Week 11 DVOA Ratings
by Aaron Schatz
Houston stays on top of the Football Outsiders DVOA ratings this week. That's not much of a shock since they didn't play this week. Green Bay remains at number two, but falls back a bit after a close game with Tampa Bay. Obviously you can't fully judge teams based solely on how they played against a single shared opponent, but it's interesting to note that Tampa Bay has now played three of the top teams in the league this year, and they played the Packers closest. They got clobbered in San Francisco by 45 points. They lost to Houston at home by 28 points. But they only lost in Green Bay by nine.
I've got a little graph to help show people why Houston has a higher DVOA rating than Green Bay, and why despite this it is perfectly legitimate to say that Green Bay is better than Houston. The graph below shows the week-to-week single-game DVOA for both teams this year. You can see that Houston has absolutely been on fire in its last four games, starting with the 41-7 whipping of Tennessee in Week 7. However, Green Bay has been really consistent this year. This week's Tampa Bay game was Green Bay's lowest DVOA of the year, and the Packers have been above 0% in all ten games.
Green Bay also is far ahead of the league in the "estimated wins" statistic. Estimated wins uses a formula that gives extra consideration to consistency as well as certain splits of DVOA that correlate strongly with a team winning more games. Estimated wins suggests how many wins a team would have if it played an average schedule with average luck. Green Bay is number one in the league in DVOA on both offense and defense in the second half of close games (score within a touchdown). The Packers are also third in first-quarter offense. All that means that Green Bay has 8.3 estimated wins. Houston and New England are tied for second with 7.3 apiece. However, it is worth noting that total DVOA and weighted DVOA have a stronger correlation with future performance than estimated wins, which is why the main ranking of teams here is in DVOA and not estimated wins.
I will admit that the chances of Green Bay going 16-0 in our playoff odds report seem a little low, only around five percent. Upsets happen all the time in the NFL, but I probably should look in the offseason to make sure our equation isn't expecting upsets to happen a little too much.
I've written numerous times about how Green Bay, while a great team, is not an "all-time great" team. The opposite is also true this year. The Indianapolis Colts, although bad and threatening to go 0-16, are not an all-time horrible team. The current Colts DVOA of -43.7% does not even qualify as one of the ten worst DVOA ratings we've ever measured through Week 11. The worst DVOA ever through Week 11 belonged to the 2-7 Tampa Bay Bucs of 1993, at -61.3%. That team gets a bit of an asterisk, since that was the year of two bye weeks and the team had played only nine games. They won their tenth game and their DVOA improved significantly. The worst team we've ever measured through ten games, regardless of week, would be the 2-8 2005 San Francisco 49ers at -60.1%.
A couple of other notes about teams and players on historic pace:
- The Chicago Bears' current special teams DVOA of 10.6% is the best ever for a team through Week 11. The previous record was 10.2% by the 2004 Buffalo Bills. The highest all-time final special teams DVOA is 10.0% by the 2002 New Orleans Saints. The 2007 Bears at 9.5% were the only other team to ever finish above 9.0%.
- New England tight end Rob Gronkowski (294 DYAR) is on pace to break the all-time record for most DYAR in a season by a tight end. This one comes with a bit of an asterisk. The record was set last year by Antonio Gates, with 371 DYAR. But Gates did that in only ten games before an injury ended his season. So while Gronkowski's current pace would give him 470 receiving DYAR over 16 games, easily passing Gates for the record, we really should be comparing Gronkowski's current numbers to Gates' 2010 season. In that case Gronkowski falls short. Gronkowski's current DVOA of 47.5% is the second-highest DVOA rating ever for a tight end with at least 60 passes. Only three different tight ends with at least 60 passes have ever put up DVOA above 40%: Gronkowski if he continues his current pace, Mark Chmura with 46.6% in 1995, and Antonio Gates with -- this is not a misprint -- 79.5% in 2010.
* * * * *
All individual stats, team stats, and playoff odds report now updated. The playoff odds report this week has two different versions. One version is the normal version based on weighted DVOA. The second version tries to estimate the effect of the injuries to Matt Cassel, Jay Cutler, and Matt Schaub. I've reduced the weighted DVOA for each of those teams by 13.3%, which is a rough estimate of the difference between an average quarterback and a replacement-level quarterback. Obviously, those guys aren't necessarily average quarterbacks -- Schaub is above-average, Cassel below -- the quality of teammates means that Tyler Palko and Matt Leinart probably won't play at precisely replacement level either. (Palko was worse, starting with last night, and I'm assuming Leinart will be better. Actually, Palko may turn into Kyle Orton before the week is over.)
* * * * *
These are the Football Outsiders team efficiency ratings through 11 weeks of 2011, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)
OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. WEIGHTED DVOA represents an attempt to figure out how a team is playing right now, as opposed to over the season as a whole, by making recent games more important than earlier games.
As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.
To save people some time, please use the following format for all complaints:
<team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>
- 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).
