Week 6 DVOA Ratings

Week 6 DVOA Ratings
Week 6 DVOA Ratings
Photo: USA Today Sports Images

by Aaron Schatz

Minnesota stays at No. 1 in the Football Outsiders DVOA ratings after a week off, with Seattle moving up to No. 2 after its win over Atlanta combined with Philadelphia's loss to Washington. There's a gap of a few percentage points between Minnesota and Seattle, and then another gap between Seattle and the rest of the league. However, the team sitting at No. 3, the top team in the AFC, is probably quite a surprise to most readers. Say hello to the suddenly playoff-likely Buffalo Bills.

Yes, with their big win over San Francisco, Buffalo has climbed all the way up to No. 3 in DVOA. The Bills are winning, and they're doing it with a great amount of balance. Right now, Buffalo is the only team in the league to rank in the DVOA top 10 for all three phases of the game: offense (6), defense (10), and special teams (8). The offense is primarily led by the running game, which ranks second in DVOA behind Dallas. The pass offense is only 15th. The defense is well-balanced, ranked 12th against both the run and the pass.

Does Buffalo's easy schedule have something to do with their high DVOA rating? Perhaps. Of course, DVOA does adjust for opponents, but those adjustments are only at 60 percent strength right now. That's also part of the problem with judging Buffalo based on its schedule: just who are these teams, anyway? We can be pretty sure that the 49ers and Jets are horrible, but we really still don't know how good the Ravens will be all year, or how good the Cardinals will be all year. Buffalo fans can also take satisfaction in the fact that Buffalo's schedule doesn't get any harder the rest of the season. What looked like a very difficult midseason run looks a bit easier now. Cincinnati doesn't seem like the tough opponent we thought it was before the season, and Oakland may not be the tough opponent we thought it was just a couple weeks ago. (The Raiders drop from 10th to 18th in DVOA after losing to Kansas City.) The Bills also still have two games against Miami on the docket, plus the rematch with the Jets and home games against Cleveland and Jacksonville. Based on average DVOA of opponent, only Tennessee and San Diego have easier schedules the rest of the way.

There's a bit of an asterisk on this whole performance, which is Buffalo's 16-0 win over New England in Week 4. As the season goes along, that game is going to stand out more and more because the adjustment for stopping the New England offense will get stronger and stronger. However, as we all know when we apply a little common sense, that wasn't really the New England offense in Week 4. That offense was led by an injured third-string quarterback. If we remove that game from Buffalo's defensive DVOA, it drops all the way from -7.2% (10th) to 0.4% (18th). However, Buffalo's overall DVOA would still be sixth in the NFL at that point -- or seventh, if we also remove the game from New England's offensive DVOA. That would still make New England and Buffalo the top two teams in the AFC, in either order.

(An aside: Every year we end up with a situation like this, where there are a couple of games against backup quarterbacks which seem a bit out of whack because our opponent adjustments are based on how well that opponent has played for the entire season. As I note every year, trying to do separate opponent adjustments based on which players were healthy or out in each week is just way too complicated. Do you adjust based only on quarterbacks, or on other players as well? How do you handle major injuries on defense? What if quarterbacks play, but aren't 100 percent healthy? How do you handle the fact that the sample size for the backup quarterback may be only one game, and based on one game we would have surmised that in 2015 a Saints team led by Luke McCown was a much tougher opponent than a Saints team led by Drew Brees? This is where you have to use your common sense and not just quote our statistics as if they were the tablets Moses brought down from Sinai. Ironically, the Patriots' defensive DVOA is probably going to have this exact same problem after this week, because its performance against Landry Jones will be adjusted based on a Pittsburgh offense that will be quarterbacked for most of the season by the much better Ben Roethlisberger.)

Buffalo's strong run and easy remaining schedule set the Bills up nicely to finally break their NFL-long postseason drought of 16 seasons. This week's playoff odds simulation gives Buffalo a 75.7 percent chance to reach the playoffs, second in the AFC behind only New England at 90.6 percent. The Patriots are looking really good right now; since Tom Brady's return they have climbed from 17th to seventh in total DVOA and from 13th to fourth in offensive DVOA. Both AFC East teams got a lot of help this weekend when the other three 4-1 teams in the AFC -- Denver, Oakland, and Pittsburgh -- all lost. Either New England or Buffalo wins a wild card in 68 percent of this week's simulations.

Those three losses by the top AFC teams also emphasize another surprising turn of events for the NFL in 2016: the NFC is dominant over the AFC so far. Although the NFC had three of the top four teams in 2015, the AFC had the next five teams in a row, giving them six of the top 10. However, this week the AFC has only three teams in the top 10: Buffalo, New England, and No. 9 Denver. The AFC also has the three lowest teams in DVOA and four of the bottom five: Indianapolis at 28, Houston at 30, Cleveland at 31, and the New York Jets moving back behind the Browns and into last place. Believe it or not, the Houston Texans are not the worst 4-2 team in DVOA history. I wrote last week about two 3-2 teams that had a lower DVOA than Houston, and one of those teams, the 2007 Lions, actually then went out and won the next week as well. So the 2007 Lions had a 4-2 record and -27.1% DVOA. However, the Texans are the lowest-ranked 4-2 team in DVOA history, as the Lions were 27th that year, and the Vikings were actually 28th at 4-2 a year ago before improving significantly over their final 10 games.

* * * * *

Once again this season, we have teamed up with EA Sports to bring Football Outsiders-branded player content to Madden 17 Ultimate Team. Each week, we'll be picking out a handful of players who starred in that week's games. Some of them will be well-known players who stood out in DVOA and DYAR. Others will be under-the-radar players who only stood out with advanced stats. We'll announce the players each Tuesday in the DVOA commentary article, and the players will be available in Madden Ultimate Team packs the following weekend, beginning at 11am Eastern on Friday. We will also tweet out images of these players from the @fboutsiders Twitter account on most Fridays. The best player of each week, the Football Outsiders Hero, will require you to collect a set of the other four Football Outsiders players that week, plus a certain number of Football Outsiders collectibles available in Madden Ultimate Team packs.

