOA and Defense Rank. I'm going to come back later and check everything else out, but my hopes are high! Here's to hoping that FO projects a bounce back year for Eli and the Giants offense! Go Big Blue!

by Karl Cuba :: Wed, 09/03/2014 - 7:54pm

Can someone explain why the Giants are expected to have an elite defense? I see a decent secondary and not a great deal else, what am I missing?

by Perfundle :: Wed, 09/03/2014 - 9:08pm

As Barnwell pointed out in his season predictions, the Giants faced a lot of QBs who were much worse than the opponent's overall pass DVOA suggests, so they're probably getting overrated on that front. Getting Pierre-Paul healthy for the whole season is also boosting their rating.

by nickbradley :: Thu, 09/04/2014 - 1:56am

maybe there were a lot of random plays that didn't go their way last year?

Karl, Good news for SF fans: Arizona is expected to regress A LOT, mostly on defense.

by Perfundle :: Wed, 09/03/2014 - 5:50pm

Whoa, didn't know that Kansas City lost both their return guys. That would explain their massive plunge in special teams DVOA. Who's replacing McCluster and Demps?

by Vincent Verhei :: Wed, 09/03/2014 - 6:13pm

De'Anthony Thomas. Scored four KR TDs and one PR TD at Oregon, and has an 80-yard PR TD this preseason.

by Paydro :: Wed, 09/03/2014 - 11:13pm

No green?? What witchcraft is this?? Who are you and what have you done with Vince?!

by Vincent Verhei :: Wed, 09/03/2014 - 11:48pm

Heh. Oh, the joys of having one account under "Vince" and one under "Vincent."

I assure you it's me.

by JoeyHarringtonsPiano :: Thu, 09/04/2014 - 6:03am

I'm sure Dave Toub will work his usual magic and they'll be fine.

by cstoos :: Thu, 09/04/2014 - 1:07pm

I am surprised by the large DVOA regression of KC's offense and defense.

While I don't see the offense scoring 35 per game (which they averaged over the last 7 games, 5 of which were against playoff teams), I do see them ending closer to the positive side of offensive DVOA. I also don't see the defense regressing at all, let alone +7.7% from last season.

by Will Allen :: Wed, 09/03/2014 - 6:23pm

With each passing year I disagree with FO's projected wins for the Vikings less and less. Obviously, Aaron, you've adjusted your secret sauce to better match my oh-so-sensible view of the known universe.

by Luke McKenna :: Wed, 09/03/2014 - 6:32pm

Think I'm right in saying Buffalo took the biggest hit in terms of drop from the book to the latest mean wins projection. 0.7 drop.

by Will Allen :: Wed, 09/03/2014 - 6:35pm

I will say this, however. If the Vikings do beat FO's and my projection, it'll likely be due to (beyond the obvious good injury luck) the value of coaching. The Vikings really haven't been well coached across the board since Dennis Green had Dungy and Billick as coordinators. Zimmer really seems to be a teacher who speaks in a a non-blustery direct manner that players respond to, and Norv's pretty proven as an OC. I don't know if Edwards is a good dc, but Zimmer is heavily involved on that side of the ball, it would appear. I don't think they'll be outcoached often, and they may get the better of more than few teams in that department, which would be quite a switch.

The first 6 weeks are really tough, however. If they can go 3-3 that'll give'em a substantial leg up, and even 2-4 would have a chance to make December worth watching.

by Chip :: Thu, 09/04/2014 - 11:48pm

Agree on all accounts Will (and I'm a Bears fan).

Zimmer is a great hire and a refreshing change from past regimes. The Vikes have drafted well for the last 2-3 years accumulating a lot of 1st round picks. All they've needed is a QB.