77 comments, Last at 27 Nov 2011, 5:01pm
#1 by Bay Area Bengal (not verified) // Nov 22, 2011 - 5:35pm
Have we seen any team fall the way that Buffalo has fallen since week 4? It seems like once their defensive weaknesses were exposed, they have been unable to perform at even a mediocre level ... which is surprising given how explosive their offense started out the season.
#13 by MJK // Nov 22, 2011 - 6:46pm
On offense, in the early part of the season, teams didn't know how good the offensive skill players were apart from Steve Johnson. Now they know...all you have to do to stop Buffalo's offense is double cover Johnson, bring your other safety up to stack the box to stop their running game, and voila! Offense shut down.
On defense, they were living mostly on the strength of turnovers. They were very good at getting their hands up and clogging inside throwing lanes, and getting INTs. Either that wasn't sustainable, or teams have compensated, or turnovers are just fluky (or all of the above), and now their defense, which was "bad but produces lots of turnovers" is just "bad".
#49 by Mike B. In Va // Nov 23, 2011 - 9:38am
Without the turnovers, the D is being revealed as the bad D we expected this year.
The offensive struggles aren't a mystery, given the state of the line. I've seen lots of bad Fitz decision making, more than a little of which seems to stem from both lack of trust in the O line and trying to dig a team that's a notoriously slow-starter on offense out of 21 point holes. The team's strength is the running game, and that's been off the table for the last two games. Make them one-dimensional, and they're pretty easy to handle, seeing as they're down to street FA guys as receivers.
On the plus side, at least Lee Evans is being just as bad in Baltimore as he was in Buffalo last year, so there hasn't been a lot of questioning of that trade. Otherwise, I didn't expect the wheels to come off this fast, but I knew they would. This is a team starting on the right path, though, I think.
#24 by Stats are for losers (not verified) // Nov 22, 2011 - 8:32pm
You know, at the beginning of the season, I was thinking that this had all the hallmarks of a mid-'00s 7-9 season for the Bills, but I didn't want to jinx it and watch 'em go 4-12. At this point, I'll take 7-9 after the last few years of what they've subjected me to. I have many not-so-fond memories from my childhood of watching the Bills wait until the Super Bowl to collapse like this.
Most of what you probably saw from the early-season offensive explosiveness was the week 1 KC blowout, then the comeback against a NE defense that, in addition to being pretty bad, was also missing several (seemingly at the time) key players. Other than that, it was close wins and losses against average to below-average defenses (NE, KC, OAK, CIN, PHI, NYG).
7-9 might be asking a bit much at this point.
#2 by Boots Day // Nov 22, 2011 - 5:43pm
Was there really only one year with two bye weeks? Seems to me like it lasted longer than that, though it was a truly dumb idea.
#53 by jebmak // Nov 23, 2011 - 11:07am
I don't have a problem with it. Extra week of football.
#3 by QQ (not verified) // Nov 22, 2011 - 5:45pm
Pretty amazing how GB can literally almost never trail during games (have still not even trailed or been even tied in the 4th Quarter) and yet not be that far ahead of even of #3-#5 teams in DVOA. While GB is almost always ahead by 7-14 Points during their games, I am guessing their DVOA when ahead by 14+ or so is really bad which hurts their ratings.
#9 by ammek // Nov 22, 2011 - 6:16pm
The problem is that, as soon as they get ahead by more than 14, the Packers' defense has a tendency to give up a really quick touchdown. In week one against New Orleans, they were ahead by 15 points for precisely 3 plays and 1:32. The following week in Carolina, a 17-point lead lasted for 8 plays and 1:37. In Minnesota, they led by 16 points for 5 plays that took 1:30. In San Diego, the Packers established a 21-point lead in the fourth quarter, which lasted a whole 7 plays and 2:55. Against Denver in week four, Green Bay went ahead in the second quarter by 18 points, only to give up a TD on a 4-play drive in 1:51. That's the only one of these games where the Packers re-established a 14+ point lead.
Last week against Tampa the Packers' defense began experimenting with giving up quick touchdowns while ahead by less than 14 points. This is not a championship defense.
#12 by Bay Area Bengal (not verified) // Nov 22, 2011 - 6:29pm
Yeah, I saw QQ's post and my first thought was the game against the Chargers. The Packers' D made 2011 Philip Rivers look like 2008 Philip Rivers.
#16 by QQ (not verified) // Nov 22, 2011 - 6:57pm
It actually IS a Championship Defense since by Definition it is largely the same Players on Defense that won the Super Bowl Last year. Also, being #1 in DVOA in close situations and leading the league in INT's isn't a bad combination.
Are they a lock to Win the SB no, but the Defense while not Great when winning by Double Digits IS Great at closing out games
#45 by ammek // Nov 23, 2011 - 4:48am
Many Packer fans are in denial about the defense, using the same arguments you do. DVOA factors interceptions and late-and-close into the numbers, and the Packer D still ranks 27th. It's not unreasonable to assume that, excluding late-and-close, Green Bay has the worst pass defense in the NFL.
