Week 19 DVOA Ratings
by Aaron Schatz
Once again, it is time for postseason DVOA ratings. As always, the following rules apply:
- All 32 teams are ranked, whether they made the playoffs or not.
- Teams are ranked in order of weighted DVOA, not total season DVOA. Since weighted DVOA is meant to lower the strength of older games, these ratings do not include Weeks 1-5, and Weeks 6-11 are somewhat discounted. "Last week" here refers to last week's rank in weighted DVOA, not total season DVOA.
- Teams are treated as having a bye week in any week where they did not play. Since most teams haven't played in two weeks, that means some of the ratings for non-playoff teams can start getting a little unreliable. Really, this is only to be used for playoff teams, the other teams are just there for ranking comparison purposes.
- DVOA, as always takes a long-term view of an NFL team's performance. That means that the games of the last two weeks are just two games among many, so teams may be listed below other teams that they have beaten in the playoffs.
With that in mind, we had a lot of readers ask last week why we update the weighted DVOA ratings in the playoffs, but don't post total DVOA ratings that include both the regular season and the playoffs. Well, with more and more weeks of data, DVOA ratings change less and less each week. If you want total DVOA ratings for the season, you have them right here. Adding two more weeks of data really doesn't change much.
The playoff odds report is updated for the final four. Roster data is also updated, as are the scores for the FO Playoff Challenge game. (If you picked LeGarrette Blount and Keenan Allen, you're feeling pretty good right now.) You will find DVOA matchup pages for the AFC and NFC Championship games on the FO Premium page. Remember that the equation used to determine win probabilities for the playoff odds report is not as complex as the one used for FO Premium picks, so picks may differ.
* * * * *
To save people some time, we remind everyone to put their angry troll hatred into the official zlionsfan angry troll hatred Mad Libs form:
<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>
If you are new to our website, you can read the explanation of how DVOA is figured here. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.
Teams in yellow are still alive in the playoffs. Teams in gray lost this past weekend.
| TEAM | WEI. DVOA |
LAST WEEK |
W-L | WEI OFF DVOA |
OFF. RANK |
WEI DEF DVOA |
DEF. RANK |
WEI S.T. DVOA |
S.T. RANK |
|
| 1 | SEA | 45.6% | 1 | 14-3 | 8.0% | 10 | -29.1% | 1 | 8.4% | 5 |
| 2 | NE | 31.5% | 2 | 13-4 | 30.4% | 1 | 7.0% | 22 | 8.2% | 6 |
| 3 | DEN | 27.2% | 3 | 14-3 | 25.2% | 3 | -8.3% | 9 | -6.3% | 28 |
| 4 | PHI | 26.0% | 5 | 10-7 | 27.0% | 2 | -0.9% | 14 | -2.0% | 24 |
| 5 | SF | 24.2% | 6 | 14-4 | 10.1% | 8 | -5.2% | 12 | 8.9% | 4 |
| 6 | CAR | 22.8% | 4 | 12-5 | 4.0% | 12 | -15.2% | 5 | 3.6% | 11 |
| 7 | KC | 22.8% | 7 | 11-6 | 12.1% | 7 | 1.1% | 18 | 11.8% | 3 |
| 8 | STL | 21.0% | 9 | 7-9 | -5.7% | 21 | -13.6% | 6 | 13.0% | 2 |
| 9 | NO | 21.0% | 8 | 12-6 | 20.0% | 5 | -6.3% | 10 | -5.4% | 27 |
| 10 | SD | 16.8% | 10 | 10-8 | 24.0% | 4 | 8.2% | 23 | 1.0% | 17 |
| 11 | CIN | 15.1% | 11 | 11-6 | 1.5% | 15 | -11.6% | 7 | 2.0% | 14 |
| 12 | ARI | 12.7% | 12 | 10-6 | 2.7% | 13 | -18.9% | 2 | -9.0% | 29 |
| 13 | PIT | 7.9% | 13 | 8-8 | 8.9% | 9 | 3.6% | 19 | 2.6% | 12 |
| 14 | BAL | 0.6% | 14 | 8-8 | -23.9% | 30 | -11.0% | 8 | 13.5% | 1 |
| 15 | CHI | -0.7% | 15 | 8-8 | 12.6% | 6 | 17.0% | 31 | 3.7% | 10 |
| 16 | DET | -1.3% | 17 | 7-9 | -8.6% | 24 | -5.3% | 11 | 2.0% | 13 |
| TEAM | WEI. DVOA |
LAST WEEK |
W-L | WEI OFF DVOA |
OFF. RANK |
WEI DEF DVOA |
DEF. RANK |
WEI S.T. DVOA |
S.T. RANK |
|
| 17 | MIN | -4.7% | 21 | 5-10-1 | -0.5% | 17 | 11.1% | 26 | 6.9% | 9 |