The Football Outsiders stars for Week 6 are:

  • LG Richie Incognito, BUF (FOOTBALL OUTSIDERS HERO): Buffalo RB had 19 carries for 172 yards (79% success rate) up the middle; no sacks allowed.
  • WR Cole Beasley, DAL: No. 4 wide receiver with 61 DYAR this week (6-for-6, 58 yards, 2 TD); No. 1 wide receiver in DYAR for the season.
  • DT Johnathan Hankins, NYG: sack, 2 run TFL
  • RG Jermon Bushrod, MIA: Miami RB had 9 carries for 109 yards to the right and 15 carries for 59 yards up the middle; no sacks allowed.
  • CB Jason McCourty, TEN: 3 PDs, including 2 on third down, and 5 tackles after receptions which all prevented first downs.

* * * * *

All stats pages are now updated through Week 6 of 2016. Snap counts and playoff odds are also fully updated.

* * * * *

[ad placeholder 3]

These are the Football Outsiders team efficiency ratings through six weeks of 2016, 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. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE. Because it is early in the season, opponent adjustments are only at 60 percent strength; they will increase 10 percent every week through Week 10. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.

DAVE is a formula which combines our preseason projection with current DVOA to get a more accurate forecast of how a team will play the rest of the season. Right now, the preseason projection makes up 19 percent of DAVE for teams with six games played, and 27 percent of DAVE for teams with five games played.

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 MIN 35.6% 1 25.7% 2 5-0 4.5% 11 -25.9% 2 5.3% 5
2 SEA 29.3% 3 28.2% 1 4-1 -0.8% 17 -27.6% 1 2.5% 10
3 BUF 22.6% 8 18.6% 3 4-2 12.1% 6 -7.2% 10 3.3% 8
4 PHI 21.8% 2 13.5% 7 3-2 0.0% 16 -16.5% 5 5.3% 4
5 DAL 19.8% 9 15.9% 4 5-1 23.5% 1 2.5% 19 -1.1% 20
6 ATL 18.8% 5 14.4% 6 4-2 22.9% 2 10.2% 26 6.1% 3
7 NE 13.5% 13 15.4% 5 5-1 14.9% 4 3.5% 21 2.1% 11
8 GB 12.2% 6 11.3% 9 3-2 0.7% 14 -9.4% 7 2.1% 13
9 DEN 12.0% 7 10.0% 10 4-2 -6.0% 23 -17.5% 4 0.5% 17
10 ARI 10.2% 14 11.5% 8 3-3 -7.5% 24 -20.9% 3 -3.2% 24
11 WAS 8.5% 12 5.4% 14 4-2 6.4% 9 1.5% 18 3.6% 7
12 PIT 8.3% 4 9.7% 11 4-2 12.4% 5 6.9% 24 2.7% 9
13 SD 7.8% 11 5.9% 13 2-4 7.0% 8 -5.1% 12 -4.3% 26
14 KC 6.7% 17 8.8% 12 3-2 -3.8% 20 -8.5% 8 2.1% 12
15 BAL 4.2% 15 4.6% 15 3-3 -11.8% 27 -16.4% 6 -0.4% 18
16 TEN 3.0% 16 0.8% 16 3-3 5.4% 10 -8.0% 9 -10.5% 32
17 NYG 0.0% 18 -0.5% 17 3-3 -1.2% 18 -1.6% 16 -0.4% 19
18 OAK -0.2% 10 -0.7% 18 4-2 15.6% 3 17.6% 29 1.9% 15
19 CHI -2.8% 20 -3.8% 19 1-5 -1.3% 19 -2.2% 15 -3.7% 25
20 MIA -4.0% 28 -5.8% 21 2-4 -5.9% 22 0.0% 17 1.9% 14
21 NO -6.0% 21 -4.8% 20 2-3 10.8% 7 14.5% 28 -2.2% 22
22 LARM -7.8% 25 -6.0% 22 3-3 -15.5% 29 -2.7% 14 5.0% 6
23 JAC -9.6% 22 -9.1% 25 2-3 -15.8% 31 -4.7% 13 1.5% 16
24 CAR -10.6% 23 -7.7% 23 1-5 0.0% 15 3.6% 22 -7.0% 30
25 CIN -11.3% 19 -8.5% 24 2-4 1.3% 13 7.7% 25 -4.9% 28
26 DET -15.0% 24 -12.5% 26 3-3 3.4% 12 26.1% 32 7.7% 2
27 SF -17.2% 26 -17.7% 29 1-5 -10.1% 26 2.6% 20 -4.6% 27
28 IND -18.1% 27 -15.0% 27 2-4 -4.4% 21 21.9% 31 8.2% 1
29 TB -21.2% 29 -16.7% 28 2-3 -14.3% 28 3.8% 23 -3.0% 23
30 HOU -23.2% 31 -20.4% 30 4-2 -20.5% 32 -5.1% 11 -7.8% 31
31 CLE -27.9% 32 -26.9% 31 0-6 -8.7% 25 12.6% 27 -6.6% 29
32 NYJ -36.1% 30 -30.2% 32 1-5 -15.8% 30 18.5% 30 -1.8% 21
  • 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 MIN 35.6% 5-0 39.0% 6.0 1 -3.7% 22 0.8% 13 6.3% 10
2 SEA 29.3% 4-1 33.1% 4.6 2 -9.3% 32 2.5% 10 10.1% 18
3 BUF 22.6% 4-2 27.7% 3.9 8 -5.5% 25 -4.2% 30 13.1% 25
4 PHI 21.8% 3-2 32.1% 4.0 6 -5.8% 28 12.5% 1 11.5% 23
5 DAL 19.8% 5-1 21.0% 4.4 4 -1.8% 19 3.6% 8 4.0% 7
6 ATL 18.8% 4-2 15.6% 4.5 3 0.6% 17 -0.4% 18 10.6% 19
7 NE 13.5% 5-1 17.2% 3.7 10 -5.6% 26 -2.5% 27 17.1% 29
8 GB 12.2% 3-2 8.5% 3.5 12 6.2% 6 5.0% 6 8.3% 12
9 DEN 12.0% 4-2 21.5% 3.8 9 -5.8% 27 -0.1% 16 10.1% 17
10 ARI 10.2% 3-3 20.1% 3.2 15 -7.7% 30 7.6% 2 10.7% 20
11 WAS 8.5% 4-2 8.1% 4.0 7 4.4% 9 6.0% 4 8.9% 14
12 PIT 8.3% 4-2 10.3% 3.3 13 -2.4% 20 -2.1% 25 34.8% 31
13 SD 7.8% 2-4 11.8% 3.1 16 -2.5% 21 -4.7% 32 2.3% 3
14 KC 6.7% 3-2 6.8% 4.4 5 -8.7% 31 -1.1% 21 41.6% 32
15 BAL 4.2% 3-3 10.7% 3.3 14 -1.1% 18 -1.9% 24 3.0% 5
16 TEN 3.0% 3-3 5.9% 3.6 11 -5.8% 29 -4.3% 31 9.1% 16
17 NYG 0.0% 3-3 -10.2% 2.8 20 12.4% 1 1.5% 11 0.8% 1
18 OAK -0.2% 4-2 -1.4% 2.7 21 5.8% 7 -2.1% 26 11.0% 21
19 CHI -2.8% 1-5 -0.5% 2.6 22 -4.0% 23 5.4% 5 5.5% 8
20 MIA -4.0% 2-4 -7.6% 2.9 18 2.5% 12 -1.6% 23 15.5% 27
21 NO -6.0% 2-3 -0.9% 2.9 19 3.2% 11 -1.4% 22 2.5% 4
22 LARM -7.8% 3-3 -4.2% 3.0 17 1.5% 14 -0.2% 17 14.4% 26
23 JAC -9.6% 2-3 -4.9% 2.3 23 0.7% 16 0.3% 14 6.1% 9
24 CAR -10.6% 1-5 -10.9% 1.9 28 3.7% 10 4.6% 7 8.1% 11
25 CIN -11.3% 2-4 -6.6% 2.2 24 2.2% 13 -0.9% 20 8.5% 13
26 DET -15.0% 3-3 -14.0% 2.0 27 1.4% 15 7.0% 3 9.0% 15
27 SF -17.2% 1-5 -20.7% 1.6 29 10.6% 3 -0.6% 19 16.3% 28
28 IND -18.1% 2-4 -14.3% 1.6 30 -5.1% 24 0.0% 15 2.2% 2
29 TB -21.2% 2-3 -23.4% 2.1 26 4.5% 8 3.6% 9 19.8% 30
30 HOU -23.2% 4-2 -20.4% 2.1 25 6.3% 5 -2.9% 29 11.1% 22
31 CLE -27.9% 0-6 -29.0% 0.6 32 7.8% 4 1.2% 12 3.2% 6
32 NYJ -36.1% 1-5 -39.2% 1.4 31 11.0% 2 -2.5% 28 11.6% 24