To what extent are those final-drive interceptions 'situational' — a product of being ahead (largely thanks to the offense)? The Patriots are joint-second in interceptions, but nobody thinks they have "playmakers" in the secondary or that their crappy pass defense is particularly opportunistic. It's just a lot easier to pick off mediocre quarterbacks when you're 10 points ahead with two minutes left.
#66 by QQ (not verified) // Nov 23, 2011 - 3:16pm
I don't think many Packers fans are not worried about the Defense. However, in their 16 Game Win Streak,the # of times that the Defense has been able to make Late stops likely has to be something more than simply good fortune/randomness.
Additionally, Defense is not played in a vaccuum. The goal of a Defense is probably not to shut out every opponent but rather to hold them to fewer points than the Offense allows.
I would propose that as likely the Best Team, GB should employ a High Risk/High Turnover causing Defense. The better the team, the more possessions it should want in a game. A High Risk/High Turnover Defense should maximize the number of possessions in a game.
#72 by ammek // Nov 23, 2011 - 6:19pm
What makes it extra difficult to get a tab on the Packer defense is that we've never seen this kind of season-to-season change before.
Green Bay's DVOA for pass defense is down (or, rather, up) by 36.9% compared with 2010. If that continues to the end of the year, it will be the sixth-largest such decline by a team that finished the previous season in the top five.
Let's widen the field a bit. Since 1992, there have been 19 teams whose pass defense DVOA was in the top five one year, and declined by 25 percentage points or more the following season. None of those teams had more interceptions in the second year than the first; in fact, on average these teams picked off 8.4 fewer passes. (The Titans had the same number of interceptions in both 2008 and 2009.) Green Bay is on course for 30.4 interceptions, compared with 24 last season.
Meanwhile, the Packers are giving up 7.0 net yards per pass attempt (NY/A), up from 5.4 last year. That extra 1.6 net yards per pass would be the fourth-most among the sample: only the 1996 Jets, the 2000 Seahawks and the 2001 Titans had a bigger swing. Just behind are the 2007 Ravens and 2006 Redskins (both +1.5 NY/A), and the 2009 Titans (+1.3). It's difficult to think of three teams less similar to the current Packers.
Most of the teams on this list had higher-than-average defensive variance in their second year. (Variance includes rushing of course — I don't have access to the pass-only figure.) The only teams among the least variant were the 1996 Jets again (5th) and the 2006 Redskins again (2nd). They won six games between them. Green Bay is #1 in variance, and 10-0.
Another element with these teams is that their run defense was often well below average, too. As an extreme example, the 2000 Rams dropped to 28th in run defense from 3rd the previous year. The 2002 Jets finished dead last. The Packers aren't that bad yet, but their run defense is in a spin.
A full list of the 19 teams, with their decline in pass-defense DVOA and the number of interceptions in year A and year B in parentheses:
2001 Ten +48.8 (17, 13)
2006 Was +48.3 (16, 6)
2004 StL +47.0 (24,6)
1994 Hou +41.2 (26, 14)
2000 Sea +38.9 (30, 17)
2011 GB +36.9 (24, 30.4)
2009 Ten +36.8 (20, 20)
2010 Buf +36.1 (28, 11)
2005 NE +35.8 (20, 10)
1993 Buf +35.0 (23, 16)
2005 Buf +34.1 (24, 17)
2007 Bal +32.8 (28, 17)
2000 StL +31.5 (29, 19)
2009 Min +31.2 (12, 11)
1995 NE +30.8 (22, 15)
2008 NE +29.7 (19, 14)
1996 NYJ +28.2 (17, 11)
2008 SD +28.2 (30, 15)
1999 NYG +26.2 (19, 17)
2002 NYJ +25.5 (20, 15)
It's probably a coincidence, but there are more teams from the AFC East on this list than from the entire NFC.
#14 by dmstorm22 // Nov 22, 2011 - 6:50pm
Green Bay is constantly ahead, but other than a few blowouts (Denver, Minny #2) have rarely just blown out teams the way most dominant teams have. A lot of their wins are in that comfortable, but still relatively close category.
As for never trailing or being tied in the 4th, that is an insane stat of consistency. The only two other ones I can think of that are in the realm of being that impressive in recent years are the 2005 Colts who won each of their first 13 games by at least a touchdown, and the 2003 Pats who went from Week 13 all the way to the 4th quarter of the Super Bowl without trailing at any point in the game. They trailed against Houston in Week 12, and then the next time they trailed was when they were down 22-21 in Super Bowl XXXVIII.
#23 by LionInAZ (not verified) // Nov 22, 2011 - 8:29pm
I heard similar crap last year about how the Packers never trailed by more than 7 points at any time, and how that made them such a special team. Yet they finished the regular season with a 10-6 record, and yes, I know they won the Super Bowl.
This year we're talking about a new kind of consistency, and yet the Packers have been struggling to stave off teams like Carolina nd Tampa Bay. And yes, I know they're 10-0.
It's a team with obvious vulnerabilities that no so far seems to have figured out how to capitalized on.