| 18 | MIA | -5.3% | 19 | 8-8 | 0.1% | 16 | 0.3% | 16 | -5.1% | 26 |
| 19 | TB | -5.4% | 22 | 4-12 | -7.6% | 23 | -4.4% | 13 | -2.2% | 25 |
| 20 | TEN | -5.4% | 18 | 7-9 | 4.2% | 11 | 8.9% | 25 | -0.8% | 21 |
| 21 | NYJ | -6.2% | 23 | 8-8 | -6.6% | 22 | 0.9% | 17 | 1.3% | 15 |
| 22 | NYG | -6.9% | 24 | 7-9 | -24.9% | 32 | -18.9% | 3 | -0.9% | 22 |
| 23 | BUF | -7.4% | 20 | 6-10 | -13.8% | 26 | -16.6% | 4 | -10.2% | 30 |
| 24 | IND | -10.6% | 16 | 12-6 | -3.9% | 20 | 6.4% | 20 | -0.3% | 19 |
| 25 | DAL | -13.8% | 25 | 8-8 | 2.4% | 14 | 23.1% | 32 | 6.9% | 8 |
| 26 | ATL | -14.9% | 27 | 4-12 | -2.8% | 19 | 13.4% | 28 | 1.2% | 16 |
| 27 | GB | -15.8% | 26 | 8-8-1 | -0.8% | 18 | 14.3% | 30 | -0.7% | 20 |
| 28 | JAC | -16.9% | 28 | 4-12 | -17.5% | 28 | 6.7% | 21 | 7.3% | 7 |
| 29 | CLE | -25.0% | 29 | 4-12 | -12.8% | 25 | 12.3% | 27 | 0.2% | 18 |
| 30 | HOU | -33.7% | 31 | 2-14 | -23.9% | 31 | 8.3% | 24 | -1.5% | 23 |
| 31 | WAS | -35.7% | 30 | 3-13 | -18.0% | 29 | -0.7% | 15 | -18.4% | 32 |
| 32 | OAK | -42.7% | 32 | 4-12 | -16.6% | 27 | 13.6% | 29 | -12.5% | 31 |
There's been a lot written out there about how we ended up with the teams we all thought we were going to end up with. The big surprise of the weekend was the lack of surprises. The final four gives us four of the top five teams in weighted DVOA as of the moment, and four of the top six teams in total DVOA for the season. Seattle, New England, and Denver were the top three teams in the Football Outsiders preseason DVOA projections. San Francisco was seventh, but that was in part because of the Michael Crabtree injury, the 49ers played out as expected too: they got a little bit better once Crabtree came back.
This has nothing to do with the playoffs, but one thing that hit me when running these weighted DVOA ratings is that I had not realized how good the Baltimore and St. Louis special teams were over the second half of the season, especially St. Louis. Not that it has much to say for next year, given the year-to-year inconsistency of special teams, but that was a huge part of why Baltimore almost got into the playoffs and the Rams went 4-3 in their final seven games.
[ad placeholder 3]
The other item that will stand out is that the Patriots now rank No. 1 in weighted offensive DVOA, despite losing Rob Gronkowski in Week 14. The Patriots have had a consistently good but not dominating offense now for two months, though. The Pats had a season-high 74.8% offensive DVOA against Pittsburgh in Week 9, took Week 10 off, and then put up a fabulous 57.5% offensive DVOA against Carolina in Week 11. The haven't matched those two games, but they have been between 10% and 40% on offense in all seven games since. Denver, after adjusting for opponent, has actually been below that three times since midseason: Week 12 against the Patriots (-3.2%), Week 15 against the Chargers (5.2%), and then this week against the Chargers (6.2%), although obviously part of that are huge opponent adjustments based on how bad San Diego was on defense in the first three months of the season.
Here are the one-game DVOA ratings for this weekend's games.
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Seattle-New Orleans ends up a bit closer than you probably expected, and New Orleans even ends up with the higher rating once we account for opponent adjustments. You may be wondering how that happened when the Saints looked so bad for most of the game. Here is the New Orleans offensive DVOA by quarter: 29.5%, -31.4%, -2.4%, and 122.7%. You may refer to that last quarter as "garbage time," but it wouldn't have been very trashworthy if Marques Colston had hopped out of bounds like a normal person and Drew Brees had managed one more big pass to tie the game.
