Week 12 DVOA Ratings
by Aaron Schatz
Our top two teams on bye means our top two teams stay the same this week, with the Kansas City Chiefs still No. 1 in DVOA, followed by the Los Angeles Rams. The New Orleans Saints move up to the No. 3 spot thanks to a big Thanksgiving night win over the Atlanta Falcons. In weighted DVOA, which drops the strength of September games, the Saints and Rams are essentially tied at No. 2.
I've written a lot in the last couple weeks about the disagreements between DVOA and other advanced metrics, which now tend to have the Saints at No. 1. But the difference between DVOA and other advanced metrics may be bigger for the Chiefs than it is for the Saints. And that huge difference between the Chiefs and the Rams/Saints in DVOA is also responsible for the huge difference between the Chiefs and Rams/Saints in our odds for winning Super Bowl LIII.
To see how much, I ran an additional simulation this week which gave the Chiefs the same weighted DVOA as the Rams and Saints (31.1%). In that simulation, Kansas City's chances of representing the AFC in Super Bowl LIII drop from 53.7 percent to 42.5 percent. We still would give the Chiefs the strongest odds of winning the Super Bowl, because unlike the Saints or Rams, they don't have to play the Saints or Rams on their way to Atlanta. But they wouldn't have the strongest odds by much:
| Team | Original Simulation |
If KC = NO/LAR Simulation |
| KC | 31.9% | 22.4% |
| LAR | 20.4% | 22.0% |
| NO | 19.9% | 21.5% |
| NE | 5.4% | 7.0% |
| CHI | 5.1% | 5.9% |
| LAC | 4.8% | 6.3% |
| PIT | 4.0% | 5.0% |
| HOU | 3.3% | 4.0% |
| Others | 5.2% | 5.9% |
The difference between the Rams and the Saints here has to do with remaining schedule, even though the Saints currently have the No. 1 seed thanks to the head-to-head tiebreaker. The Rams have the easiest remaining schedule in the league by average DVOA of opponent, while the Saints' schedule ranks 14th. That being said, the Rams have the hardest remaining single game, playing at Chicago in Week 14. The Saints' hardest remaining opponent, Pittsburgh, is a little bit below the Bears in DVOA and will come to New Orleans instead of the Saints going on the road.
While we're on the subject of the Saints, let's talk about the Saints defense. Last year, after a slow start, the Saints ended up with one of the best defenses in the NFL. The same thing seems to be happening this year. Thanks in part to the opponent adjustments for playing Atlanta, the Saints had the highest single-game defensive DVOA of Week 12 at -50.7%. It was their best defensive performance of the year, and emphasizes a major change in defense since their Week 6 bye. In Weeks 1-5, the Saints ranked 26th with 9.4% defensive DVOA. (Remember, defense means more scoring so a positive number is worse.) In Weeks 7-12, the Saints rank fifth in the NFL with -10.7% DVOA.
I took a look through other numbers on both offense and defense and found teams that have improved significantly in recent weeks. First, on offense:
- Indianapolis was at -12.1% DVOA on offense through Week 6, 24th in the NFL. Since Week 7, the Colts have 26.5% offensive DVOA, fifth in the NFL.
- Pittsburgh's turnaround was earlier. For September, through Week 4, the Steelers had 0.6% offensive DVOA, 16th in the NFL. Since Week 5, the Steelers rank fifth with 22.0% DVOA.
- Cleveland's inflection point is actually Week 8 against Pittsburgh, before Todd Haley and Hue Jackson were let go. In Weeks 1-7, the Browns had -26.8% offensive DVOA, which ranked 30th. For Weeks 8-12, the Browns have 16.5% offensive DVOA, which ranks eighth. Every Cleveland game since Week 8 has had a better offensive DVOA than each of the seven games prior to Week 8.
And two improved defenses besides New Orleans:
- Minnesota ranked 25th through Week 5 with 6.7% defensive DVOA. In Weeks 6-12, Minnesota has the No. 1 defensive DVOA in the league at -23.1%.
- Denver ranked 22nd through Week 5 with -0.2% defensive DVOA. In Weeks 6-12, Denver has the No. 2 defensive DVOA in the league at -22.4%.
There are also defenses that have faltered over the last few weeks, including Jacksonville, Philadelphia, Seattle... and Buffalo. Although the Bills are second in defensive DVOA for the entire season, that splits into -24.6% in Weeks 1-6, first in the league, and -2.8% in Weeks 7-12, 10th in the league.
Of course, not everything in Buffalo is getting worse. Their offense has suddenly gotten much, much better. So much better. Prior to Week 10, the Bills didn't have a single game with offensive DVOA above -20%. But Buffalo managed to score a single-game offensive rating above 20% in both Week 10 and Week 12. That improvement has them almost sliding off our list of the worst offenses in DVOA history.
| WORST OFFENSIVE DVOA THROUGH 11 GAMES, 1986-2018 |
||
| 1992 | SEA | -44.3% |
| 2005 | SF | -43.7% |
| 2010 | CAR | -43.1% |
| 2002 | HOU | -42.3% |
| 2013 | JAX | -39.9% |
| 2018 | ARI | -39.6% |
| 2004 | CHI | -39.3% |
| 2004 | MIA | -39.2% |
| 1997 | NO | -38.9% |
| 2018 | BUF | -37.9% |
| 2008 | STL | -35.8% |
| 1992 | IND | -35.5% |
I feel like the Bills have really ruined my fun, because it was kind of cool seeing a team put up a (when adjusted for era) historically bad offense. But this is not something I have against the Bills in particular. If the Bills want to continue playing actual good offense and the Arizona Cardinals want to suck extra hard in December to get down below -40%, that would be fun to watch too. And by "fun to watch," I mean fun to monitor the stats, not fun to actually watch on television. Arizona has yet to put up a positive offensive DVOA in any game this year, although the Cardinals' 28-18 Week 5 win over San Francisco comes very close.