113 comments, Last at 25 Oct 2016, 4:55pm

1 Re: Week 6 DVOA Ratings

Buffalo's strong run and easy remaining schedule set the Bills up nicely to finally break their NFL-long postseason drought of 16 seasons.

Well now you've gone and done it.

113 Re: Week 6 DVOA Ratings

Jinxed Shady McCoy? haha

The Week 7 BES Rankings are out already - http://besreport.com/week-7-bes-rankings-2016/

Bills (4-3) rank sixth in BES Overall (3.00), seventh in BES Offense (2.83) and third in BES Defense (3.05). That'll definitely be impacted by any short or long term absence of McCoy.

2 Re: Week 6 DVOA Ratings

Richie Incognito, Football Outsiders Hero, I am just glad Catholic Match Girl isn't around to cast her disapproving gaze on this.

3 Re: Week 6 DVOA Ratings

The top offense in the league is missing its QB and WR who are its two highest paid players.

7 Re: Week 6 DVOA Ratings

Yes, that's...eww.

I'll be honest - I'm glad that's not a normal description for football players. Just imagine the pre-draft discussions; 'That Jurgen McGurp, OLB out of Ohio State - he's quite the feces beater! Great engine on that kid!'

20 Re: Week 6 DVOA Ratings

and Minnesota of course lost Bridgewater and Peterson. Just shows how much football is a team game. And, how hard it is to figure out a single individuals contribution is to winning or losing games.

5 Re: Week 6 DVOA Ratings

I'm trying to figure out how the Cowboys scored a better pass offense than rush offense vs. the packers, and how their rush defense scored so much better than their pass defense in the same game. I'm wondering if it's because of the fumbles.

Does a fumble after a reception count against rush DVOA or pass DVOA? What about a fumble while scrambling, like when Dak just dropped the ball, or when Rodgers got stripped near the goal line? What about when Dak got stripped while attempting to pass?

11 Re: Week 6 DVOA Ratings

1) Green Bay's run defense was ungodly great going into that game, so Dallas' rush offense gets a huge boost from opponent adjustments. I mentioned this in Quick Reads, but the Packers have now allowed eight runs of 10 yards or more all season -- six by Dallas, two by everyone else. Dallas had three 20-yard runs against Green Bay -- nobody else had any. Elliott was only stuffed for a loss three times -- the rest of the league has been stuffed 26 times by Green Bay. Elliott, by himself, is responsible for more than 40 percent of the rushing yards Green Bay has allowed this year. Etc., etc., etc.

2) A fumble on a reception or a sack counts in pass DVOA. A fumble on a scramble also counts as a passing play, with the obvious exception of when a quarterback crosses the line of scrimmage and then becomes a runner.

28 Re: Week 6 DVOA Ratings

For (1), I think you're completely right -- but that's not what DVOA is showing. The DVOA for the game shows Dallas passing offense as being better than rushing offense yesterday which is what I'm trying to figure out.

And Dallas' rushing defense scored insanely well despite good-but-not-great conventional stats.