The other three games were fairly one-sided. I'm actually a little surprised there isn't more of a gap between San Francisco and Carolina. It's a little strange that the Colts don't get more of an opponent-adjusted boost for playing the Patriots, but it's a related to the down/play combinations. For example, the Pats' offense was actually very poor running the ball on third down for most of the season, so the Colts defensive DVOA actually gets extra-penalized, not improved, on the Pats' seven third-down runs.
* * * * *
 |
During the 2013 season, we'll be partnering with EA Sports to bring special Football Outsiders-branded items to Madden 25 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 standard stats. Others will be under-the-radar players who only stood out with advanced stats, including DYAR, Defeats, and our game charting coverage stats for cornerbacks. We'll announce the players each week in the DVOA commentary article, and the players will be available in Madden Ultimate Team packs the following weekend, beginning Friday night.
The Football Outsiders stars for the Divisional round are:
- Keenan Allen, WR, SD (Limited Edition): Led all WR with 77 DYAR, 142 yards and 2 TD. We've been trying to find the right week to do Allen all year, so it's nice to finally get to him the last week it was possible.
- Jamie Collins, OLB, NE: sack, interception, run TFL, 3 QB knockdowns.
- Dan Connolly, RG, NE: Helped lead Patriots running backs to 197 yards on 27 runs listed as middle/guard.
- Anthony Davis, RT, SF: Held Charles Johnson without a sack or a QB knockdown.
Other players we considered this week included Joe Staley, Julius Thomas, Bobby Wagner (5 Defeats), and Ryan Wendell.
Comments
91 comments, Last at 16 Sep 2014, 10:24am
#1 by Vincent Verhei // Jan 13, 2014 - 7:41pm
So based on VOA (which I think is the best way to determine who "should have" won), New Orleans was better than Seattle on offense and defense, and the Seahawks only won because they had the better kicker. Lovely.
#72 by Perfundle // Jan 14, 2014 - 4:00pm
It's a bit surprising that Seattle's Special Teams DVOA is so high then. Tate and Baldwin didn't do anything special returning punts and kickoffs, and Seattle's kicking coverage was about average as well. Was it all about Hauschka's field goals, then?
#77 by Vincent Verhei // Jan 14, 2014 - 5:43pm
Second quarter, with Seattle up 13-0, Ryan had a 24-yard punt from his own 28. NO was set up at their own 48. They got one first down, then the drive ended on a fourth-down incompletion from the SEA 29.
Third quarter, Seattle up 16-0, Ryan had a 26-yard punt from the SEA 48. NO took over at their own 26, and went on to score a touchdown to make it 16-8.
#78 by Will Allen // Jan 14, 2014 - 6:24pm
I think there is a substantial difference between getting a drive started on the opponents' 40, and getting a drive started on your own 48. A sixteen yard punt from your own 24 is a lot more catastrophic, in a low scoring game, than a punt of 24 yards from your own 28, plus a 26 yard punt from your own 48, especially since the 16 yard punt occurred so early. For the road team, getting the lead early in a contest like this is extremely important. The 16 yard punt, combined with the fumble, made the 1st quarter a disaster for the Saints. I don't think in the context of this game it makes sense to say the Saints got better production from their punter than the Seahawks.
This leads me to think about how a single turnover in a low scoring contest is a lot more expensive than one in a high scoring contest, which suggests yet another way in which rules emphasis penalizes a roster which is defensive oriented; the margin for error shrinks.
#44 by rich006 (not verified) // Jan 14, 2014 - 6:23am
Each punter had a bobble followed by a terrible punt leading to excellent field position for the opponent. The main differences in the game were:
1) the early turnover, which gave Seattle a short field leading to their only TD in the first 3 quarters.
2) the two missed field goals by the Saints.
#45 by Will Allen // Jan 14, 2014 - 8:19am
Payton not having a 4 down strategy on the drive leading to the first missed field goal was a real error, I thought, and I said so when the drive started. You can't be thinking of longish field goal attempts in conditions like that.
#47 by BJR // Jan 14, 2014 - 8:45am
It plain p*sses me off when coaches fail to adjust their 4th down strategies away from attempting field goals in adverse weather. Partly because it is just bad strategy, but also because it falls into the category of potential blame-avoidance. "If only our kicker (in this case signed off the street two weeks ago) wasn't so crappy we might have won!"