* * * * *
 |
Once again this season, we have teamed up with EA Sports to bring Football Outsiders-branded player content to Madden 19 on a monthly basis. Today, we get to announce the Football Outsiders November players for Madden Ultimate Team on consoles, which will go live at 10:30am Eastern on Sunday. These players will also go live in Madden Overdrive for mobile devices sometime soon.
- QB Baker Mayfield, CLE: Second in passing DVOA for November (65-of-88, 771 yards, 9 TD, 1 INT).
- HB Phillip Lindsay, DEN: Third in rushing DYAR for November (42 carries, 249 yards, 3 TD).
- TE Nick Vannett, SEA: Fifth among tight ends in receiving DYAR for November (10-of-14, 99 yards, 2 TD).
- LG Quenton Nelson, IND: Colts ranked second with 5.97 adjusted line yards per carry on left-side runs in November; allowed just one sack.
- RG Larry Warford, NO: Saints ranked fourth with 5.46 adjusted line yards per carry in November; allowed just two sacks.
- RE Mario Addison, CAR: Tied for third in NFL with 8 defeats in November (7 TFL including 3 sacks, plus tackle to prevent third-down conversion).
- RE Chris Jones, KC: Led NFL with 10 defeats in November: 5 sacks, two third-down PDs, two run TFL, and run tackle to prevent third-down conversion.
- LOLB Leighton Vander Esch, DAL: Second in NFL with 45 total plays made and 6 passes defensed in November.
- CB Steven Nelson, KC: Only allowed 4.4 yards per pass in November by Sports Info Solutions charting, 4 PD, 2 INT.
- CB Coty Sensabaugh, PIT: Only allowed four catches for 45 yards (and 3.2 yards per pass) in November according to Sports Info Solutions charting.
- K Justin Tucker, BAL: 8-for-8 including two FG over 45 yards; 71 percent touchback rate.
- P Tress Way, WAS: Second in NFL with 44.7 net average in November; first in league with 7.4 points of estimated field position value on gross punts.
(Note: Late change here, as we had to remove Roquan Smith from our list of players for this month, we're subbing in Kansas City cornerback Steven Nelson.)
* * * * *
Playoff odds are currently updated through Week 12, as well as the FO Premium DVOA database and snap counts.
* * * * *
These are the Football Outsiders team efficiency ratings through 12 weeks of 2018, 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.
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.
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>
| RK | TEAM | TOTAL DVOA |
LAST WEEK |
WEI. DVOA |
RANK | W-L | OFFENSE DVOA |
OFF. RANK |
DEFENSE DVOA |
DEF. RANK |
S.T. DVOA |
S.T. RANK |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | KC | 41.6% | 1 | 42.4% | 1 | 9-2 | 38.4% | 1 | 5.2% | 25 | 8.5% | 1 |
| 2 | LAR | 32.6% | 2 | 31.1% | 2 | 10-1 | 31.5% | 2 | -1.1% | 17 | 0.1% | 17 |
| 3 | NO | 27.1% | 4 | 31.1% | 3 | 10-1 | 23.4% | 4 | -1.5% | 14 | 2.2% | 9 |
| 4 | LAC | 26.2% | 5 | 28.2% | 4 | 8-3 | 26.4% | 3 | -5.5% | 9 | -5.7% | 32 |
| 5 | CHI | 21.1% | 3 | 20.9% | 6 | 8-3 | 2.7% | 14 | -22.5% | 1 | -4.0% | 29 |
| 6 | DEN | 19.7% | 7 | 21.8% | 5 | 5-6 | 8.8% | 9 | -12.5% | 3 | -1.6% | 23 |
| 7 | PIT | 14.5% | 6 | 17.0% | 7 | 7-3-1 | 14.2% | 6 | -2.4% | 12 | -2.1% | 25 |
| 8 | BAL | 14.2% | 9 | 12.7% | 9 | 6-5 | 1.6% | 15 | -9.9% | 5 | 2.7% | 6 |
| 9 | NE | 11.6% | 10 | 13.5% | 8 | 8-3 | 12.3% | 7 | 0.8% | 19 | 0.2% | 16 |
| 10 | GB | 11.5% | 8 | 11.5% | 10 | 4-6-1 | 14.7% | 5 | 1.6% | 20 | -1.5% | 22 |
| 11 | SEA | 9.5% | 12 | 10.3% | 11 | 6-5 | 5.1% | 11 | -2.8% | 11 | 1.5% | 11 |
| 12 | HOU | 7.7% | 14 | 8.5% | 14 | 8-3 | -6.6% | 21 | -12.0% | 4 | 2.2% | 8 |
| 13 | IND | 7.0% | 11 | 8.7% | 13 | 6-5 | 4.5% | 12 | -1.2% | 16 | 1.4% | 13 |
| 14 | MIN | 6.7% | 13 | 8.9% | 12 | 6-4-1 | -0.5% | 17 | -9.8% | 6 | -2.6% | 26 |
| 15 | CAR | 5.4% | 15 | 5.4% | 15 | 6-5 | 10.7% | 8 | 5.8% | 26 | 0.6% | 15 |
| 16 | JAX | -5.4% | 17 | -7.6% | 17 | 3-8 | -15.4% | 28 | -7.0% | 7 | 3.0% | 5 |
| 17 | WAS | -7.0% | 16 | -8.3% | 18 | 6-5 | -11.2% | 25 | 0.1% | 18 | 4.4% | 3 |