"A foolish consistency is the hobgoblin of little minds."
#25 by dmstorm22 // Nov 22, 2011 - 8:41pm
I agree that this is, with the exception of maybe the 2008 Titans, the least great team to start 10-0 in quite some time.
The Packers have just had a great string of success in not giving up that last score to lose games.
Here are drives that opponents had against the Packers in the 4th quarter with a chance to tie the game or take the lead:
NO; Down 34-42, Turnover on Downs
CAR; Down 16-23, Turnover on Downs
MIN; Down 27-33, Turnover on Downs
SD; DOwn 38-45, Punt
SD; Down 38-45, Interception
#26 by QQ (not verified) // Nov 22, 2011 - 8:48pm
"I heard similar crap last year about how the Packers never trailed by more than 7 points at any time, and how that made them such a special team. Yet they finished the regular season with a 10-6 record, and yes, I know they won the Super Bowl.
This year we're talking about a new kind of consistency, and yet the Packers have been struggling to stave off teams like Carolina nd Tampa Bay. And yes, I know they're 10-0."
Setting an NFL ALL TIME Record of 16 Straight Games without ever Even Trailing in the 4th Quarter is a Pretty Clear Sign of Dominance. Even more so when you factor in that the same Team won the Super Bowl and has 1 of the Longest Win Streaks in NFL history, even if you disregard the never trailing.
#31 by Treima // Nov 22, 2011 - 10:19pm
Speaking in ALL CAPS with Randomly Capitalized Words is a Pretty Clear Sign of a Comment Troll.
#34 by dbostedo // Nov 22, 2011 - 11:30pm
"...a Pretty Clear Sign of Dominance."
I disagree. It's probably more of a statistical oddity and coincidence. If I was trying to measure "dominance", I'd put all the weight in the win streak, and not give much creedence to the "never trailing in the 4th quarter" thing.
And then I'd subtract for the fact that while they've won, it wasn't often in dominant fashion. Better teams typically have some more blowout wins, even if they take a loss or two.
#35 by QQ (not verified) // Nov 22, 2011 - 11:46pm
"I disagree. It's probably more of a statistical oddity and coincidence. If I was trying to measure "dominance", I'd put all the weight in the win streak, and not give much creedence to the "never trailing in the 4th quarter" thing.
And then I'd subtract for the fact that while they've won, it wasn't often in dominant fashion. Better teams typically have some more blowout wins, even if they take a loss or two."
It is pretty amazing to not consider a Team Dominant that is basically always in the lead. I think many people would consider Hypothetical Team A who gets a 7-10 Point every game and never relinquishes that lead to be the more Dominant Team than Hypothetical Team B that beats 7 Opponents 50-0 and then in their other 3 games falls behind and needs Last Second FGs/TDs to win the games. Yes Team B will have the better Point Differential but they were much more in threat of losing 3 times.
#38 by RickD // Nov 23, 2011 - 12:38am
Enough with the Caps already!
And asserting your opinion strongly isn't the same thing as making an argument.
#39 by bravehoptoad // Nov 23, 2011 - 12:41am
Maybe QQ's native language is German?
#60 by DGL // Nov 23, 2011 - 12:35pm
Or he's a time traveler from the 18th century?
#67 by QQ (not verified) // Nov 23, 2011 - 3:21pm
If the size of letters that someone uses to type upsets you, then it would seem that a message board is not the place for someone with such Thin Skin.
Additionally, simply refusing to acknowledge an argument is in no way the same was refuting it.
**I see you not surprisingly have not addressed how a team that Never Trails is less Dominant than a team that sometimes Stomps and otherwise needs Late Comebacks to pull out a game they were in Danger of Losing
#64 by Noahrk // Nov 23, 2011 - 2:25pm
If the game is on the line -and #25 just quoted 5 of those for GB- then there's a risk of losing, no matter who happens to have the lead.
Besides, the point is that blowout wins are a better predictor of future success.
We are number one. All others are number two, or lower.
#68 by QQ (not verified) // Nov 23, 2011 - 3:23pm
"Besides, the point is that blowout wins are a better predictor of future success."
Why are you mentioning future success when that has nothing to do with the discussion? The Question is whether GB has been Dominant during their 10 Game Win Streak and 16 Game Win Streak?
#69 by dbostedo // Nov 23, 2011 - 3:26pm
"It is pretty amazing to not consider a Team Dominant that is basically always in the lead."
Again, "always in the lead" when it's a slim lead is not a very good sign of dominance. I would definitely take the second team in your example to be the better team. It's been shown that defeating bad teams soundly is a better sign of future wins than winning by small margins - even if those wins are against good teams.
Close wins are never as good as wins by large margins. There is way too much luck involved in the game to consider that to be predictive, or indicative of the true quality of the team. A team that let's everyone that it plays stay close even toward the end of the game is risking losing. 7-10 points can be overcome by luck. And it's somewhat lucky that the Packers haven't fallen victim to that yet.