#10 by formido // Jan 13, 2014 - 8:28pm
Nope, this is just one that DVOA gets wrong. The only reason the offensive VOA for New Orleans is so high in the 4th quarter is because they moved to a high risk/high reward strategy and got every break: Two dropped INTs, another INT that was makeable but tough, and a 50 yard pass completion off of one of those INTs. They'll be rewarded by that in DVOA in the same way an NBA player would be rewarded for making 3 straight half court shots, but such a streak would be utterly irrelevant for predicting future success. Anyone who watched the game didn't for a second think New Orleans was the better team.
And that's all aside from the fact that Seattle purposely played a low risk second half strategy due to the weather conditions, whose degree continues to be ignored. When the Saints were playing their planned strategy, in the first half, future Hall of Famer Drew Brees was held to 38 yards passing by the defense plus weather conditions. Nothing can be taken from this game other than that Seattle is built to win in this type of weather.
#33 by dbostedo // Jan 13, 2014 - 10:27pm
"Nope, this is just one that DVOA gets wrong. The only reason the offensive VOA for New Orleans is so high in the 4th quarter is because they moved to a high risk/high reward strategy and got every break..."
He wasn't talking about prediction, he was referring to judging who should have one based on how they just performed. So this one VOA gets right. Catching every break, whether lucky on not, means you performed well in the game that already happened.
#41 by Ghost Shock // Jan 14, 2014 - 3:51am
The other team dropping INTs means you performed well? In what world? Luck happens.
The reason FO stats seemingly favor NO is because of the 4th quarter, just like Schatz already said, not because NO outplayed them overall. NO outplayed them in the 4th, which Seattle was fine with. VOA and DVOA aren't perfect evaluative stats for this very reason - they're just very, very good.
#79 by dbostedo // Jan 14, 2014 - 7:31pm
Numerically it does mean you performed well. For instance, if your opponent had, say, 3 dropped interceptions they are the same (judging performance by the numbers as VOA does) as incompletions. Your stat line would show zero interceptions - which is a good thing. That's what I mean by "you performed well".
You could watch that and say they are getting lucky, but your VOA would not show that - it would say you played well and didn't throw any picks.
And in reality, those "almost picks" are exactly the same as incompletions in hind-sight, which is what we're talking about.
#83 by Vincent Verhei // Jan 14, 2014 - 8:34pm
And in reality, those "almost picks" are exactly the same as incompletions in hind-sight, which is what we're talking about.
Actually, one of them turned into a 52-yard completion and New Orleans' biggest play of the day.
#84 by Perfundle // Jan 14, 2014 - 8:41pm
"And in reality, those "almost picks" are exactly the same as incompletions in hind-sight, which is what we're talking about."
In this case, they are not, because one of those "almost picks" was a 52-yard completion.
#7 by Purds // Jan 13, 2014 - 7:59pm
I heard that, so I rewound the DVR to make sure. Yup. Now, I can't say it was manning. It sounded like a different voice, even one from the sideline, but it would be funny if it were Peyton. Thomas had just caused a snap problem the play before because he lined up on the wrong side. I think they burned a TO. Wondering if Thomas has been investigating the new law in Colorado!
#5 by nat // Jan 13, 2014 - 7:53pm
For example, the Pats' offense was actually very poor running the ball on third down for most of the season, so the Colts defensive DVOA actually gets extra-penalized, not improved, on the Pats' seven third-down runs.
Whoa. Opponent adjustments are that situation specific? Doesn't that introduce a lot of noise?
#31 by theslothook // Jan 13, 2014 - 10:10pm
I'm assuming they've tested to see if such specificity helps or hurts the predictive value of dvoa and that's why they've kept it in. That said, I have no idea how robust any of their regressions are and with all of the inter complexities with different series, I'm fairly certain you've got all kinds of heteroskedasticity and autocorrelation problems.
All that to say, without access to their formula, we'll have to accept it with a fair amount of skepticism.
#6 by Perfundle // Jan 13, 2014 - 7:58pm
"You may be wondering how that happened when the Saints looked so bad for most of the game."
More like the Saints' score looked so bad for most of the game. New Orleans was moving fairly easily between the 30's for much of the game, although DVOA doesn't know about the dropped interceptions. Barnwell said that New Orleans decided to sacrifice a tight end on Sherman's coverage, and it seemed to work in the second half; since Sherman has shown that he can cover the left side of the field, the coaching staff needs to prevent opponents from exploiting this strategy; I believe Houston used it to great effect as well.