I wouldn't be surprised if the NFCN is the toughest division in the NFL by season end.

by 3.141592653 not... :: Sat, 09/06/2014 - 12:15am

If you are putting any eggs in a Norv Turner basket, well, you will far fewer chickens than you were counting before the hatching process.

by Will Allen :: Wed, 09/03/2014 - 6:45pm

The number that leaps out at me is the Eagles having a 57% chance of making the playoffs, in a division with what appears to be three corpses. I think Romo is going to really decline under the pressure of having to score eleventy bazillion points a game, it appears like it'll be midseason at least before the Giants look better than a MAC team on offense, and I think there's good chance that RG3 has been ruined, and rendered uncoachable by Mr.Redskin. I think the Eagles have about 75% chance of winning the division.

by Thok :: Wed, 09/03/2014 - 8:00pm

It wouldn't take much for any of Dallas, NY, or Washington to have enough of a fluke improvement on defense to be 10-6. And it's easy to see Philly's offense crater if Foles regresses significantly or if the wide receivers underperform.

by herewegobrownie... :: Wed, 09/03/2014 - 10:53pm

Yes, but Dallas's offense will crater for the time Weeden has to see the field (which is likely to be a fairly non-zero amount given Romo's health.)

by MilkmanDanimal :: Wed, 09/03/2014 - 11:32pm

Well, the offense will clearly suffer from a productivity standpoint when Romo gets hurt, but from a general humor standpoint, it's going to just be spectacular when Weeden hits the field.

by commissionerleaf :: Thu, 09/04/2014 - 1:24pm

Dallas has one of the better offensive lines in football now, which should help with Romo's health. And scoring has never been the Cowboys' problem. The problem is that they have to win shootouts every week a la the post-Gregg Williams Saints. Romo could put up very Drew Brees numbers this year.

by bingo762 :: Thu, 09/04/2014 - 10:30am

There were 3 corpses in the NFC East last year and it's not like the Eagles ran away with it. It still came down to a last minute INT by Kyle Orton for the Eagles to make the playoffs. Now throw in some turnover and health regression from the Eagles, and Washington and NY on the uptick and that's what you get

by Will Allen :: Thu, 09/04/2014 - 11:01am

The Redskins on the uptick is a proposition I have some doubt in. I think the owner may have created an untenable situation which brews disaster. The Giants may be on the uptick, but in this league it is hard, hard, hard, to consistently win games 13-10.

by jonnyblazin :: Wed, 09/03/2014 - 7:58pm

Ouch, Ravens with the projected 29th best offense.

I'm a bit more optimistic, I think with a non-horrendous C, healthy guards (knock on wood), and Kubiak as the OC, they'll get decent O-line play. They should get better production at WR (with Smith Sr.) and TE (healthy Pitta), and Flacco will probably do better under a controlled, structured offense than under whatever it is that Caldwell put together last year.

by herewegobrownie... :: Wed, 09/03/2014 - 11:29pm

Remember that, as explained, it mistakenly treats the Ray Rice-less team of the first two games as the team they will field for the whole season. (Although he struggled enough last year that I'm not sure how much this would adjust the equation.)

by Shattenjager :: Thu, 09/04/2014 - 1:01pm

"[T]eams penalized for missing players for half the season are essentially half-penalized for the entire season."

by Aaron Schatz :: Thu, 09/04/2014 - 1:29pm

The projections aren't adjusted for players who miss two games. I've only adjusted for players missing at least half the season, except for Mike Pouncey (one-fourth).

by herewegobrownie... :: Thu, 09/04/2014 - 7:24pm

Ok, re-reading your line and this, I understand now.

So you're saying, essentially, that Ray Rice's, Robert Mathis', and Ray McDonald's values are included in each and every game, but in each game only 7/16th of Aldon Smith's value is included, regardless of whether it's during or after the suspension?

by tuluse :: Wed, 09/03/2014 - 9:16pm

FO keeps projecting the Bears to make the playoffs, and the most important players keep getting hurt.

by coremill :: Wed, 09/03/2014 - 9:17pm

I know defense regresses more than offense, and is harder to predict, but the projected range for defense DVOA seems extremely narrow. Seattle is projected best at -10.6, and Dallas worst at 6.3. Last year 15/32 teams' final defense DVOA fell outside that range. Dallas' 6.3 would have been good enough for 24th last season and would represent a significant improvement for them (they finished last year 30th with 13.8). I only went back 10 years, but the 32nd-ranked DVOA defense has never been better than 14.8 during that time and averaged 18.0.