To look at it another way, they've lost leads in other quarters. If they had instead lost them in the 4th quarter, but still had the same final scores, I wouldn't think of them as less dominant. So the 4th quarter stat doesn't hold much weight for me. It's just a fun statistic, and doesn't say anything really about the team that the 16 wins don't already.
#73 by QQ (not verified) // Nov 23, 2011 - 9:48pm
At the minimum, among the longest Win Streaks in NFL history, through 16 Games GB has been further away from losing and having their streak snapped than any other team.
During their Streak, the Closest that teams got to ending it was:
1. The Philly Wild Card Game where Philly had No Time Outs and Roughly 30 Seconds when Vick got Intercepted on a 25-30 Yard Pass to Cooper in the End Zone
2. Week 1 vs NO, when NO was Down by 8 at the 1 Yard Line with :00 Left and needed the TD + 2 Point Conversion in order to go to Overtime
*I do think that from a strategy and tactics standpoint that there is a Difference between protecting a lead in Q2 vs Q4. Yes it is preferable to never give up a lead, but giving up a 2nd Quarter Lead is not lower your expected win value nearly as significantly
#75 by dbostedo // Nov 24, 2011 - 6:31pm
Streaks are inherently a bad way to judge the quality of teams players. This is because small changes that obliterate the streak don't really say anything about the team in question... If Green Bay had the same win streak but lost a fourth quarter lead along the way, they would be no less dominant.
But maybe the issue here is that we're using dominant to mean different things? I'm equating dominance with overall team quality (which i think others are up above as well). That's why predictivity comes into it.
#17 by Karl Cuba // Nov 22, 2011 - 7:09pm
I was quite suprised to see that the Packers' defense is number one in variance, which isn't a good thing when you're ranked 27th.
What is a little odd is that they were ranked 18th last week and the gap between the two spots is approximately 6%. I don't get how those two facts combine to make any sense, though it is possible that I don't understand variance.
#43 by ammek // Nov 23, 2011 - 4:25am
Don't forget changes in opponent adjustment. As, for example, the Vikings' offense drops down the rankings, and the Panthers' and Broncos' settle a bit, previous weeks' DVOA are subject to change.
What's weirder is that Green Bay ranks #1 in variance on offense and #1 in variance on defense, yet #2 in overall variance, behind Cincinnati which is #4 in both. Presumably the Packer units are varying at different times, while the Bengals' are varying in tandem.
#47 by M // Nov 23, 2011 - 8:52am
Read comment #21 from the Week 10 DVOA ratings and the subsequent responses. My take is that it is partly due to Special Teams variance (not available in Special Teams detail) and partly due to covariance among the three components (see comments #122 & #133).
Stated another way, let's say offensive DVOA is "up" and defensive DVOA is "down" in a 52-38 game relative to an average game, the composite DVOA will be "average" even though the primary components are both deviating far from average).
Not sure if you'll like to see this presented algebraicly or not...
A = Offensive DVOA
B = Defensive DVOA
C = Special Teams DVOA
#55 by Karl Cuba // Nov 23, 2011 - 12:03pm
Thanks for the help there. So does that mean that the following statement is true? The Packers offense is consistently great while their defense is consistently poor.
#57 by M // Nov 23, 2011 - 12:25pm
I would think that is a valid statement based upon last week's thread. To really see that, though, I think it would be nice to see the weekly DVOA figures for Offense & Defense as an eyeball test. Can anyone who's watched all of each GB game attest that they also continually witness that in on-the-field play?
#65 by Jimmy // Nov 23, 2011 - 2:39pm
If you purchase premium FO you can get all those numbers. Only posting because before I got premium I had no idea what sort of stats were available. Your question is exactly the sort that you get good splits for.
EDIT: Had a quick look at the numbers and basically VOA indicates that their passing game has either been good to amazing every week, that their running game is rather hit and miss and that their defense is generally miss but with no real consistency in how it sucks.
#58 by DisplacedPackerFan // Nov 23, 2011 - 12:32pm
You can't actually make that claim without looking at each data point. Like I said last week I understand the math but since I haven't actually used it for like 15 years I'm not expressing anything in those terms because I'm sure I'll make a stupid mistake.
In this specific case that statement is mostly true. Trying to pull the game numbers from the graph (so again these are approximate total game DVOA numbers based on reading a graph). Then I'll add my subjective comments.
Week 1 (NO) 42% Offense was great, defense was bad
Week 2 (CAR) 31% Offense was good, defense was bad
Week 3 (CHI) 20% Offense was OK, defense was OK
Week 4 (DEN) 28% Offense was great, defense was poor
Week 5 (ATL) 27% Offense was good, defense was good
Week 6 (STL) 17% Offense was OK, defense was poor
Week 7 (MIN) 42% Offense was great, defense was awful
Week 9 (SD) 8% Offense was great, defense was awful
Week 10 (MIN) 48% Offense was great, defense was good
Week 11 (TB) 3% Offense was good, defense was awful
So you get variance, but in general the offense has at worst been above average and generally great. The defense has generally been bad to awful. They don't don't always covary.