| 18 | MIA | -7.3% | 20 | -9.0% | 19 | 5-6 | -8.8% | 23 | 2.8% | 21 | 4.3% | 4 |
| 19 | CIN | -8.1% | 19 | -10.6% | 23 | 5-6 | 0.4% | 16 | 9.3% | 28 | 0.7% | 14 |
| 20 | NYG | -8.2% | 21 | -7.4% | 16 | 3-8 | -2.2% | 18 | 8.6% | 27 | 2.6% | 7 |
| 21 | DAL | -9.6% | 22 | -9.4% | 20 | 6-5 | -9.4% | 24 | -1.5% | 13 | -1.7% | 24 |
| 22 | ATL | -9.6% | 18 | -10.2% | 21 | 4-7 | 8.6% | 10 | 19.6% | 32 | 1.4% | 12 |
| 23 | CLE | -11.9% | 25 | -10.4% | 22 | 4-6-1 | -12.7% | 26 | -6.3% | 8 | -5.5% | 31 |
| 24 | PHI | -12.0% | 23 | -11.6% | 24 | 5-6 | -6.3% | 20 | 4.2% | 24 | -1.4% | 21 |
| 25 | TEN | -14.9% | 24 | -14.4% | 25 | 5-6 | -13.2% | 27 | 3.7% | 22 | 1.9% | 10 |
| 26 | TB | -18.5% | 28 | -18.9% | 26 | 4-7 | 3.3% | 13 | 18.8% | 31 | -3.0% | 27 |
| 27 | NYJ | -18.7% | 27 | -19.4% | 27 | 3-8 | -25.7% | 30 | -1.2% | 15 | 5.8% | 2 |
| 28 | SF | -24.0% | 26 | -25.2% | 30 | 2-9 | -16.9% | 29 | 4.1% | 23 | -3.1% | 28 |
| 29 | DET | -24.1% | 29 | -22.8% | 29 | 4-7 | -7.2% | 22 | 16.1% | 30 | -0.8% | 19 |
| 30 | BUF | -24.6% | 31 | -21.8% | 28 | 4-7 | -38.3% | 31 | -14.8% | 2 | -1.0% | 20 |
| 31 | OAK | -27.2% | 30 | -30.1% | 31 | 2-9 | -6.1% | 19 | 15.8% | 29 | -5.3% | 30 |
| 32 | ARI | -37.4% | 32 | -37.9% | 32 | 2-9 | -39.6% | 32 | -2.9% | 10 | -0.7% | 18 |
- 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).
| RK | TEAM | TOTAL DVOA |
W-L | NON-ADJ TOT VOA |
ESTIM. WINS |
RANK | PAST SCHED |
RANK | FUTURE SCHED |
RANK | VAR. | RANK |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | KC | 41.6% | 9-2 | 38.6% | 10.4 | 1 | 3.4% | 9 | -0.9% | 16 | 9.9% | 13 |
| 2 | LAR | 32.6% | 10-1 | 31.9% | 9.9 | 2 | 5.7% | 5 | -15.3% | 32 | 6.0% | 2 |
| 3 | NO | 27.1% | 10-1 | 29.7% | 8.2 | 3 | -2.8% | 21 | -0.5% | 14 | 9.3% | 10 |
| 4 | LAC | 26.2% | 8-3 | 32.6% | 7.4 | 6 | -5.8% | 29 | 16.4% | 3 | 8.3% | 7 |
| 5 | CHI | 21.1% | 8-3 | 26.4% | 7.7 | 4 | -10.5% | 32 | 3.7% | 10 | 12.7% | 15 |
| 6 | DEN | 19.7% | 5-6 | 18.1% | 7.5 | 5 | 9.5% | 1 | -9.0% | 28 | 16.2% | 23 |
| 7 | PIT | 14.5% | 7-3-1 | 13.9% | 6.4 | 10 | 2.7% | 10 | 5.9% | 5 | 17.0% | 25 |
| 8 | BAL | 14.2% | 6-5 | 12.0% | 6.9 | 8 | -1.3% | 18 | 5.6% | 6 | 16.7% | 24 |
| 9 | NE | 11.6% | 8-3 | 13.0% | 6.5 | 9 | -0.6% | 16 | -5.9% | 22 | 14.3% | 20 |
| 10 | GB | 11.5% | 4-6-1 | 7.9% | 6.9 | 7 | 0.1% | 13 | -13.7% | 31 | 13.3% | 18 |
| 11 | SEA | 9.5% | 6-5 | 7.9% | 6.4 | 12 | 4.6% | 8 | -7.4% | 26 | 7.2% | 5 |
| 12 | HOU | 7.7% | 8-3 | 13.5% | 6.4 | 11 | -4.9% | 26 | -8.2% | 27 | 5.5% | 1 |
| 13 | IND | 7.0% | 6-5 | 17.2% | 6.4 | 14 | -9.6% | 31 | -6.1% | 23 | 15.4% | 21 |
| 14 | MIN | 6.7% | 6-4-1 | 4.9% | 6.4 | 13 | -3.4% | 24 | 2.2% | 13 | 6.3% | 3 |
| 15 | CAR | 5.4% | 6-5 | 14.1% | 6.0 | 15 | -5.4% | 28 | 2.8% | 12 | 18.5% | 27 |
| 16 | JAX | -5.4% | 3-8 | -6.2% | 4.1 | 25 | -0.5% | 15 | -2.9% | 20 | 8.1% | 6 |
| 17 | WAS | -7.0% | 6-5 | -1.9% | 5.2 | 16 | -3.1% | 23 | -10.5% | 30 | 8.4% | 8 |
| 18 | MIA | -7.3% | 5-6 | -7.3% | 4.2 | 23 | -4.8% | 25 | -7.2% | 25 | 15.4% | 22 |
| 19 | CIN | -8.1% | 5-6 | -11.6% | 5.1 | 17 | 7.0% | 2 | 4.3% | 8 | 19.2% | 28 |
| 20 | NYG | -8.2% | 3-8 | -3.9% | 4.6 | 19 | -5.3% | 27 | -0.7% | 15 | 10.0% | 14 |
| 21 | DAL | -9.6% | 6-5 | -0.1% | 4.0 | 26 | -6.0% | 30 | -0.9% | 17 | 6.8% | 4 |
| 22 | ATL | -9.6% | 4-7 | -8.8% | 4.5 | 20 | -0.1% | 14 | -5.0% | 21 | 9.2% | 9 |
| 23 | CLE | -11.9% | 4-6-1 | -6.7% | 4.1 | 24 | 5.1% | 6 | 7.8% | 4 | 9.6% | 11 |
| 24 | PHI | -12.0% | 5-6 | -9.2% | 4.4 | 22 | -2.6% | 20 | 3.4% | 11 | 9.8% | 12 |
| 25 | TEN | -14.9% | 5-6 | -16.3% | 4.7 | 18 | 1.4% | 12 | -6.4% | 24 | 21.3% | 30 |
| 26 | TB | -18.5% | 4-7 | -19.9% | 3.2 | 28 | -1.1% | 17 | 5.5% | 7 | 13.6% | 19 |
| 27 | NYJ | -18.7% | 3-8 | -20.0% | 2.9 | 29 | -1.3% | 19 | -1.7% | 18 | 24.0% | 31 |
| 28 | SF | -24.0% | 2-9 | -22.6% | 2.9 | 30 | -3.1% | 22 | 18.5% | 2 | 13.1% | 17 |
| 29 | DET | -24.1% | 4-7 | -22.1% | 3.5 | 27 | 2.5% | 11 | -2.2% | 19 | 21.2% | 29 |