44 Re: Week 6 DVOA Ratings

What's even weirder is that Dak was the #16 passer and Elliott was the #2 RB. It makes sense, since despite the 9.1 yards/attempt and 3 TDs, Dak had an INT and 2 fumbles... but you'd think that would reflect in the team passing DVOA as well.

Edit: Unless, of course, it's just that passing is just so much more efficient overall than running that a good running performance like that is just not going to register.

45 Re: Week 6 DVOA Ratings

OK, I finally have time to look at this and answer it, and the guess in your edit is correct. Dallas had a pass offense DVOA of 32.6% this week, and a rush offense DVOA of 13.1%. So on the surface it looks like they were better passing the ball. However, the pass offense ranked just 13th for the week, while the rush offense ranked ninth. (I expected them to rank higher there too, but remember that teams like Buffalo, Miami, and Pittsburgh this week were average 6, 7, 8-plus yards per carry.) Because an average passing attack is still more efficient than a good rushing attack.

Reminder: All this data is available to our Premium Subscribers. Click the "Online Store" button at the top of this page for more info.

8 Re: Week 6 DVOA Ratings

How can the vikings' estimated wins be greater than the number of games they have played? Is this stat pro-rated? Guess it must be...

67 Re: Week 6 DVOA Ratings

I'm pretty sure estimated wins is calculated based on weeks completed in the season, not games played. Then at the end it's factored over the whole season.

Edited. Originally said 17 games to a season but I was confusing/melding NFL and CFL as I was listening to some blue bombers sports talk radio about the 18 games.

21 Re: Week 6 DVOA Ratings

"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."

57 Re: Week 6 DVOA Ratings

So at the end of the season, when all teams have had a bye... is it projected as if each team had played 17 games? Or does it reset back to 16 at some point.

I understand the usefulness of being able to compare one team to another, but feel the common sense utility of being able to compare a team's actual wins to estimated wins is muted by doing this. I know we have stats like "eyards" stat. I'll admit all this time I've been seeing Estimated Wins (despite it being defined elsewhere) I had mentally been comparing it to actual wins and making a snap judgement about the coach and team chemistry and other intangibles based on whether it was higher or lower.

I would suggest the statistic be more useful if it were simply pro-rated to the number of games played.

"I was corrupt before I had power!" - Random

62 Re: Week 6 DVOA Ratings

EST wins is a predictive stat in that it strips out useless noise to indicate how a team should have been rewarded so far (not how it was actually rewarded); so its utility is in using the ratio to calculate forward for the remainder of the season.

That the stat allows for a Bye means it's easier to compare team to team at any given point in time, than it would be if it didn't (allow for the Bye).

Comparing it to actual wins for any given team would be possible if you do your own ratio based adjustment to the EST win figure. Ie 3.0 after 6 for a team that's played 5 is actually 2.5 EST wins.

It wouldn't be more useful if the Bye were not factored in - it would just have a different (raw) use, but could also be reverse engineered to allow team to team comparison.

I (strongly) suspect it is in its current format as a useful tool for betting. You want to compare teams at a glance you don't want to be trying to remember who had a Bye and who didn't.

Added: in the spirit of full disclosure the weekly EST wins numbers are why I pay to use this site. That and the opportunity to read grown men arguing with each other on these comments sections.

66 Re: Week 6 DVOA Ratings

Using estimated wins on a uniform baseline (even down to against a league-average schedule), means that what the column is effectively telling you is the Forest Index. So why not publish the Forest Index alongside an estimated wins that's not adjusting for byes?

10 Re: Week 6 DVOA Ratings

The Week 6 BES Rankings are out - http://besreport.com/week-6-bes-rankings-2016/

DVOA and BES again in relative agreement on the top-10 except for Arizona and Green Bay. The BES instead still has the Raiders and the Steelers in its top-10. BES and DVOA also agree on the Skins at No. 11...love the run they're on. Unlike DVOA, the BES isn't quite ready to put the Bills in the top 3 but they did land at No. 5 in this weeks BES Rankings. The BES is also less forgiving toward the Panthers, ranking at No. 29 where DVOA has them 24th. When is Cam Newton going to grow up?

Good stuff as always!

12 Re: Site Layout

Not exactly on topic here.

For all the people who complain about how this site is ancient in terms of layout, I for one appreciate that the site is always consistent.

Earlier today Belichick ripped the Surface Tablets from Microsoft and the NFL's radio equipment. He said there's no consistency and that creates a lot of headaches.

14 Re: Site Layout

In reply to by Rich A

Remaining off topic, you'd think Microsoft, having spent 400 million to have the NFL flak their stuff, woulda' spent a couple million more a year to have their own employees in charge of running the communications in every NFL stadium, in order to avoid the counterproductive message, from the NFL's most prominent coach, saying that the Microsoft products are unreliable, accompanied by the video clip of him slamming the product down on a table.

23 Re: Site Layout

In reply to by Will Allen

He didn't say Microsoft products are unreliable. He said he doesn't like using the tablets and is going back to paper.

I have found when you have 3+ sheets of paper worth of info it is way easier to just look at it on paper than flip back and forth on the tablet.

What he did say was unreliable are the radio systems and that has nothing to do with Microsoft. They only handle the "sideline viewing system" and again his problem appears to be how cumbersome the tablet is to cycle between a bunch of pages rapidly, not any sort of reliability on delivering the data.

It is Bose that provides the headsets and all the communications stuff is happening over small bandwidth ranges and that is why they have all kinds of problems. The following link on NFL tech has a link to an article on "NFL Frequency Coordinators" but to be sure a USMC radio telephone operator would be way more useful to Bill than anyone at Microsoft since they just handle gimmick tablets and XBOX advertisements.


PS - That link says the Browns were at the forefront of this radio tech. Fire up your quip generators!!!

25 Re: Site Layout

In reply to by Jerrytron

It really is a network thing. Belichick doesn't like that the signal drops without warning several times during the game. Photos are immune to dropped signals.

Sounds like the NFL should have a partnership with Cisco as well as with Microsoft. I know some people who know some people...