#13 by formido // Jan 13, 2014 - 8:33pm
It's pretty easy to have passing success when you're willing to throw interceptions. The fact that the interceptions weren't caught doesn't mean we should appreciate the strategy. That's celebrating results over process, and we have very good evidence Seattle will usually catch those interceptions.
#15 by Perfundle // Jan 13, 2014 - 8:44pm
If you aren't willing to throw anything that might end up as an interception then you're Alex Smith. Wilson has thrown plenty of passes that could've been intercepted if the defender put in a heroic effort. He has become overly conservative these last few games, even when the pocket has held up decently.
#9 by young curmudgeon // Jan 13, 2014 - 8:26pm
"Anthony Davis, RT, SF: Held Charles Johnson without a sack or a QB knockdown."
Well, of course, if Davis held Johnson, of course Johnson didn't have a sack or knockdown!
Sorry, but when I read it at a glance, that was my first impression--sort of whatever the reading equivalent of hearing a Mondegreen.
#18 by Will Allen // Jan 13, 2014 - 9:37pm
I really don't like the Broncos chances against the Seahawks at all. The Broncos, absent Miller and Harris, are likely to be generous enough on defense to make the Seahawks offense look dynamic, and the Seahawks have enough depth in the secondary to avoid, or at least substantially lessen, the weak link effect that the Manning alien tends to ruthlessly exploit.
I think the Patriots, on a cold windy day in Jersey, would make the Seahawks offense look like what it is, and Patriots offensive line could ensure the ball is run effectively. The Broncos passing game is so dynamic that it is hard to know how much of their running success is based on that.
(edit) On the other hand, the Niners get such a consistent push up the middle, especially with the hominid/rhino hybrid named Justin Smith, that they might just give Mr. Bundchen a very, very, difficult time, given his lack of receiving help these days.
#89 by beargoggles // Jan 15, 2014 - 6:19pm
I've been thinking that for weeks. Seattle's defense seems like it was designed to beat this particular Bronco offense. Especially if the weather isn't good. (not cold necessarily, but windy/rainy).
Any of the other possible match-ups seem less clear cut to me. Denver would probably due OK with SF's relative pedestrian corners. Either SF match-up could be a shoot-out. NE-SEA would be very interesting. I don't think Brady is as likely to force doomed throws and they could probably run the ball. They could be well built for the 'Hawks especially if Seattle continues it's trend of uninspired offense.
#21 by Keith_1 // Jan 13, 2014 - 9:42pm
With Moreno and Ball, the Broncos have a formidable, if consistent and capable, running game. If we call Manning/Brady equal in cold weather (Manning has had, I believe, only a single "terrible" game in cold weather, with a sprinkling of fair, middling, and good games otherwise), and say the receivers for the Broncos are better, I think the Broncos offense could match up against any opposing defensive weakness as well or better than the Patriots. (That is not to say that Fox could compete with The Hoodie in correctly using the personnel, which I would grant is an equally important aspect.)
I think the biggest concern for both the Patriots and the Broncos is the defensive side of the ball, with all the injuries they each have. Even with their DVOA rankings, SEA/SF offenses do not exactly inspire fear; however, for all their faults, they are perfectly capable of winning a 20-17 game against either AFC competitor.
#24 by theslothook // Jan 13, 2014 - 9:51pm
I disagree about SF's offense. They have all of the elements to be very good. Their run game is ok, but their pass game has capable weapons and a good strong armed qb. They are the best overall team imo in the nfl, but unfortunately must play seattle at seattle. If this were a neutral site, I would take sf over any team in the nfl. But alas, seattle at home might be a bit too formidable.
#26 by Will Allen // Jan 13, 2014 - 9:53pm
I don't think there is any comparison between the Patriots' ability to hide their defensive weaknesses, and the Broncos, now that Miller and Harris are out.
If Crabtree can go nearly full-tilt, the Niners offense becomes much more formidable.
#27 by theslothook // Jan 13, 2014 - 9:58pm
THe broncos are strange. I still feel like their defense is capable of being ok. Their defensive suckitude is really coming from completely blown coverage assignments and that seems to be the result of in game injuries. Against ne, the suckitude in the pass game started because DRC got hurt. In this game, things came unglued with harris out. Of course, harris isn't coming back, but I wonder if with a full week of practice, they won't be caught in quite the bind as the keenan allen explosion might have you believe.