| 30 | BUF | -24.6% | 4-7 | -32.8% | 4.4 | 21 | 6.1% | 4 | -9.2% | 29 | 31.8% | 32 |
| 31 | OAK | -27.2% | 2-9 | -31.2% | 2.6 | 31 | 5.0% | 7 | 21.8% | 1 | 17.8% | 26 |
| 32 | ARI | -37.4% | 2-9 | -41.0% | 2.0 | 32 | 6.8% | 3 | 4.0% | 9 | 12.9% | 16 |
Comments
95 comments, Last at 03 Dec 2018, 8:39pm
#1 by Kelly Mamer // Nov 27, 2018 - 6:25pm
I love that Seattle is 11 in DVOA, Weighted DVOA, Offense, Defense, and Special Teams. Feels to me like they're about the 11th best team in football too. They may as well be #1 in Variance (sadly, they're only #5).
#2 by ammek // Nov 27, 2018 - 6:43pm
Wait, the best team in the NFC East by weighted DVOA is the Giants? Then again, I watched the Thanksgiving showdown in Arlington and it made my eyes hurt.
So, real wins versus estimated wins. Two teams stand out. Denver (2.5 fewer of the former than the latter) has played the hardest schedule. Green Bay (2.4 fewer) doesn't have that excuse. By season-long DVOA it's the fourth best team in the conference, yet nine NFC teams have more wins and better playoff odds. To read the hugely negative Packer press you'd never imagine the team was 5th in total offense nor 9th in passing by DVOA. Yet it's not only that writers and fans don't understand how to adjust for defense (the Packers passing offense is 14th by unadjusted VOA, narrowly ahead of the NFC East's awesome checkdown Giants). The crucial stat might be variance, where the Packers are #1 on offense, reflecting the groundhogdayish experience of seeing Rodgers take another third-down sack.
Meanwhile, the Patriots are on course to have one or more extra real wins vis-à-vis estimated wins for the sixth straight season. Ho hum.
#29 by big10freak // Nov 28, 2018 - 7:58am
I don't know why I get the barbs for posting a fairly innocuous statement. I am not some drive by Packer poster aka oaktoon taunting other teams/posters, etc when things go well and then disappearing when convenient.
#74 by big10freak // Nov 28, 2018 - 7:42pm
So why not write that versus 'playing our tiny violins for you' which is snark in any forum.
And I am not 15 years old who has only known 12 at qb. I sat through my share of Bart Starr, Forrest Gregg, Lindy Infante ineptitude. But hey you got in your shot. Enjoy.
#57 by xydux // Nov 28, 2018 - 1:34pm
Because people who aren't fans of the Packers find this hilarious.
Incidentally, as somebody who doesn't root for the Lions, I still think they screwed up by firing Caldwell (he actually has the best winning percentage of any Lions coach since the 60s aside from an interim guy who went 4-3...which itself speaks volumes about Lions coaching), BUT every time I watched a Lions game with him coaching, I found myself immensely frustrated by the decisions/play of the Lions.
#84 by Will Allen // Nov 29, 2018 - 9:57am
All snark aside, there have been some Rodgers' throws recently that were very reminiscent of Bad Smokin' Jay. Open receiver, Rodgers unpressured, but inaccurate, while displaying mechanics that would be charitably described as sloppy. Now, I don't want to overstate the case here; Rodgers would have to travel parsecs to get into the neighborhood of The Indifferent One. However, the best argument for replacing McCarthy is that he seems to have lost the ability or willingness to coach bad habits from recurring with the team's most important player. Which is ironic, in that McCarthy was originally hired to kick Stubbleface in the ass.
#87 by Richie // Nov 29, 2018 - 1:05pm
Just some fun (and surprising to me) facts:
McCarthy now ranks 27th all-time in coaching wins (5 behind Marvin Lewis).
He is tied with Joe Gibbs, Bud Grant and Buck Shaw for 24th all-time in winning percentage at .621.
He is 20th all-time at 49 games above .500.
He is tied with Bud Grant, George Seifert, Pete Carroll, Bill Walsh and John Harbaugh for 13th all-time with 10 playoff wins.