40 Re: Site Layout

In reply to by RickD

The photos are faxed and the Broncos know fax machines always work when you need them to.

The headsets are wireless radios. Microsoft isn't going to help. The overwhelming majority of the rant is about the radios. The tablets just got the ball rolling.

The link I gave above with the subsequent link to the frequency coordination even has a picture of Bill's Telex TR-1.


Here is a link to the manual where there are RJ-45 connectors on the back but everything I am reading says they are pushed into a very narrow radio band due to FCC regulations and that since they have so many parties whose communications they have to keep separate it becomes a mess.


Some cool charts in there. Or you know, blame Microsoft cause Windows 10 sucks!

58 Re: Site Layout

In reply to by Jerrytron


And it came to pass, as soon as he came nigh unto the training camp, that he saw the decaf, and the end zone celebration; and Bellicheck's anger waxed hot, and he cast the tablets out of his hands, and broke them beneath the mount laurel.

63 Re: Site Layout

In reply to by andrew

Very good.

That reminds me what happened to the guy who used to write Dearest Arlette from Captain Chip.

70 Re: Site Layout

In reply to by Bright Blue Shorts

It mighta' been fun to see that literary tradition continue, what with the qbs under Captain Chip's command this year.

65 Re: Site Layout

In reply to by andrew

If I see a pillar of fire preventing Tyrod Taylor from executing a naked boot in a few weeks, I'll know I've miscast Belichik as Darth Hoodie. Or if Bill starts carrying a staff.

75 Re: Site Layout

In reply to by Will Allen


Moses Bellicheck
Hates Tablets - Hates Tablets
Employs a staff - Employs coaching staff
Made name for himself defeating Ramses - Made name for himself defeating Rams
Overcame Sea of Reeds obstacle - Overcame Ed Reed obstacle
Successor was Joshua - Successor is Josh McDaniels?
Used spies to scout (Numbers 13) - Used spies to scout

76 Re: Site Layout

In reply to by Jerrytron

It sounds a lot like he's calling them unreliable:

"I’ve given them as much time as I can give them. They’re just too undependable for me. I’m going to stick with pictures, as several of our other coaches do as well, because there just isn’t enough consistency in the performance of the tablets. I just can’t take it anymore... I’ve tried to work through the process, but it just doesn’t work for me, and that’s because there’s no consistency to it."

I guess because he uses "undependable" and "no consistency," you can technically say he didn't call them "unreliable." You know, if those are the points you like to chase.

103 Re: Site Layout

In reply to by Karma Coma

Did you read the whole answer he gave? Almost the entire thing is about how complicated and unreliable communication in general is at the stadiums. He discusses everything from problems with infrastructure, to getting them too late on game day to troubleshoot problems, to the league not seeming to have any sort of urgency in fixing problems when they develop.

He makes it very clear that the tablets themselves aren't the actual problem.


46 Re: Site Layout

In reply to by Will Allen

When I read BB's 5+ min rant on the communications systems, I immediately thought it meant it was a Microsoft issue. That's the trouble with prominently attaching your name to the tablets.

The funny part of the rant was that he apologised at the end of it for the long answer!

85 Re: Site Layout

In reply to by Bright Blue Shorts

We really need a good way to measure the quality of the tablet performance separate from the overall IT deployment and management. How dependent on the user education, wireless infrastructure, etc. is the Surface's success and failures? Some may say you could just remove the Surface and put an iPad in it's place and everything would be better, but I think that's too simplistic.

When Belichick says, "They're just too undependable for me," what he really means is, "in 2016, the Surface tablet, within the confines of Gillette Stadium, while connected the the NFL managed networking system, is just too undependable for me."

Unfortunately I don't think there is any single metric that can isolate the success of a given tablet. Perhaps we'll just have to make do with a combination of advanced stats and scouting, with a good measure of common sense factors.

104 Re: Site Layout

In reply to by InTheBoilerRoom

"When Belichick says, "They're just too undependable for me," what he really means is, "in 2016, the Surface tablet, within the confines of Gillette Stadium, while connected the the NFL managed networking system, is just too undependable for me.""

Exactly - and in the full quote he basically says the same thing about the radios.

The difference between the tablet and the radios is that he has an easy way to get similar functionality as the tablet - pictures. There's no way to replicate the radio function

102 Re: Site Layout

In reply to by Rich A

"For all the people who complain about how this site is ancient in terms of layout, I for one appreciate that the site is always consistent."

Consistent? It's layout is broken in a completely different way almost every week.

13 Re: Week 6 DVOA Ratings

I could see it as not being terribly outlandish for the Bills to only lose two more games the rest of the way, including beating the Patriots in Buffalo, thus giving them the tiebreaker over The Empire. That means Darth Hoodie needs to have his minions lose two other games besides the one in Buffalo. At Denver is an obvious candidate, of course. Will Seattle cross the country and beat them at Foxboro? I don't like the Steelers chances this Sunday. I think you really need a great, great, defensive performance to beat the Pats, especially in Foxboro, and I suppose the Ravens are a candidate, health permitting. None of the other Pats' road games look too difficult, but the week 17 game in Miami has to conjure bad memories.

The smart bet for the division remains on The Dark Side, but it wouldn't be the most shocking development to finally see The Death Star destroyed for the first time since 2008.

29 Re: Week 6 DVOA Ratings

Bills schedule: @Mia, NE, @SEA, @CIN, JAX, @OAK, PIT, CLE, MIA, @NYJ
Pats schedule: @Pit, @Buf, SEA, @SF, @NYJ, LARM, BAL, @DEN, NYJ, @MIA

(Yeah, both of them have their third game vs. Seattle - the Pats have the earlier bye week.)

It certainly would be easier for the Pats if they could win in Buffalo.

Let's say Buffalo loses in Seattle and NE loses in Denver. Those would balance out. The Pats are catching a break if Roethlisberger doesn't play on Sunday. Buffalo won't have that advantage. The Bills also have road games at Oakland and at Cincinnati. DVOA hates the Bengals but I'm less convinced. As for Miami and the Jets, those are games that could be dangerous, esp. the road games in Miami. Buffalo has the advantage of getting the Dolphins in Buffalo in December. The Dolphins historically hate cold weather games.