Also equally important, Knighton didn't play in the first meeting and neither did Julius thomas. And ofc, gronk is out.
#32 by Will Allen // Jan 13, 2014 - 10:12pm
Secondaries, when injuries deplete them past the tipping point, tend to be pretty ugly things to observe. Is Jammer absolutely finished, to the point where you'd do just as well with a street free agent? Guess we'll find out, as Brady strenuously explores the question after kickoff.
#39 by Hummingbird Cyborg // Jan 14, 2014 - 3:12am
As a Broncos fan, I've been hoping all season to see them use Tony Carter again. He was pretty good last season lined up with Harris and Bailey most of the year. He just has to be better than Kayvon Webster who has looked like an outclassed rookie all too often and Quentin Jammer.
#30 by theslothook // Jan 13, 2014 - 10:07pm
Just as a sidenote: one of the underrated things about an elite qb is how they are able to adjust when an injury takes away one of their main weapons. I remember re-watching the NE game and it was clear how much gronk just affected all levels of the broncos defense. They consistently tried to cloud him and he not only beat those coverages, but he allowed so many things to amendola and edelman because he sucked in the safeties. Of course, now that he's gone, its not as if NE will try and duplicate the same exact offense sans gronk. NE has always shown great adaptability in ways I don't think even patriot fans appreciate enough.
As for the other guy, his response to losing a player is to just assume more of the burden. In 2010 when his o line was shitty and his receivers were hurt, he ran the same offense but compensated with a quicker release and more threaded passes. Its admirable that he does this as I think almost no one in the nfl could.
That said, I do wonder sometimes if this is the right approach to take. One great example of this to me is with reggie wayne. Reggie wayne was starting to lose his ability to completely separate on the outside against top corners. Once bruce arians got there, they moved wayne in the slot, used him in motion and different designs to take advantage of his knowledge.
Its two different ways to attack a problem, but I think peyton ought to consider allowing scheme to help him compensate rather than take on additional responsibilities.
#34 by RickD // Jan 13, 2014 - 10:44pm
Manning's completion percentage, over his career, in games when the temperature is below freezing, is 41%. Brady's is above 62%.
See
http://espn.go.com/blog/statsinfo/post/_/id/79050/tom-brady-thrives-in-new-england-weather
and
http://www.cbssports.com/nfl/eye-on-football/24315549/the-week-in-overreactions-peyton-manning-coldweather-qb
And keep in mind that any of Manning's career stats have to be viewed with an understanding that most of them were accumulated before his surgery.
#36 by Keith_1 // Jan 14, 2014 - 12:17am
Without looking at individual games, it is really hard to get a grasp of what actually has transpired. With that being said, I queried PFR for relevant games from 1998 to 2013 of all outdoor games played at temperatures below freezing, where Peyton Manning's team played, and then tossed single-digit attempt nights for pointless games (such as week 17s). The first three are as a Bronco; the other seven are as a Colt.
Dec 8, 2013, 18 degrees, Titans: 39/59, 397 yds, 4 TD, 0 int
Nov 24, 2013, 22 degrees, @Patriots: 19/36, 150 yds, 2 TD, 1 int
Jan 12, 2013, 13 degrees, Ravens: 28/43, 290 yds, 3 TD, 2 int
-------------------------------------------------------------
Jan 3, 2010, 11 degrees, @Bills: 14/18, 95 yds, 0 TD, 1 int
Jan 16, 2005, 27 degrees, @Patriots: 27/42, 238 yds, 0 TD, 1 int
Jan 18, 2004, 31 degrees, @Patriots: 23/47, 237 yds, 1 TD, 4 int
Nov 24, 2002, 27 degrees, @Broncos: 27/44, 229 yds, 0 TD, 1 int
Dec 3, 2000, 29 degrees, @Jets: 27/51, 339 yds, 2 TD, 2 int
Nov 19, 2000, 26 degrees, @Packers: 25/44, 294 yds, 3 TD, 1 int
Dec 26, 1999, 31 degrees, @Browns: 27/43, 276 yds, 0 TD, 0 int
I see, at first glance, two great games (Titans 2013, @Packers 2000), two terrible games (@Bills 2010, @Patriots 2004), three good-to-middling games (@Patriots 2013, @Ravens 2013, @Jets 2000), and four fair games. You can re-define them however you would like. Manning's total line:
256/427 (59.95%), 2545 yds (5.96 YPA), 15 TD, 13 int (1.15:1), over 10 games, and he would be [roughly] a league-average quarterback.