McCarthy is 55 years old. If the Packers were to move on from him, would he get (or want) to coach again someplace else? If he coaches 5 more seasons and averages going 8-8, that would be 40 more coaching wins for him. That would bump him up to 16th all-time in wins, unless Mike Tomlin, Pete Carroll or Sean Payton can beat him to 165 wins. I'm assuming Marvin Lewis doesn't have 35 wins left in him.
#88 by theslothook // Nov 29, 2018 - 1:22pm
On a related point, I keep going back and forth on what to think of Marvin Lewis. On the one hand, he's stuck in perennially tough division, with a perennially terrible owner, with most of his tenure spent in the harder conference and has done quite well. Jokes aside, the Bengals during the Lewis era have been pretty good
And yet.... they've never once been great and have fallen flat every single year in the postseason. Sure, each one can be explained and reasoned through and playoffs are random events, but it's still pretty remarkable. The usually hapless Jags by comparison have gotten past the first round three times in the last 15 years. I mean, even the stumbling Jets and snakebit Chargers have done it.
Somewhere in my head, I respect Lewis' accomplishments a lot and I do feel like there's a stronger possibility the next hire with the Bengals does a lot worse than a lot better. But, Lewis as a coach seems to have a prohibitive hard ceiling.
#92 by Richie // Nov 29, 2018 - 2:30pm
The 2015 Bengals were pretty good. Through week 12 they were 10-2 and number 1 in DVOA. https://www.footballoutsiders.com/dvoa-ratings/2015/week-12-dvoa-ratings
Then Andy Dalton got hurt and they went 2-2 down the stretch with AJ McCarron (who actually played pretty well). Then they blew that home playoff game to Pittsburgh when Jeremy Hill fumbled while running out the last 90 seconds of the clock.
#90 by Will Allen // Nov 29, 2018 - 2:03pm
Oh, I'm not on the Fire Omar McCarthy! Train. The data indicates that he likely a well above average coach. If, however, the CEO decides, by close observation, that McCarthy is going to be ineffective in coaching Rodgers, well, McCarthy is easier to replace than Rodgers, which is not to say it is easy to replace McCarthy.
#4 by Cythammer // Nov 27, 2018 - 7:37pm
Most people who don't know anything about DVOA would probably be surprised to see how high it ranks the Broncos this year. They would really be surprised to know DVOA actually thinks this squad is better than the Super Bowl champion 2015 team. Those Broncos were ranked 8th in the league and had a DVOA of 17.7%. The 2018 Broncos are ranked 6th and have a 21.8% DVOA.
#5 by merlinofchaos // Nov 27, 2018 - 8:26pm
To be honest, I know a bit about DVOA and I'm still surprised. I've always throught they were better than their record, but not by as much as DVOA says. They've got some pretty glaring weaknesses.
That said, the last 2 weeks they cleaned up a lot of the mistakes. Bolles is getting a lot fewer flags. If their 3rd CB weren't "receiver can't catch" I'd be a lot more comfortable with their chances of running the table the rest of the way in. That said I do like them to win 4 of their 5 next. Oddly the matchup I'm most concerned with is the Browns. After firing Hue they're actually doing pretty well and could give more trouble than conventional wisdom would say.
#7 by Aaron Schatz // Nov 27, 2018 - 9:04pm
I'm surprised too but that No. 1 past schedule means a lot of adjustment. And I think people forget just how horrendous the 2015 offense was for much of the regular season. That team had only 9.7 Pythagorean wins (https://www.footballoutsiders.com/stats/teameff2015).
Also, Denver's only slightly ahead of 2015. You're looking at weighted DVOA, not full-season DVOA. Full-season DVOA is 19.7%.
#21 by xydux // Nov 28, 2018 - 12:51am
As a Broncos fan, THE OFFENSE STILL SUCKED WITH PEYTON [in 2015].
At least when Osweiler was in, I wasn't worried about the next pass being a lame-duck INT (although, in retrospect, that totally should've been a worry). With Peyton, after week...4, basically, that was a constant concern, because at that point it became pretty obvious that he'd hit the wall Father Time put in his way.
In retrospect, the Broncos were much better when he was in (and when he realized he didn't have to do it all himself and basically let the defense take care of business in the playoffs, that was the best iteration of the team). But the offense was still so horrifying to watch that I was always more relaxed when the defense was on the field.
#45 by Aaron Brooks G… // Nov 28, 2018 - 12:11pm
The 2015 Broncos offense was about the same as the 2000 Ravens offense (better passing, not as good running), and not much worse than 2002 Tampa or 2006 Chicago.
The common thread being attached to historically-good defenses.
#53 by Will Allen // Nov 28, 2018 - 12:42pm
Tampa had an above average passing offense in 2002, and if you have to choose between being above average in either passing or rushing, pick the former. Tampa, in my view, was significantly better on offense than those other teams. That 2002 Bucs team is kind of an underrated champion, I think.
#54 by merlinofchaos // Nov 28, 2018 - 12:44pm
Right, and Denver's defense this year his not historically good.
Interestingly, I think it's one player away from that. Sadly that player was let go for Reasons. (Actually I understand the Reasons but I think they'd have 3 more wins this season if they'd kept Talib).
#8 by Cythammer // Nov 27, 2018 - 9:15pm
It has as much to do with the 2015 team being not as good as one might think. They had a very weak offense, of course, and they kinda skated by all season, collecting a lot of narrow wins that ultimately garnered them the home field advantage they needed to succeed in the playoffs. I think with just slightly different luck they could have easily been a wild card team that went out right off the bat in the playoffs. It seemed like they were in crisis mode for much of the season thanks to their offensive struggles, which always made their eventual success more remarkable to me.