OK can be viewed as a average or a bit better. Good means good enough to win the game in most cases. Great is good enough to win even with the rest of the team playing like they are still in college. On the other side you go from poor to bad to awful. Awful play needs great play in other phases for you to have a chance or for the other team to be playing worse than OK.
#4 by JJohnson (not verified) // Nov 22, 2011 - 5:58pm
Is Miami the best 3-7 team ever?*
*By ever I mean as far back as DVOA goes of course.
#20 by LionInAZ (not verified) // Nov 22, 2011 - 8:23pm
No, the Eagles are the best 3-7 team. They just happen to be 4-6 right now.
#33 by DisplacedPackerFan // Nov 22, 2011 - 10:53pm
Well let's see most year all bye weeks are done by week 11 or 12. Miami is at 0.0%. I'm just looking for teams better than -5% in a quick scan in the DVOA weekly rankings.
So we start going backwards and we go here: http://www.footballoutsiders.com/dvoa-ratings/2010/week-11-dvoa-ratings
We see a 3-7 Cleveland squad sitting at #13 with a %8.6 DVOA.
2009 and 2008 didn't have anything close.
2007 had a -2.3% Cincy so not too far off.
2006 nothing of note.
Then I hit week 12 in 2005 and things became less clear to me because of this
IMPORTANT: Beginning with Week 12 of 2005, DVOA is based on second-order opponent adjustments. Until other years are updated with this new system, only first-order DVOA (second table) can be compared to previous seasons.
So while I expect that the database they have will have been updated with their week to week records, I doubt the articles were ever edited. So the next results I don't know. But there is a crazy one in them.
04 and 05 had nothing of note but not so with 2003.
Week 11 2003 coming in ranked #6 at a 19.0% DVOA is your 3-7 Kansas City Chiefs!
That year also gave us the -1.9%, 19th ranked 3-7 Carolina Panthers
http://www.footballoutsiders.com/dvoa-ratings/2004/week-11-dvoa-ratings for the link on that one.
I think that ends it for what I can poke at anyway. There could be more going back to 92 but I don't have access to that data.
So while it looks like they are one of the better 3-7 teams last year had a 3-7 team that was even better and depending on how the changes to DVOA retroactively shake out, 2003 may have had two, one of which was, well, really crazy.
#36 by PaddyPat (not verified) // Nov 23, 2011 - 12:06am
Week 11 2003 coming in ranked #6 at a 19.0% DVOA is your 3-7 Kansas City Chiefs!
You're getting something terribly wrong here. KC in 2003 was 13-3 and looked to be the best team in football through the first 6 weeks or so. That was the year of Dante Hall's total dominance with Priest Holmes in best form etc. etc. They eventually went down to the Manning Colts in that game that had no punts (or almost none, I can't recall) and then the Colts screwed up the punt the following week against NE from... lack of practice? In any event, you must be referring to a different year.
#40 by JJohnson (not verified) // Nov 23, 2011 - 12:45am
He must have been referring to 2004. If you look at the year in the link he provided, its says 2004. Thank you Displaced Packer Fan for the info.
I was looking at year end rankings and saw that Miami finished third in 2002 behind the two teams that reached the Super Bowl that year (TB and Oak) and Miami missed the playoffs. That's why Norv Turner was fired as OC from Miami.
#59 by DisplacedPackerFan // Nov 23, 2011 - 12:35pm
Yeah I typo'd in there and just kept going along the tracks. They don't have weekly numbers back in 2003 so I'm not sure exactly how I got there. I meant 2004, and I'm glad I stuck the link in there. :)
#5 by Richard G (not verified) // Nov 22, 2011 - 5:58pm
I liked the playoff estimated DVOA, but how about including one with Carson Palmer with his career average DVOA instead of -8% driven mostly by one half (I'm guessing Denver, San Diego, Minnesota probably leave him in his career zone).
#22 by Jerry // Nov 22, 2011 - 8:27pm
Your point is reasonable, but it quickly becomes difficult to reflect all the important changes in any given week, and then the arguments start about what the proper adjustments should be. They're probably better off leaving well enough alone; this week's three-QB special is unusual.
#42 by Eggwasp (not verified) // Nov 23, 2011 - 3:56am
I dont think that many teams lose their starting QB and have a retired guy play the second half the following week that often, I think an exception could be made...
#30 by Richard G (not verified) // Nov 22, 2011 - 10:17pm
OK so how about just exclude the KC game if you are gonna do a manual adjustment. I think excluding the KC game would make the projections far more accurate (for the same reason you exclude the preseason, which it essentially was for them).
#6 by tunesmith // Nov 22, 2011 - 6:02pm
Wow, Denver has climbed a lot these last few weeks. It also looks like the first time they've caught up to their estimated wins.
#46 by Danish Denver-Fan // Nov 23, 2011 - 6:10am
Yes, they've climbed a lot. I can't figure out why their weighted DVOA ranking is equal to their total ranking.