I think if the Pats win in Buffalo they'll get the division.

I'm not really convinced the Bills are a top-tier team. Beating up the Rams and 49ers only goes so far. I could see them going 0-4 in their next four games. Probably won't, but that's a tough stretch. Which is why I find the comment about the easy schedule odd. I guess it's easy because they still have games against Cleveland, Jax, and the Jets. But a small number of games against a few very bad teams isn't as useful as a large number of games against mediocre teams.

32 Re: Week 6 DVOA Ratings

The Bills have a high variance of outcomes I think. I could see them losing 6 of those final games. (@mia, ne, @sea, @cin, @oak, pit) But if they are as good as they've been playing lately, only 1 or 2 more losses is possible.

35 Re: Week 6 DVOA Ratings

Oh, I think if the Bills fail to beat the Pats in their next meeting, the Bills' chances of winning the division drop a large amount. Darth knows this, of course, so I expect he will employ his dark arts to full effect. My guess is that he'll try to force the Bills offense of 2016 to play like the Broncos offense of 2012.

54 Re: Week 6 DVOA Ratings

I peruse some other sites that allow comments to articles where one can just "star" a comment to like it. When I come here, I often want to do that and always spend a few moments looking for some sort of "like" button before I realize that there isn't one at FO.

47 Re: Week 6 DVOA Ratings

I'm pretty certain Buffalo will manage to lose some games to the Jets and/or Dolphins or Browns or someone just at the point we're all beginning to believe that Rex has turned them into contenders and they'll turn into an 11-5 team at best.

Remembering 2003 ... didn't the Patriots bookend their 31-0 opening loss with a 31-0 beating of Buffalo in week 17. Or did I misremember.

16 Re: Week 6 DVOA Ratings

The Vikings dreadful opener against the dreadful Niners last year, in a time zone nightmare special, really pulled them down on the DVOA chart until later in the season. Weird stuff happens with only a 16 game regular season.

18 Re: Week 6 DVOA Ratings

The top half of the league by DVOA has more teams that might just be bad than I can ever remember. PHI, GB, ARI, SD, WAS, BAL, TEN. I wouldn't feel good betting any of those teams to get to 10 wins right now. I'm sure a few of them will. But it feels like the league's middle class has suffered a massive drop off in quality.

19 Re: Week 6 DVOA Ratings

With every passing week the Jets win over the Bills in Week 2 looks stranger and stranger. Incredible how far each team has gone in the opposite direction since that result. I'm not sure I believe the Jets are really the worst team in the NFL, but they're obviously very bad, and perhaps the most disturbing part is how terrible the defense is. If Bowles can't even coax a mediocre performance on defense, which is his strong suit, I think he'll definitely be done at the end of this season, and no one could really argue with that decision.

56 Re: Week 6 DVOA Ratings

I have to admit, I'm shocked that firing Roman had this much effect. The rematch will be interesting.

As for the Jets, yeah, the D has been bad, although it's had missing players and depth really was a question going into the season - I guess it's been answered. The D didn't even play especially well against Buffalo, the O just had everything working and Roman's play calling for Buffalo was freaking AWFUL until the game was really out of reach.

68 Re: Week 6 DVOA Ratings

I have to say that my impulse was to think, from afar, that Ryan was scapegoating Roman without much cause. It seems a lot more likely now that Ryan had been communicating to Roman for a long time as to what kind of game he wanted called, to compliment the rest of the team, and Roman just wouldn't do what his boss told him to do. That's a pretty legitimate reason to fire anybody.

22 Re: Week 6 DVOA Ratings

The issues with NE's defense and DVOA are well traversed, so I wasn't going to mention anything.... until I saw that DVOA has then virtually identical to Carolina. Even granting all the known explanations that seems hard to reconcile.

41 Re: Week 6 DVOA Ratings

NE has the 4th best scoring defense thus far in the league (not per drive, etc.) Hard to understand how that doesn't translate more into efficiency stats. My guess is that it's been similar to that dominant win over the Jets they had in 2011 when DVOA insisted that the Jets were better because the Jets went 3-and-out a ton of times and put together a few long drives with a lot of 3rd down conversions. NE has tended to either force quick punts or surrender long drives this season (at least to the eye test). That translates as poor efficiency, even though it's really not.

59 Re: Week 6 DVOA Ratings

Check out the drive stats. Carolina's D is slightly better at the big three: Yds/Drive, TOs/Drive, and DSR (aka moving the chains). New England's advantage in points allowed seems to come primarily from better field position.

Consider that DVOA is a per play stat. We humans tend to focus on drives and especially whether they score, not on each play separately.

Consider that DVOA is opponent-adjusted. Our human eyes tend not to consider the opponent's strength, but to just see what they see.

And - this one is often forgotten - consider that DVOA by its nature grades on a curve. The New England defense has played with an average of a nine point lead. This means DVOA is comparing it to the "average defense with a nine point lead": comparing it to better than average defenses, in other words.

The first three mean that the Patriots D may not be as good as you think it is.

The last point means it may be better than DVOA says.

106 Re: Week 6 DVOA Ratings

One of the problems that DVOA constantly has with the Patriots defense (both good and bad) is that DVOA largely makes the assumption that defense is context independent, and that is something that BB clearly doesn't agree with.

As an example, there was a game a while back (and I'm having a hard time finding it) that I think was Pats/Miami, and Aaron wrote an article about it. The Patriots led the whole game by a significant margin, and gave up a roughly 80 yard 8+ minute drive to Miami in the 4th quarter while up 20+ points. Miami ran more than they passed on that drive, and chewed up almost all of the time left in the game.

Aaron said that drive had significantly affected New England's defensive DVOA - despite the fact that Miami had probably gone from a small chance of winning to almost no chance of winning.

DVOA is probably the best of the football stats, but its still a very rough tool.