#39 by BJR // Nov 28, 2018 - 11:11am
As always, injuries were a critical factor. Pittsburgh and New England were both pretty beaten up by the time they went to Denver in the playoffs that year. Pittsburgh had endured a brutal game in Cincinnati the week before, and were without Antonio Brown, among others. Denver barely scraped past them. The Patriots OL line was ravaged by injury late in the season, and they limped into the playoffs. It was only some Brady/Gronk brilliance that allowed them to put up any offense at all in the AFCCG.
And crucially in both those games, and in the Super Bowl, Denver were able to take an early lead and allow their pass rush to take over, whilst turtling on offense. I would not have fancied their chances of playing from behind.
Interestingly DVOA had the two best AFC teams that year as the Bengals and Chiefs. But the Bengals lost Dalton to injury late on in the year. And the Chiefs narrowly lost the division to Denver, and were pretty beaten up themselves. They had to go on the road, and inevitably lost in Foxboro.
#41 by Cythammer // Nov 28, 2018 - 11:35am
Denver taking an early lead and then turtling on offense the rest of the game is probably a fair description of the Panthers and Patriots games, but not the Steelers game. Denver scored first, but Pittsburgh took the lead by the end of the first quarter and kept it until there were just three minutes left in the game. It was a comeback win for the Broncos in that one.
#52 by theslothook // Nov 28, 2018 - 12:42pm
Your points are all valid. I just dislike doing revisionist history because unless you are the 85 Bears, every team has had favourable breaks go there way when it comes to winning the SB.
I love Zach Lowe's comment on this. "They don't give championships for season long point differentials."
I guess it's a matter of taste, but I like the randomness of the NFL more than the near certainty of the NBA. Even pre super teams, there were realistically only 4 to 5 teams with any shot to win the title. A NBA equivalent of the 2007 NYG would never have stood a chance.
#55 by Will Allen // Nov 28, 2018 - 1:18pm
Oh, definitely, the predictability of the NBA finalists really inhibits my interest in the sport. How they get anybody to spend a hundred bucks or more to attend a regular season game, between even the two third best teams in their respective conferences, is a bit puzzling to me.
#68 by Thomas_beardown // Nov 28, 2018 - 2:32pm
You can get 100 level Pacers tickets for like $60 some nights. It's great.
Also, it's still a phenomenal athletic display for most NBA teams. I guess in some ways it's similar to the reason people go see ballet.
#69 by theslothook // Nov 28, 2018 - 3:25pm
Me personally, I don't tailgate and I generally dislike attending NFL games. I find there too many obnoxious fans ( having been to several 49er games) and you just can't capture the game Dynamics despite being able to see the whole field.
I enjoy attending basketball games more
#72 by Will Allen // Nov 28, 2018 - 4:44pm
I wouldn't pay to attend either, at this point. Too many drunks at NFL games, and not enough urgency at regular season NBA games. I'll go for the socializing on those occasions when I've been invited to a luxury suite; it's kind of like attending a high end sports bar. Baseball, at a good stadium, on nice day, I'll attend anytime, since I just like being outdoors in the summer.
#82 by Aaron Brooks G… // Nov 29, 2018 - 9:46am
Jordan's playoffs weren't far removed from those of his later Bulls years. He had 17 games of the regular season to return to form; his regular season numbers were more depressed than his playoff numbers.
That aside, Olajuwon's Rockets posed matchup problems that other Bulls opponents had not. Once Grant left, Chicago had a hard time answering teams who had a dominant center and anything else. The 1994 Rockets weren't great at anything but center, but they were deep and solid everywhere. The 1995 Rockets weren't as deep, but Drexler was better than the two SGs he replaced.
The best center the Bulls beat was Shaq in 1996, when he and Hardaway showed up and everyone else wet themselves (like they had in the 1995 Finals). The next best center they beat was either Ewing or Mourning on structurally similar Knicks/Hornets/Heat teams who had no shooting around their star center. The Rockets were like a better version of the Pacers -- all the perimeter shooting of the Pacers, but with Olajuwon at center instead of Rik Smits. The Pacers gave the Bulls all kinds of problems. The Rockets were like a rich man's Pacers. Olajuwon's Rockets would have been the best team the Bulls faced on their runs since the '87 Celtics or the '88 Pistons.
#85 by Will Allen // Nov 29, 2018 - 10:10am
I agree that a great center posed the most significant challenge to the Jordan Bulls. I really disagree that 17 regular season games, after a multiyear layoff, provides adequate preparation for the NBA playoffs. Just as significantly, roster construction with knowlege that Jordan will be available for the season differs in important ways from what is likely when the team doesn't have Jordan.
The Rockets were great back to back champs, and Olajuwon beyond great. The task was made considerably easier by Jordan taking a couple years to discover that he couldn't hit a damned thing.
#83 by Eddo // Nov 29, 2018 - 9:48am
Orlando did beat Jordan's Bulls, though it was a version of Jordan that was barely 20 games removed from playing baseball.
And while I think a Bulls/Rockets matchup would have been very interesting, on what are you basing the Bulls struggling with the Knicks. The Bulls beat the Knicks in the playoffs each year of their first threepeat, never even going to a seventh game.
#60 by E // Nov 28, 2018 - 1:46pm
I immediately also thought of the '95 Rockets. But they had the prior season's MVP (Hakeem), won the title again in '96 to prove how good they really were and were the only team I could think of in the last 40 years that even arguably fit the description. I'm calling them the exception that proves the rule.
#73 by whateverdude // Nov 28, 2018 - 6:50pm
They won in '94, not '96. '94 was a much more traditional run as they were the 2nd seed and never had to play the 1st seed (Seattle -- who became the first top seed to lose in the first round). Totally different teams, too. '94 had Otis Thorpe and were paint-dominated, '95 had Drexler and played small ball. Olajuwon was unbelievable in both playoff runs, though.