Is this the same issue as the projected wins being more conservative and not matching the historical distribution of wins? Seems like DVOA should be easier to project, since it is less random than wins.

by big10freak :: Wed, 09/03/2014 - 9:47pm

Ted Thompson and Mike McCarthy have a fair amount of accomplishments to their credit but for the life of me I don't understand why the Packers are unable to consistently generate decent special teams. Green Bay has been above average I think once in the past 5-6 years. (2011)

by DisplacedPackerFan :: Thu, 09/04/2014 - 9:07am

There are two other things that are in common to look at.

1) Injuries - I don't want to dig up all the AGL numbers for the Packers since TT and MM took over, but they have been the most injured team several years and I think at best they have been middle of the league. That causes a lot of shuffling on the special teams and lowers your overall talent level. The Packers have completely revamped their training camp and regular season practice schedules, contracted with a 3rd party company to data collect information about athletes, and they had significantly fewer training camp injuries this year.
There is a lot of random in injures, but if there 30% you can control it is possible they have flipped from being about as bad as you can be with that 30% to about as good as you can be.

2) Shawn Slocum - He hasn't been the ST coordinator for the whole tenure of TT and MM, but for quite a few years now. I think Toub in Chicago and now KC proves that coaching matters. The Packers brought in Ron Zook this year as an assistant and from what I saw in the preseason it has mattered.

So it's possible they have done as much as they can to improve special teams this year. I'm hopeful as a fan but still have trouble believing it will happen.

by justanothersteve :: Thu, 09/04/2014 - 9:24am

Shawn Slocum.

by Jetspete :: Wed, 09/03/2014 - 10:22pm

i could've listened to Mike and Mike and got the same generic result; 10/12 teams returning to the playoffs. DVOA is definitely a victim of its' own success, and most of their analysis is considered mainstream. But no matter how mainstream you get, football fans still know that the turnover in playoff teams is always closer to 50%.

by herewegobrownie... :: Wed, 09/03/2014 - 11:12pm

Is it, recently?

2012 playoff teams that missed in '13: Bal, Hou, Wash, Min, Atl

2011 playoff teams that missed in '12: NYG, DET, PIT

That's just one third playoff turnover (small sample, I know.) Overall division standings turnover has also been pretty slight, compared to reputation. (It is interesting to see the "estimated division standings" from the playoff odds; the 2nd place Bills finish is in line with the good "worst to first" odds they got, even though the Pats have been so dominant for so long in that division. It is also interesting to see the Colts being given <50% playoff odds, and in about one fifth of their playoff makes they do not win their division; despite the perception of the enormous gap between them and their AFCS peers, it seems to reflect the limitations of Luck's supporting cast.)

by Perfundle :: Wed, 09/03/2014 - 11:06pm

This time, though, there seems to be a really large gap between the predicted playoff teams and everyone else. The least likely playoffs team to make it is Indianapolis, at 48.6%, and then it drops all the way to 36.0% for Tennessee. The 12 playoff teams are the only ones predicted to have more wins than losses too.

by RickD :: Thu, 09/04/2014 - 12:10am

I feel like these simulations overestimate he Super Bowl win probabilities of some of the weaker teams. The Jaguars are given a .2% chance of winning. Now this may seem reasonably (only 1/500), but the Panthers are only given a 2.6% chance of winning.

Would anybody be interested in the following bet? If the Jaguars win the Super Bowl, you get $26, but if the Panthers win, I get $2. If somebody else wins, the bet is off.

It just seems like the relative odds there should be more than 13:1.

by nickbradley :: Thu, 09/04/2014 - 1:11am

Curious to know whether the STL projections were done with Bradford in there or Shaun Hill.