109 Re: Week 6 DVOA Ratings

That doesn't sound like a very typical example, considering the losing team usually wants to hurry in those situations. And I'm rather surprised that it changed their DVOA that much. Seattle gave up two straight TD drives against SF this year in garbage time and it didn't seem to hurt their #1 rank in defense.

In any case, it's still true that teams that allow a lot of garbage-time drives tend to be worse defenses. Maybe in this case their winning probability went down, but if they had been quicker with the drive, and gotten some stops, who knows might have happened. The Patriots nearly gave up a 21-point 4th-quarter lead to Miami just this year.

110 Re: Week 6 DVOA Ratings

They tend to, sure, but not always.

DVOA works on the assumption that every defense is basically the same - and has the same goals, tendencies and philosophies. This simply isn't true - and it means that sometimes DVOA can think pieces of data are predictive when they're not, and aren't predictive when they are.

61 Re: Week 6 DVOA Ratings

Are you counting TD returns on Carolina's offensive turnovers? Are you counting safeties given up? Are you counting punt returns?

That could explain a lot of your confusion, right there.

53 Re: Week 6 DVOA Ratings

Actually, Carolina leads the league to being able to force 3-and-outs on a per-drive basis.

Comparing success rates of the opponents of the two games shows that Carolina has indeed been better (and that's with a harder offensive schedule too), but I think a big difference is that Carolina has given up far more long plays than New England; they've allowed 8 plays of over 40 yards to New England's 1, with both teams having faced about the same number of plays. And since long plays are weighed less, perhaps Carolina isn't as penalized as much as they should be.

There are a few other reasons why New England has worse defensive DVOA numbers than their points per drive surrendered suggests. Their opponents have had five drives that ended on downs (Carolina has none), but all of them gained at least 30 yards, including the 89-yard drive by Cincinnati that ended on the 1. All but one of their turnovers were forced after the opponent gained at least 20 yards on the drive (compared to 4 of 8 for Carolina). And they gave up 63-yard and 33-yard drives that ended in missed FG.

On the other hand, those reasons also help explain why New England has better offensive DVOA numbers than their points per drive suggests. They've had nine drives ending with missed FGs, downs, or TOs, with an average drive length of 41.7 yards, compared to the league average of 26.6 yards.

24 Re: Week 6 DVOA Ratings

A future schedule of 12.5%! Wow, that is brutal for Philadelphia. Only two remaining games against teams with negative DVOA. Median opponent is 8th. Their time in the limelight may have been and wentz.

Because several of the top teams have had bye weeks, the average future schedule is about 0.7% this week. Peculiar!

Preseason projections often talked about the Packers having the league's easiest schedule. What happened there? Really it's nonsense to project strength of schedule prior to the season, except for the Patriots', whose always seems to be friendly because AFC East.

The AFC may be weaker this year, and the AFC South may be the dregs of the conference, but the Bears have contrived to go 0-3 against it. It's the rise of the NFC East which has caught everyone by surprise; how can we have failed to foresee the impact of two rookie QBs, or such quality play by the Washington special teams?

27 Re: Week 6 DVOA Ratings


If you get a chance I would love to know where this current Detroit Lions 26.1% defensive DVOA currently rates as compares to all other years after 6 weeks of an NFL season.

Also, I could be wrong about this but I thought the 2015 New Orleans Saints ended up with the worst defensive DVOA ever over a full season as they also were at 26.1% DVOA. It looks like this current 2016 Lions defense is even worse right now than the winless 2008 Lions as the 2008 group finished with a 24.3% defensive DVOA.

Amazing to think just how far this Lions defensive unit has fallen in the course of just two years as the Lions had the best defensive DVOA for most of the 2014 NFL season (IIRC) only to fall to #3 overall at the end of the regular season with a -13.9% DVOA for the year. Looks like it all adds up to nearly a 40 point swing.

87 Re: Week 6 DVOA Ratings

This isn't the first 6 weeks, but it does show the worst defenses of all time, and the Saints are worst with 26.1%.

49 Re: Week 6 DVOA Ratings

If you watch any Lions game this year, that'll answer your question. Yes, the Saints have given up more points, but the Lions have very good special teams this year, so the opponents on average have poor starting field position...yet still score on the majority of their drives. The defense has also forced very few fumbles and has only intercepted 3 passes.

78 Re: Week 6 DVOA Ratings

You took words out of my mouth as I've watched more or less every Lions game for the past 25 years (no laughing please) and I don't recall them every playing this poorly on defense before.

I had a feeling what we've been watching thus far this season was historically bad and their current defensive DVOA more or less confirms it. They've made Hoyer, Wentz and Keenum look like Marino, Brady and Manning.

Overall opposing quarterbacks are hitting 74% of their passes on the season and have averaged a 119 passer rating.

79 Re: Week 6 DVOA Ratings

Yea, even the 0-16 2008 crew was never this bad against the pass. Luckily, they have a competent offense this time.

They have some good individual players that were part of an excellent defense in 2014 and at least an average defense in 2015, so I really can't figure out what the issue is other than injuries and lack of depth.

Oh, I guess now would be a good time to mention that they play at New Orleans this year. If the defense remains this bad, and Stafford continues to play this well, Vegas might set the over/under to near triple digits.

31 Re: Week 6 DVOA Ratings

How could James White RB not get a football outsiders this week? He is the definition of a football outsider. Far more deserving this week than TWO linemen.

74 Re: Week 6 DVOA Ratings

This has been discussed before. Their FO Madden highlights are done withint what EA wants to do, and are supposed to focus on players whose contributions might not normally get a lot of attention because they're not racking up traditional stats. Hence a lot of lineman. White might be getting some special buff from EA?

36 Re: Week 6 DVOA Ratings

Brady and Belichick have been very consistent over the years about what kind of offense they want to win with.

The last year the Pats had a truly dominant offense (30%+ DVOA) was 2012, featuring two dominant TEs (Gronk and Hernandez), a great small receiver paired with a similar clone (Welker and Edelman), two small shifty RBs with a lead power back (Vereen/Woodhead, and Ridley), a low catch % deep threat (Brandon Lloyd) and an offensive line anchored with 3 high draft picks (Vollmer, Solder, Mankins).