#62 by Yu Narukami // Nov 28, 2018 - 1:54pm
You also want to add the "importance" of every single game on the NFL schedule. You have so few games and so few PO spots that (almost) every game matters. Plus the one-an-done system for the PO.
On comparison, with NBA, you have to go through 82 games, then the long PO series where you don't even know when the decisive game is!
This is one of the factor on why an european (used to soccer, where the highest tier (the Champions League) follow a similar structure) like me, falls for the NFL.
#27 by Vincent Verhei // Nov 28, 2018 - 4:21am
I just checked the video, and all three fumbles are the kind of plays that confuse our parser:
* In Week 3 against Atlanta, Brees dropped the snap, then picked it up and handed off to Alvin Kamara, who ran for a 14-yard gain.
* In Week 7 against Baltimore, Brees dropped the snap, then picked it up and completed a pass to Ben Watson for 3 yards.
* In Week 12 against Atlanta (again), Brees dropped the snap, then picked it up and pitched to Mark Ingram, who ran for a 19-yard gain.
Those are the kinds of things that need to be cleaned up in postseason editing. It's rare for one player to get two of those in a season, let alone three.
#40 by RoninX // Nov 28, 2018 - 11:21am
Interesting - what is the standard for recording a fumble on a player change of possession?
Based on this, I assume the fumbled is charged to the QB on a mishandled snap. Is the fumble also charged to the RB on a botched QB-RB exchange? Is there a difference between a classic hand off and an RPO? Or is there no standard and the scorekeeper makes a judgement call in each case?
Seems like recording these kind of things a .5 fumbles might be more accurate.
#49 by Travis // Nov 28, 2018 - 12:19pm
Generally speaking, charge all those fumbles to the QB. (An exception would be a shotgun snap so off-target that the QB never touches the ball; in that case charge the fumble to the center and credit the QB with a rush for 0 yards.)
#61 by E // Nov 28, 2018 - 1:49pm
I actually think it would be better to not consider those fumbles -- even though they may disrupt the timing of the play or even lead to a sack, they never lead to a lost fumble and therefore skew the fumble numbers (and any related stats like DVOA, DYAR, etc.). A low snap that the QB allows to touch the ground and immediately picks up is really no different than a very high snap that he has to jump for, except one is recorded as a fumble and one isn't. I'd be in favor of FO ignoring these "fumbles" not unlike hail mary INTs.
#64 by Richie // Nov 28, 2018 - 2:15pm
Looking at the NFL Statistician's Manual, they don't spell it out explicitly, but it appears that the QB should usually get charged with a fumble on fumbled snaps, or fumbled handoff/pitches, unless it is clear that it is the other player's fault.
#30 by Will Allen // Nov 28, 2018 - 8:02am
The Vikings defense is a great example of how even the best metrics don't capture team strength, due to injuries. When Everson Griffin can't play or isn't his normally productive self, they aren't the 6th best defense, when he is, they are likely considerably better than that. If Rhodes is seriously hurt, his loss, at this point, with their #1 draft pick corner on IR, is probably as debilitating as losing Griffen. On the other hand, losing Sendejo, or even Barr, due to position and depth, has been pretty manageable.
If you really want to have insight as to the strength of teams at any given time, you need to really know the quality of players in the 2nd slot of the depth charts, and the quality of their position coaches. Very, very, very, few people do.
#34 by Lost Ti-Cats Fan // Nov 28, 2018 - 10:06am
This is such a good point.
Making DVOA adjustments based on the average strength of the opposing unit over the course of a full season is weakened when the strength of the opposing unit in that particular game is either significantly better (healthier) or significantly worse (less healthy) than the season average.
I have no way how you'd parse the data to make these finer adjustments. Sometimes it's the loss of a single key player. The bigger impact, though, seems to be when you have multiple injuries at the same position (corners, WRs, OL, DL, etc.)
#36 by Will Allen // Nov 28, 2018 - 10:16am
I don't know if it can be done, but injury analysis is where huge strides in performance measurement may be available. I've pretty much given up on preseason win projections, because I came to realize that predicting the winners of competitions, when I don't know the identities of the competitors, is kind of silly.
#13 by ryan5581 // Nov 27, 2018 - 11:19pm
No "average" teams: there are no teams with Total DVOA between -5% and +5% (weighted or not). Every other 5% "band" is represented from -30% to +30%. Which also means the biggest gap in rankings is between Carolina and Jacksonville. In fact, the DVOA gap between CAR #15 and JAX #16 is bigger than the gaps both between PIT #7 and CAR #15, and between JAX #16 and TEN #25. Just found this interesting!
#20 by theslothook // Nov 28, 2018 - 12:28am
I have not watched a single Cardinals game this year(hard to believe, I've seen every team including the woeful bills twice), why is their offense this terrible? Bad qb play, bad coaching, no talent??? I'm guessing it's a combination of all three but there has to be a dominant factor. Is Rosen having a Goff like rookie year?
#31 by Yu Narukami // Nov 28, 2018 - 8:26am
You can do something like the HHI
https://www.investopedia.com/terms/h/hhi.asp
Sum of every (OL's snaps/5)^2
Maximum value = 1,000 (better)
Up until Week 12 ->
LARM 0,973
CLE 0,927
NYJ 0,913
SF 0,905
TB 0,902
DET 0,899
CAR 0,885
SEA 0,872
DAL 0,868
ATL 0,864
LACH 0,862
GB 0,853
CIN 0,846
BUF 0,836
NE 0,828
NO 0,818
PIT 0,817
CHI 0,813
OAK 0,799
PHI 0,793
BAL 0,780
HOU 0,780
TEN 0,777
KC 0,775
DEN 0,736
JAX 0,718
MIA 0,718
NYG 0,712
WAS 0,710
MIN 0,709
ARI 0,678
IND 0,645
#33 by Will Allen // Nov 28, 2018 - 9:51am
Intuitively, my sense is that this formula doesn't capture what we are looking for, as well as HHI captures the state of competition within an industry. Maybe because every snap has 5 entities with 20% of the "market". My intuition may be wrong. Thanks; you've given me something to flummox me.