Im a big Skeptic on Bradford, but it seems like a significant dropoff to Shaun Hill, based on what I remember of him in San Francisco

by JoeyHarringtonsPiano :: Thu, 09/04/2014 - 6:08am

They both throw a lot of checkdowns and don't attempt many deep passes. Except Hill takes fewer sacks and turns the ball over less. I don't think it will be as much of a dropoff as some people think. Of course he's now 34 and hasn't seen live action since attempting 13 passes in 2012, so it's possible his skills have significantly diminished.

by t.d. :: Sat, 09/06/2014 - 5:21am

Cowboy Brad Johnson agrees

by MC2 :: Thu, 09/04/2014 - 1:25am

I know it doesn't really tell us much about this year's Broncos, but it's still interesting to note that it's been over 40 years since a team lost the Super Bowl one year, and then won it the next year.

In fact, it's only happened twice, and those came back-to-back, with the Cowboys losing SB V and winning SB VI, beating the Dolphins, who then won SB VII.

by nickbradley :: Thu, 09/04/2014 - 1:47am

yeah but a lot of teams have 'been right there' the following year (e.g. SF2013), and the AFC is so weak that Denver can pull it off

by nickbradley :: Thu, 09/04/2014 - 1:53am

So this is basically regression towards mean performance with some slight adjustments.

So, I decided to look at deviation from regression. No surprise that GB has the largest positive deviation (duh, Rodgers), while KC has the biggest negative deviation. ST had a much weaker relationship, so I excluded it.

All of these make sense except for Denver -- they didn't get any better on offense and they lost Decker, and while their defense is better it isn't *this* much better. Also, Dallas' Offense and Oakland's offense -- did the raiders recover a ton of fumbles last year or something?

TEAM Total DVOA Deviation from Regression/ Offensive DVOA Deviation from Regression/ Defensive DVOA Deviation from Regression/
GB 17.6% 13.8% -2.4%
DEN 10.1% 4.4% -2.6%
SD 7.7% 3.4% 0.1%
CHI 6.0% 1.3% -2.7%
NE 5.9% 1.2% -1.4%
WAS 4.5% 0.9% -1.8%
NO 4.1% -0.7% -2.5%
MIN 3.3% 1.9% -0.9%
SF 2.9% 0.3% -2.4%
HOU 2.6% 3.7% 0.1%
PIT 2.6% 0.3% -2.3%
JAC 1.8% 5.4% 1.1%
TB 1.0% 3.7% 0.5%
NYG 0.7% 0.9% -2.0%
PHI 0.1% -1.8% -0.4%
ATL 0.1% -0.3% 1.8%
NYJ -0.1% 1.6% -1.2%
SEA -0.6% -1.9% -2.6%
BUF -0.8% 1.3% 2.0%
IND -1.2% 3.3% 4.9%
CIN -1.8% 0.6% 0.8%
BAL -3.2% 1.4% 0.8%
MIA -3.3% -3.4% 0.1%
TEN -3.5% -2.2% 2.8%
STL -3.8% -1.8% -0.7%
OAK -4.3% -7.2% -1.4%
CLE -4.7% -2.5% 2.0%
DET -6.7% -1.3% 3.2%
DAL -6.8% -5.7% 1.9%
ARI -8.6% -3.4% 4.3%
CAR -9.0% -9.4% -1.1%
KC -12.9% -7.7% 2.0%

by nickbradley :: Thu, 09/04/2014 - 2:05am

* note - variance from regression is twice as large for offenses as it is for defenses, I assume this is due to QB play, with the biggest variations coming in GB, CAR, OAK, KC, JAC, HOU, TB, KC, and DAL.

Rodgers is back, Cam has lost all his WRs, a bunch of teams have new QBs, and Smith lost 3/5ths of his offensive line or something crazy like that.

by eggwasp :: Thu, 09/04/2014 - 3:10am

Does FO ever follow up the previous years predictions to show us how accurate these DVOA predictions are?

by Riceloft :: Thu, 09/04/2014 - 7:58am

I went back a couple of years and checked.