Now it's 2016 and the team has two dominant TEs (Gronk and Bennett), a good small receiver paired with a similar clone (Edelman and Amendola), two small shifty RBs with a lead power back (Lewis, White, and Blount), and a low catch % deep threat (Hogan). The only difference is that their offensive line isn't the same, partially due to Vollmer being injured and partially due to high draft picks being spent on defense since 2011.

91 Re: Week 6 DVOA Ratings

The big difference seems to be that Hogan appears to be a legitimate deep threat. The Pats haven't had that since Moss. Clearly, Hogan isn't going to produce like Moss, but adding some vertical stress to defenses will help the running and screen game considerably.

43 Re: Week 6 DVOA Ratings

Odds are looking up for the "Someone Would Have To Win Bowl" - MIN vs BUF - at 5.8%.

That is all.

50 Re: Week 6 DVOA Ratings

Can a Bears fan explain to me how is it that Brian Hoyer's standard stats (100 passer rating!) and advanced stats (#5 DYAR and #4 DVOA!) look so good, yet the Bears can hardly score any points?

64 Re: Week 6 DVOA Ratings

Redzone problem, almost entirely. Against he Jaguars, they had 5 drives go 30+ yards, and another start at the Jags's 28 yard line, but only got 1 TD. They also only had two 3 and outs the whole game. And no turnovers, which DVOA loves.

Personally, I think Hoyer's arm strength becomes a problem as the field compresses. With smaller windows to throw into, he can't just jam it in. I don't think this covers all the problems though.

55 Re: Week 6 DVOA Ratings

Buffalo is clearly ranked too high because we're obviously in an alternate reality. QBs throwing for 300+ yards is way better than this. That's OK, Bills fans have faith they'll somehow blow it.

69 Re: Week 6 DVOA Ratings

How does DVOA judge special teams? Specifically, for New England's DVOA, which is more important? Their consistently strong field position, or Stephen Gostowski's accuracy problems?

72 Re: Week 6 DVOA Ratings

See this: http://www.footballoutsiders.com/info/methods#specialteams

Each separate aspect of special teams is converted into a net points value. Those net points are given equal weight. Of course, their actual net point value will depend on how well the team does at each aspect.

The special teams DVOA page currently puts the positive point value of New England's great kickoffs and coverage as more than twice what it would need to be to offset the field goal and extra point weakness.

Of course, the kickoffs aren't only about Gostkowski. There's the coverage, too. But, in my opinion, Gostkowski's kickoffs earn him a lot of leeway to get his field goal kicking accuracy back.

73 Re: Week 6 DVOA Ratings

There is a way to factor qb missed(or injured or missing several stars) opponent adjustments in a systematic way: use vegas pre game odds. When Brady is not there, the line naturally drops.

80 Re: Week 6 DVOA Ratings

Been a bills fan my whole life so I usually take these early predictions with a huge grain of salt... Definitely enjoying the season thus far but there is a non-statistical factor to the Bills.. IF there is a way, they will find a way to let you down... god I hope i'm wrong.

82 Re: Week 6 DVOA Ratings

Yeah, I read this: ...the team sitting at No. 3, the top team in the AFC, is probably quite a surprise to most readers. Say hello to the suddenly playoff-likely Buffalo Bills.

And immediately I could hear this voice from the future: It was in Week 7 when the wheels came off, and Buffalo squandered its promising start by finishing the year with a 3-7 run. Another 7-9 season by the only team to play in New York State.

Because that's how this game works.


84 Re: Week 6 DVOA Ratings

The sad thing is that I'm old enough to remember when the Bills somehow won games you were expecting them lose.

Unless it was the Super Bowl, of course.

90 Re: Week 6 DVOA Ratings

Yup. I remember wondering how the Raiders managed to lose their regular season game to the Bills in 1990 - Sunday night game I think. Up by something like 10 points with 6-mins to play and a blocked punt turned it round. Obviously we don't even think about the AFC Championship game.

They did something similar to Denver that year too.

83 Re: Week 6 DVOA Ratings

Just scanned through the playoff odds and didn't know whether to laugh or cry when I saw that Cleveland dropping to 0-6 actually increased their playoff odds.

97 Re: Week 6 DVOA Ratings

Yes, but not directly.

Low variance means that you can be reasonably confident in how the team will play, and so increase confidence that they actually are better (or worse) than their opponent.

High variance means that nobody knows which version of the team is going to show up, and also means that predictions on that team are a fool's errand.

96 Re: Week 6 DVOA Ratings

If not is there a better way to predict future games with the info from the site? I figured variance was a good place to start since it meassures a teams consistency.

98 Re: Week 6 DVOA Ratings

Why would you start with variance? Variance measures a teams' performance consistency with respect to generating DVOA.

What exactly are you trying to predict? Are you trying to bet on specific games, bet on W/L over under? Anyone can correct me if I'm wrong, but my understanding is that DVOA is not explicitly designed to be predictive, but to better tell the story of what actually happened during a football game/season than traditional metrics and then use that data to form conclusion about future probabilities. Of course, this carries some "predictive" implications. But how it is used in that regard really depends on your goals.

If you believe that DVOA more accurately reflects team strength than traditional metrics you can use it to look at matchups between upcoming opponents - but you should be context aware (team DVOA handles injuries ... less than optimally), or you can look at DVOA-based projected final records based on team ratings.

99 Re: Week 6 DVOA Ratings

My goal was picking each game against the spread and I figured by using variance I would be picking the more consistent team.

100 Re: Week 6 DVOA Ratings

If you used the team with the stronger variance last week you would have been 12-3 ATS

111 Re: Week 6 DVOA Ratings

I would be cautious about trusting the volatility measures at this moment. It's really not much of a sample size and dvoa hasn't fully Incorporated opponent adjustments.

Save 10%
& Support Aaron
Support Football Outsiders' independent media and . Use promo code SCHATZ to save 10% on any FO+ membership and give half the cost of your membership to tip Aaron.