Considering Manning has played roughly 1/3 of the total [meaningful] outdoor games below freezing that Tom Brady has (10 vs 29), comparing them is difficult.
First impressions: I skimmed over Brady's games (I will post compilation data tomorrow, if desired, but it is getting late), and while his total quality is higher than Manning (in these conditions), so is his opportunity. Since they are so similar in nearly every other weather condition, if we assumed Manning's quality was near enough to Brady's to reach quantity, Manning would have a near-season's worth of good performances...in a row. Probably not going to happen.
So yes, as it is, a randomly sampled Tom Brady game in these conditions is likely to better than a randomly sampled Peyton Manning game in these conditions. That says quite a bit more about Tom Brady's consistent success in sub-freezing temperatures than Peyton Manning's shortcomings. In the end, the outlook does not appear nearly as bleak for Manning as the quoted 41% completion percentage.
#43 by theslothook // Jan 14, 2014 - 5:39am
A few other added points. In the 2010 bills game, the int was the result of a receiver slipping. Not too mention, it was one drive and then rest for peyton.
In addition, do we really know for sure cold games hurt qbs? Snow hurts the dbs footing, as evidenced in the 2010 bears vs NE game, where the bears dbs admitted they had a hard time being able to follow receivers due to the snow.
In fact, this whole weather thing feels like after the fact judgements applied one sidedly to suit an argument. Drew brees can't win the cold until he did(v Philly). I remember the dome argument being used to pick away at manning. Well, he just had a huge season while playing outside so hopefully that narrative is dead.
I don't think manning or brady are all that different from any qb. They all prefer warm controlled weather, are aided or hurt by the snow(depending on db footing), and probably all hurt by wind. Until i see an exhaustive statistical study to suggest otherwise, this cold argument is right up there with the clutch choker argument.
#48 by Ryan D. // Jan 14, 2014 - 8:47am
The other issue that is not being observed here is home/road splits. How many of Brady's cold-weather games were home games? For the first 11 years of Manning's career, playing a cold-weather home game was impossible.
Every QB plays worse on the road, so you should expect that some-to-most of his cold-weather games would rate as below his career average performance. Since 30% of these games were contested during the playoffs (against the eventual SB champion, no less), you can expect that he is playing against very good teams. How many of Brady's career cold-weather games are road playoff games? I'm guessing the percentage is lower than 20%.
Having a home game against the Bills/Dolphins in December every year probably helps keep the cold-weather stats pretty high for Brady.
#52 by Will Allen // Jan 14, 2014 - 10:13am
True, but if you had a sample size that was, say, the size typically used to evaluate a major league hitter's career, the pool would very likely be a lot less biased. Buffalo can't suck until the year 4500, can it?
#80 by RickD // Jan 14, 2014 - 7:31pm
On the other hand, "the sample size is too small" is a common complaint made by people who don't work with numbers professionally, or who don't know exactly what they are looking for. Or who may be counting the wrong thing.
To wit: if a QB had a completion rating of 41% in 10 games below a certain temperature, and of 65% or above over the rest of his career, that would most likely be a statistically significant since, in this case, n does not equal 10, but n is the total number of passes thrown in those games, which is well over 400.
Of course, tracking down that number (41%), it is clearly wrong, since Manning has never thrown that poorly in a single game, much less as an average over several games. I suspect that somewhere along the way somebody transposed the completion percentage and the incompletion percentage, which is about 41%.
A drop to 59% is far less dramatic than a drop to 41%, and I don't think I would have made the initial point the way I did had I had the correct number to start with. But given that his ordinary completion percentage is 5 points higher (and, more importantly, his ordinary completion percentage on the road is also 5 points higher), I think it's clear that Peyton suffers a bit in colder weather.
One could argue that there's a bias involved related to playing road playoff games, but only two of the games in the set in question are road playoff games.
#81 by dmstorm22 // Jan 14, 2014 - 7:45pm
I think the only variable that could be affecting the results (and to show you how long it's been since I've taken an intro-stats class, I can't even remember what that variable is called) is that other than two they're all road games.
My guess is the results aren't significant (could be wrong), but I wouldn't get upset at the notion that he performs slightly worse than he normally does in sub-freezing temps. I would argue if you remove the two playoff games against NE, the stats look a lot better.
It really doesn't matter going forward because it's not like he's never played bad games in the cold. And even to get to the point where he could be playing a sub-zero game, he has to win a game in non-sub-zero temps this Sunday.