#51 by Joshua Northey (not verified) // Nov 23, 2011 - 10:35am
Because they haven't actually climbed a lot.
By far their worst result was against DET in WK 8. That is still full strength. Other than that their results are all more or less close losses or close wins against decent teams. Not much variance between them early in the season and now.
I think you are falling into the trap of overvaluing wins. If they were 0-4 in the last 4 instead of 3-1 (which could easily be the case without changing almost anything about those games) would you still feel they "climbed a lot"?
When I see Denver I see a team with a bunch of results that make them look like they are in the middle third of the league, and one result that puts them in the bottom third. That is about it.
#61 by Danish Denver-Fan // Nov 23, 2011 - 12:36pm
They have climbed a lot in the DVOA-rankings. I'm not overstimating these wins - they could easily be 1-3 in those games without playing better or worse. If memory serves me, it's only a couple of weeks ago they were ~25th in DVOA
#7 by Richard G (not verified) // Nov 22, 2011 - 6:04pm
Oh and also, the 4.6 percent Packers number doesn't seem way out of line. They play the Lions (twice), the Bears, and the Giants-- None of whom are DVOA slouches. Cassel and Cutler's injuries just helped a lot, though. I don't think they'll sweep the Lions; I don't think the Lions are special, but they can definitely keep up in a shootout at home in two nights.
#8 by Anonymous1 (not verified) // Nov 22, 2011 - 6:11pm
Whoot! NE is in single digits defensively!
But what the hell does Buffalo need to do to be closer to 20%? Their rating actually improved this week, which is mind boggling. They might be the worst defense in the entire league.
#10 by Will Allen // Nov 22, 2011 - 6:19pm
It's taken two and a half months, but the Vikings offensive rank is finally starting to get in the area code of what my eyes have been telling me, although not all the way there yet. Having Peterson out (and if they play him with a high ankle sprain that has not healed a full, absolute, 100%, or next August, they should be sued for extreme idiocy) should do the trick of getting them into the 30s by season's end.
#54 by Birdman84 (not verified) // Nov 23, 2011 - 11:30am
Definitely agreed. AD is a competitor and I'm sure he wants to get back as soon as he can. But, being competitive when A) the games don't matter and B) you're injured, is no great virtue.
#11 by bravehoptoad // Nov 22, 2011 - 6:20pm
Boy, special teams ratings seem to swing a lot. CHI and SF have been swapping the top spots all season, and one game with 2-blocked field goals and 1 missed field goal and SF drops 3+% on the season.
#15 by kbukie // Nov 22, 2011 - 6:54pm
Those adjusted playoff odds made me feel 77.7% better about the rest of the season.
#18 by Karl Cuba // Nov 22, 2011 - 7:15pm
I feel like I should be using the template here but 19th seems low for the 49ers offense to me. Could it be that they are being dragged down by Harbaugh's preference for protecting a lead by slamming Gore into the line? That the pass offense ranks 9th and the run game, which most would regard as a strength, ranks 22nd would at least suggest that defenses have been stacking the box and daring the niners to throw.
#19 by Karl Cuba // Nov 22, 2011 - 7:21pm
Comparison with Baltimore's 23rd ranked offense reveals a little of what may be going on; the Ravens' run game is at -2% and the passing game is at 15% compared to -4.3 and 29% for the 49ers. I think that this must be reflecting usage or the 49ers would rank much higher than the Ravens. (maybe, I could well be talking utter tripe)
#21 by bravehoptoad // Nov 22, 2011 - 8:24pm
The 49ers have only had one blowout this season, right? Except for that game, the offense hasn't scored over 30 points. The Ravens can seem pretty dynamic about every other game.
With a couple more of those blowouts their offense would look a lot better on the stat sheet, and they could have had those by this point considering they've played Washington, Cleveland, and Arizona in recent weeks. I agree with Harbaugh -- they should have scored 40 points last Sunday.
It does seem that, since losing Morgan in the Tampa game, Harbaugh has tightened up like a fist when playing with a lead. Their "we're-ahead" offense is a pretty grim affair; there's not much "hurray!" in it. Or to use contemporary sports terms, when they're ahead, they don't swagger.
edit: I should say, they don't "have swagger." Did the sports world forget this is a verb?
#27 by Rhombus (not verified) // Nov 22, 2011 - 8:54pm
Really good point there. In the Niners 3 games against Arizona, Washington, and Cleveland, none of the teams had really any chance of coming back. They were essentially blowouts, but the Niners offense either sputtered, or just seemed to stop trying to score. The Harbaugh-led Niners tend to sit on leads rather than expand on them, with the exception of the Tampa Bay game, especially in the 4th quarter. This has proved effective so far, but I would really like to see another dominating start-to-finish type game. And so would DVOA, I think.