2013 AFC actual: DEN, NE, CIN, IND, KC, SD
2013 AFC predic: DEN, NE, CIN, PIT, BAL, HOU

2013 NFC actual: SEA, CAR, GB, SF, NO, PHI
2013 NFC predic: SEA, CAR, GB, SF, NO, WAS

So in 2013, DVOA correctly predicted 3 of 6 AFC and 5 of 6 NFC teams.

2012 AFC actual: DEN, NE, HOU, BAL, IND, CIN
2012 AFC predic: DEN, NE, HOU, PIT, BUF, NYJ

2012 NFC actual: ATL, SF, GB, WAS, SEA, MIN
2012 NFC predic: ATL, SF, GB, NO, CHI, NYG

In 2012, DVOA got 3 of 6 in both conferences.

2011 AFC actual: NE, BAL, HOU, PIT, DEN, CIN
2011 AFC predic: NE, BAL, HOU, PIT, SD, NYJ

2011 NFC actual: GB, SF, NO, ATL, NYG, DET
2011 NFC predic: GB, SF, NO, ATL, PHI, CHI

In 2011, DVOA got 4 of 6 in both conferences.

2010 AFC actual: NE, PIT, IND, KC, BAL, NYJ
2010 AFC predic: NE, PIT, IND, KC, BAL, MIA

2010 NFC actual: ATL, CHI, PHI, GB, SEA, NO
2010 NFC predic: ATL, CHI, PHI, GB, ARI, WAS

In 2010 DVOA got 5 of 6 in the AFC and 4 of 6 in the NFC.

Overall, I think this is a pretty good track record.

by Jeremy Billones :: Thu, 09/04/2014 - 10:36am

In 2012, 3 of their 6 misses were WAS, SEA and IND aka RG3's rookie year, Wilson's rookie year and Luck's rookie year. So except for that outlier they're hitting 4/6 pretty regularly.

by Purds :: Thu, 09/04/2014 - 9:07am

Sorry for posting here, too, but...

Um, not sure where else to put this comment, but the fact that FO posted the "news" of a #3 depth chart QB being traded for a sixth-round pick, but then didn't post the news that a team owner was fined half a million and suspended for 6 games doesn't exactly debunk the allegations of this being Patriots-centric site. (The #3 QB was a Patriot and "news." The owner was not.)

by aces4me :: Thu, 09/04/2014 - 10:15am

Maybe because it is possible (if very unlikely) that trade could impact game performance but the owner suspension (what does that even mean?) will not.

by PaddyPat :: Thu, 09/04/2014 - 10:15am

I think the Mallett trade was "news" because it means that Houston will likely have another quarterback vying for a starting roll by week 3-4. The NE part was much less important.

by Steve in WI :: Thu, 09/04/2014 - 11:06am

Just playing devil's advocate, but I could argue that the Irsay news doesn't really affect the product on the field at all. The Colts aren't going to change in any meaningful way while he's suspended. (I certainly see the point that an owner being fined/suspended is newsworthy from a general NFL perspective).

On the other hand, the Mallett trade is interesting partly because it was expected to happen much sooner and for more than a 6th/7th round pick. There's also debate to be had over what this means for who will start games for Houston, and what would happen to the Pats if Brady gets hurt and Garrappolo has to step in so soon.

by RickD :: Thu, 09/04/2014 - 1:12pm

Irsay was suspended for 6 weeks? That really doesn't matter at all.

The fine is sizeable, but on the whole I'm not seeing how a disciplinary action against an owner is comparable to a trade of a player.

by Purds :: Thu, 09/04/2014 - 4:02pm

To the group: I agree suspending an owner doesn't hurt a team. But, there were lots of trades for 6th round material this weekend, but only one was given its own article. I don't really care much, as I am interested in the Mallet thing because that is all I hear about living here in NE. I was just pointing out that FO didn't exactly change opinions about the NE-focus when they post an article about a trade for a 6th round pick.