#51 by Yu Narukami // Nov 28, 2018 - 12:28pm
It's no perfect but it's very simple (you ask one number! :D) and direct indeed. And it nailed ARI and LARM, though totally missed the great year of Indy's OL.
It also doesn't dinstiguish among bad player's continuity and good player's continuity (if you lost the 5 starters day 1 then you get a high value out of scrubs and SFA).
On the other side, the deeper you get, the more qualitative assessment enter in play.
I am taking suggestions!
#66 by Richie // Nov 28, 2018 - 2:23pm
Yeah, this method is probably overrating Miami's offensive line. In the offseason they added Josh Sitton at guard and Daniel Kilgore at center. They were supposed to solidify a line that has been pretty bad for a long time. Sitton played one game and Kilgore played 4. Ted Larsen replaced Sitton for 8 starts. So that may appear good that he's played so much, but is probably bad.
I guess if there was some way to factor in how deep a particular player was on the pre-season depth chart, that might help. But would have the wrong effect if it turns out a backup is actually a better player.
#48 by Aaron Brooks G… // Nov 28, 2018 - 12:18pm
You can interpret competition as disorder, though, in a business sense. Monopolies are very orderly and if you subscribe to the gravitational theory of wealth (wealth is attracted to concentrations of wealth), is the natural end-state of industry.
Market concentration glosses some important factors (talent, for one), but has value as an analysis tool.
#35 by Lost Ti-Cats Fan // Nov 28, 2018 - 10:11am
Buffalo's situation was entertaining because they purposefully pushed Taylor out to do … whatever that is they've been doing with their QB situation this year.
Arizona at least had some objective evidence to believe that Bradford could give them time to develop Rosen behind him. Turned out he couldn't, but as a plan coming into the season it wasn't "buy your popcorn and watch the Buffalo QB circus" crazy.
#46 by Mike B. In Va // Nov 28, 2018 - 12:13pm
They purposefully pushed Taylor out because he wasn't making the team any better than a lower salary number would by staying there. This is their "blow up the team" year, they're just being mediocre at it - though the team is definitely showing signs of life, much to my surprise.
Also, where Whaley was involved, you can bet someone's on their way out.
#67 by Richie // Nov 28, 2018 - 2:27pm
By DVOA, QBR, Passer Rating and ANY/A, Josh Rosen has been better than Sam Bradford. But, Rosen has still been pretty bad. Bradford's stats in just 3 games are generally much worse than any other season of his career.
So, either:
a) Bradford's ability and/or injuries have caused him to fall of a cliff in 2018 and Josh Rosen has been a bad QB.
or
b) Arizona is a bad situation to play QB, and maybe Bradford and Rosen aren't quite as bad as their statistics would indicate.
#25 by Yu Narukami // Nov 28, 2018 - 2:34am
These are the opponent-adjusted average game scripts until week 12
Rank-Tm-Value-Difference vs. DVOA
1 KC 6,65 (=)
2 NO 5,72 (+1)
3 CHI 5,38 (+2)
4 LARM 4,79 (-1)
5 LACH 3,21 (-1)
6 NWE 2,91 (+3)
7 PIT 2,30 (=)
8 SEA 1,96 (+3)
9 BAL 1,73 (-1)
10 DAL 1,15 (+11)
11 HOU 0,97 (+1)
12 IND 0,58 (+1)
13 DEN 0,34 (-7)
14 WAS 0,28 (+3)
15 ATL 0,12 (+7)
16 GB -0,18 (-6)
17 CLE -0,58 (+6)
18 CAR -0,64 (-3)
19 MIN -0,72 (-5)
20 DET -0,76 (+9)
21 JAX -1,81 (-5)
22 TEN -2,03 (+3)
23 PHI -2,29 (+1)
24 NYG -2,45 (+4)
25 MIA -2,61 (-7)
26 SF -2,85 (+2)
27 CIN -2,93 (-8)
28 BUF -2,99 (+2)
29 NYJ -3,09 (-2)
30 OAK -3,55 (+1)
31 TB -3,83 (-5)
32 ARI -5,38 (=)
2017 Philly and Pats were 1 and 3 this time of the year (Rams were 2nd with +7,07 so even better with KC this year).
2016 Pats and Falcons were 1 and 2.
No much difference on the top, only Chicago still a little bit of a darling of mine.
All AFCE except Pats is bottom of the barrel, and thanks to them Pats will probably complete the feat of facing sixteen different starting qb all year long (Brock-Derek Anderson-McCown-Tannehill?-Allen?-Darnold?).
Houston finally seems more than lucky. In 2016 they got in PO with #26 (last year TEN was #22 as a WC).
This system really loves Dallas and Detroit, while punishing GB and Cincy (these last two are matching with eye-test).
#94 by Chip // Nov 30, 2018 - 11:26pm
Yu - how should I interpret your rankings vs DVOA? Is one more predictive vs retrospective? If game script score > DVOA does that show a better coached team? Or is it luck? NE, CHI, are all Outperforming DVOA (both good coaches that script plays well) while GB is U/P DVOA (not surprising since McCarthy has a poor reputation given his HOF QB talent). Thanks
#95 by PTORaven // Dec 03, 2018 - 8:39pm
Anyone know how Baltimore jumped from 17th to 6th in special teams from week 11 to week 12? I can't figure out how to look up the special teams table from previous weeks so I have no way of knowing what changed so drastically, but I was surprised to see a unit jump so many spots this late in the season without having a crazy outlier performance like two kick return TDs or something.