#29 by artj (not verified) // Nov 22, 2011 - 10:02pm
I think the Niners passing offense has sputtered for most of the season. Morgan was stepping into the #1 role; since his loss, Smith hasn't had a really solid, reliable threat downfield. The passing O is simply weak and underdeveloped. Some of this is probably on Smith and some on the receivers. I think Harbaugh would love to pile on the points. Remember how Morgan got hurt: going for the end zone with less than 3 minutes to go in a game they were leading by... was it already 45 at the time, or only 38?
#37 by bravehoptoad // Nov 23, 2011 - 12:16am
Ah, the good old days. I wish he'd do that again, instead of running Dixon into the line nine times.
This is the most tantalizing team. Alex seems to get less accurate the more comfortable the game is, and his receivers seem to stop trying as hard to get those spraying balls as they do earlier in the game. I keep thinking, oooh, three slightly slightly better thrown (or slightly better caught) passes, and we win by 28 instead of scrimping by.
#28 by akn // Nov 22, 2011 - 9:58pm
Looking forward to seeing where the Bears ST finishes against largely below average ST teams (DEN surprisingly being ranked #9). I wonder how much of the Bears ST DVOA is from returns (Hester and his blockers) vs the Bears' amazing punt coverage unit (allowing an incredibly low ~3.5 yards per punt return).
#41 by JCutler6 // Nov 23, 2011 - 2:44am
On the other side of the spectrum, the kick returns have been sub-par all year (probably a league wide trend though) and the kick coverage, while still very good, has the tendency to give up a big return at bad times
Podlesh and the punt coverage are the best I've seen from the Bears since Toub took over
#32 by TomC // Nov 22, 2011 - 10:47pm
Look at that remaining schedule DVOA for San Fran. Are we going to be wondering forever what might have been if not for a couple of crazy plays in September against Dallas?
[edit: probably not. Turns out playing the rest of the NFCW makes that stat insensitive to the fact that you also have to play BAL and PIT.]
Also, the splits for the Bears and SF get closer to identical every week. Who does that Harbaugh guy think he's coaching?
#52 by jedmarshall // Nov 23, 2011 - 10:57am
Yeah I was a little surprised by the 32nd ranked schedule as well knowing they played Baltimore this week. Turns out when 2 of your final 6 are against the Rams, you have an easy schedule.
#44 by Anonymous(not … (not verified) // Nov 23, 2011 - 4:34am
Is it time for an analysis of Alex Smith yet? Is he fixed?
#48 by Kevin from Philly // Nov 23, 2011 - 9:24am
I see no mention of kids in his team bio, so maybe.
#56 by Karl Cuba // Nov 23, 2011 - 12:07pm
His first kid was born just before he began running the niners' unofficial lockout training camp in San Jose.
#50 by Zieg (not verified) // Nov 23, 2011 - 9:50am
A quick skim shows Tampa to currently have the hardest past schedule and overall schedule (past + future = 12.1). This got me wondering what the hardest schedules of all-time are? Not saying Tampa is up there (St Louis is already pretty close at 11.6), just wondering at where they fall.
#62 by M // Nov 23, 2011 - 12:47pm
The 2004 Cleveland Browns had the hardest schedule (so far) in the DVOA era, coming in at a whopping 14.2%. The next hardest seem to be the 2000 NY Jets at 12.1% and the 2010 Buffalo Bills at 11.8%.
It seems unlikely that Tampa will end up breaking the record unless the teams on their schedule improve over the remainder of the season.
While DVOA hasn't gone back that far yet, I suspect the 1991 Phoenix Cardinals or LA Rams will challenge the Browns.
#63 by Jesus (not verified) // Nov 23, 2011 - 1:26pm
My goodness! Kellen Clemens is back & ready to set the world on fire !
#70 by LittleBigAnon (not verified) // Nov 23, 2011 - 4:50pm
How did the Cowboys fall in the rankings, despite winning on the road, and the Giants climb in the rankings despite losing at home and looking god awful while doing so?
#71 by Thomas_beardown // Nov 23, 2011 - 4:58pm
The Giants rating fell from last week. They were 19.1% last week, 18.1% this week. So they climbed in the rankings because other teams fell even more.
#74 by AnonymousD (not verified) // Nov 24, 2011 - 11:24am
Dallas' pass defense wasn't good, and when adjusted for opponent it's poor.
#76 by richabbs (not verified) // Nov 26, 2011 - 12:06pm
OK, the Ravens have beaten the Texans, the 49ers, the Jets, and the Steelers (twice), yet all of those teams are rated higher in DVOA. Maybe this DVOA isn't all that its cracked up to be. Sure, the Ravens had brainfart games on the road against the Titans, the Jags and the Seahawks (always tough to play when you travel across the country), but, other than the Packers, there isn't a team that is better than them in the NFL. I predict Baltimore will beat the Patriots in the AFC championship game and the Packers will beat the Saints in the NFC, which will set up a great Super Bowl matchup between the Packers #1 offense and the Ravens #1 defense.
#77 by dbostedo // Nov 27, 2011 - 5:01pm
"...the Ravens had brainfart games on the road against the Titans, the Jags and the Seahawks..."
You answered your own question. And please use the template next time...