by BengalFaninIN :: Mon, 09/08/2014 - 6:04pm

I like to complain about the Patriots homerism sometimes. But to me the trade is newsworthy, the owner suspension isn't. I read things like FO to find out about the game itself. Not the soap opera of NFL suspensions and silliness. I also can't imagine how the suspension of an owner (Whatever that means.) is going to change what happens on the field. The Texans finding a QB on the other hand, could be a factor. If they had one they might be a playoff team again.

by Purds :: Thu, 09/04/2014 - 9:10am

On to this article's ideas: I sure am hoping the Colts defense will be better than the projected #29 rank, and I think they will be simply because that is just so awful, but I do agree that they could be the worst division winner in the league. Their offensive line is just such a mess. Lots of passing targets, but they can't run block to save themselves, and so they will be so one-dimensional that it will be painful to watch.

by Bernie :: Thu, 09/04/2014 - 10:19am

My only hope (straw clutching hope) is that Pagano was telling his line to deliberately look bad to lull the rest of the league into a false sense of security, and come game time will be the reincarnation of the Hogs. That sounds pretty likely, right? whimpers....

by turbohappy :: Thu, 09/04/2014 - 10:37am

Donald Thomas and his insta-IR quad injuries are a big blow as he seems to be really good, just can't be healthy now. Shipley going from on the street to starting center is pretty crazy, although I liked him a LOT more than Satele when he saw action before...was hoping we could do even better at the center spot though.

by jedmarshall :: Thu, 09/04/2014 - 11:37am

It's hard not to do better than Satele. It's no coincidence that they seems to block better every time he was banged up and not in the game. As long as Allen and other receiving targets stay healthy, I'm not too concerned about the offense.

The defense however...yuck. I sense a lot of shootouts this year. They could have the worst safeties in the league and if the corners get hurt again there will be lots of ugly passes.

by turbohappy :: Thu, 09/04/2014 - 10:40am

I honestly think if they can just be healthy as a line and get reps together they can be decent. Satele was just such a terrible blocker...he must have been good at calling protections and snapping and other center stuff or something to keep seeing the field.

by herewegobrownie... :: Thu, 09/04/2014 - 11:15am

Remember that the projections treat the current roster without Robert Mathis as though it will be the same for the entire season.

by Kevin from Philly :: Thu, 09/04/2014 - 11:04am

Don't know about regression to the mean, but it sure looks like the comments have all regressed to the center.

by mehllageman56 :: Thu, 09/04/2014 - 12:32pm

So the Jets projections went down, but they're still 18th in the league? I thought they were going to challenge for last place in the AFC East.

by RickD :: Thu, 09/04/2014 - 1:15pm

mean wins, AFC East

NE 9.7
Buf 7.3
NYJ 6.9
MIA 6.7

With only .2 wins more than Miami, it certainly seems like the Jets are challenging for last place in the AFC East.

by nat :: Fri, 09/05/2014 - 10:43am

Something is very screwy in the playoff odds report. Besides having the date wrong by a year, I mean.

There are eleven teams whose SB winning odds are more than half their Conference winning odds. These are the teams that the simulation would pick as favorites if they make it to the Super Bowl, considering their likely opponents.

They are: GB, SEA, NO, CHI, PHI, SF, CAR, MIN, TB, ARI, and STL.

That's just insane. If, by some freak chance St. Louis gets to the Super Bowl, they are NOT going to be the favorite. Ditto, most of the other teams. Something is broken in your simulation.

Also, you've left the Raiders out of the Playoff Scenarios table. Which, while understandable, leads me to ask...

Is Simulation Drunk?

by Eddo :: Fri, 09/05/2014 - 1:00pm

I'm pretty sure all that means is that the NFC is projected to be significantly stronger.

And I'm sure "if St. Louis gets to the Super Bowl, they are NOT going to be the favorite" was said in 1999, too. An out-of-nowhere team certainly could wind up being favored.

by nat :: Fri, 09/05/2014 - 2:03pm

The Broncos, at 23.7%, have the best projected DVOA in the league. They should be favorites in any possible SB match up.

Their projected odds of getting to the Super Bowl: 28.2%
Their projected odds of winning the Super Bowl: 13.9%

That is, the Broncos are predicted to win 13.9/28.2 = 49.3% of the times that they get there.

That's really odd, since Denver has a significant edge in project DVOA against any team they might face.

No, this isn't something magical about the simulation. It's just broken. My first guess is that the NFC teams were given a home field advantage bonus by accident.

by Aaron Schatz :: Fri, 09/05/2014 - 2:31pm

We are looking into this.

by nat :: Fri, 09/05/2014 - 2:53pm


Here's hoping it's an easy fix - or better yet - that I'm wrong.

by Aaron Schatz :: Fri, 09/05/2014 - 5:08pm

OK, yes, unfortunately we have this error. We re-wrote the playoff odds calculator this year so the code would be smoother when we need to run "what if"-type scenarios, and this error snuck in by accident. We're re-running things now to produce a new set of preseason odds that will make the Super Bowl a neutral-site game again. (We're also checking to make sure the London games are neutral-site.) New odds should be posted by Saturday afternoon.

by nat :: Fri, 09/05/2014 - 6:48pm

That's great news, much better than an "I dunno" answer.
I love how FO strives to continually improve, even if that gives a hiccough ever now and then. The end result is worth the wait.

by nat :: Sat, 09/06/2014 - 6:47pm

The new odds report looks much more sane. Thanks.

I'm not sure I agree that Seattle will regress so much more than Denver. But, hey, a method is a method. We'll just have to wait and see.

by schmoker :: Fri, 09/05/2014 - 2:40pm

Is it possible to rank the Browns below a couple of college and elite high school teams? Because, boy, you do not know how nuts it is to be here in NE Ohio and have to listen to all the "THIS IS GOING TO BE A TOP 5 DEFENSE!!!!" guff. Even people that seem to understand the Browns stink, Manziel stinks, and Hoyer stinks, still seem to buy compeltely into the notion that the Browns defense was great last year, and that they will now be elite this year. There is not a single person I know, or football writer or broadcaster in the whole state, who does not feed us this party line every single day.

If the Browns could just go 0-16 and then do the stupid thing and trade that Jameis Winston or whomever pick away so they can enjoy year two of the failed Johnny Short Guy experiment, I'll be a happy man. I cannot help but root against idiocy and corruption, and Jimmy Haslam is the poster child for both.

by herewegobrownie... :: Fri, 09/05/2014 - 3:48pm

"Is it possible to rank the Browns below a couple of college and elite high school teams?"


And you're barking up the wrong tree (if that's the right expression, given the Dawg Pound, etc. terminology) questioning Pettine that much here on defense; DVOA liked the Bills' improvement a lot last year.

I do think last year's Browns defense wasn't quite as good as many hyped - and DVOA said as much. But not "everyone" thought it was that good last year; plenty of people were critical of the team's poor 3rd down conversion allowance.

Haslam himself was effectively exonerated by the agreement Pilot Flying J made in July, FYI.

by 3.141592653 not... :: Sat, 09/06/2014 - 12:29am

Haslam himself was effectively exonerated by the agreement Pilot Flying J made in July,
How much did he pay for that again, 97 million, or so? Was that really an exoneration? Was that really an effective exoneration?

by LionInAZ :: Sat, 09/06/2014 - 1:04am

According to the reports I read, the settlement does not protect Haslam from prosecution by DOJ. That's a long, long way from 'exoneration'.

by herewegobrownie... :: Sat, 09/06/2014 - 12:16pm


I never said that Pilot Flying J as a whole wasn't very un-exonerated, to use a triple negative.

"There was apparently significant misconduct of middle management employees but it is entirely possible in a company that size that you can have middle management in a sales department running their own scheme on the side without the president or the board of directors being aware of it."

Though the agreement does not preclude the prosecution of any individual, Raybin said it is unlikely that CEO Jimmy Haslam will be prosecuted. "I would say the government would not have entered into this if Haslam were clearly culpable," Raybin said."