#86 by bravehoptoad // Jan 15, 2014 - 10:55am
No, I missed that one quite deliberately. Heart-broken over the loss of the 49ers in the NFCCG. Affected me much more than the losses to Dallas in the early '90s, for some odd reason -- and I wasn't alone in that. Half of San Francisco skipped the Super Bowl that year.
#57 by PaddyPat // Jan 14, 2014 - 11:01am
I think the real point for this discussion boils down to something much more concrete and specific than Manning or Brady being a good "Cold Weather" quarterback. The point is that football is played in the elements. Adapting to the elements accurately is part of the game. Belichick's teams (whether by his influence, or Scharnecchia or the OC) seemingly play well offensively in virtually all types of elements. (They have struggled sometimes in the Miami heat). This is about coaching and game-planning. When Brady plays in the snow, his receivers routes seem to adjust to the elements, focusing on types of throws that are more suited to that weather. When it pours, the Patriots run. My brother asked me the other day during the game, if Luck can get away with great passes in this weather, why don't the Patriots pass more? And then the second interception happened. It's hard to control the ball consistently in that weather, that's why. That seems to me to be the heart of the discussion. A better thing to look at would be coach offensive performance in the elements, and there, I guarantee that Belichick knows a lot more about what he's doing than does John Fox.
#58 by Will Allen // Jan 14, 2014 - 11:11am
The fact that dome teams as a whole perform much less well offensively, in cold weather, suggests that coaching and game planning are not the major issues, unless one wishes to make the case that owners of dome teams tend to hire coaches who aren't as good. I doubt it.
I think offensive players who play a lot less frequently in poor weather conditions just don't handle it as well, and that offensive rosters in domes are built for perfect conditions.
#60 by Keith_1 // Jan 14, 2014 - 11:36am
Preface: I looked at the stats this morning.
To echo what Will said -- based only on 10 games for Manning and 28 for Brady -- I think this might go beyond coaching.
Brady plays relatively career-average football in the cold: 62.26% completion (63.4% for career), 4.98% TD percentage (5.5% career), 1.87% INT percentage (2.0% career), and 6.94 YPA (7.5 YPA career).
Manning plays noticeably worse in the cold: 59.95% completion (65.5% career), 3.5% TD percentage (5.8% career), 3.04% INT percentage (2.6% career), and 5.96 YPA (7.7 YPA career).
Now, whether this is due to a small sample size, playing away from home, playing in the cold, playoff environments, etc., I have no idea. But Manning does go from "in the discussion for GOAT" to "good quarterback" when you add multiples of those factors together: in the cold and away from home, in the cold and away from home and in the playoffs. Two of Manning's best three games "in the cold" come at home (Denver, post-surgery).
I wish we could watch Manning play in the cold and snow for another 18 games, but I doubt it happens.
#61 by dmstorm22 // Jan 14, 2014 - 11:46am
You've basically said the major variables effecting this difference. All of these games for Manning for most of his career had to be road games, likely late in the year, possibly playoff games (I think at least three were playoff games). Stats will go down in those situations.
#62 by Keith_1 // Jan 14, 2014 - 11:54am
Yes. I understand that. That is why I will not say "Manning is worse in the cold" without understanding [very specifically] how he plays in other away games, other playoff games, and other away/playoff games. And with only 10 games at which to look, we will probably never truly know, as that is hardly half a season.
Do not get the wrong impression -- my opinion is that Manning probably plays worse in cold weather, but only to the extent (maybe slightly more) that Tom Brady plays worse in the cold, and not to the extent that is often exaggerated.
#63 by dmstorm22 // Jan 14, 2014 - 12:00pm
I agree with your statement. He definitely plays worse, but I think that's more about him playing worse in road games and playoff games, something that is true of almost any QB (especially the road part).
At least we're all having a debate with his game stats and not just record. It was so maddening when people gave the W-L record and included two losses where he played one series in Week 17.
#64 by Keith_1 // Jan 14, 2014 - 12:08pm
I actually want to go back and compare the basic stats with the DVOA for each matchup for Manning and Brady, as I have a feeling (as mentioned earlier in the thread) that a good portion (say, 50%?) of Brady's exceptional cold games are against relatively bad teams. The AFC East was not always competitive for his career, and he has multiple games in the cold against them all. (I think I listed four games @Bills for Brady, and I believe either all four or three-of-four were in the exceptional category.)
#69 by Bobman // Jan 14, 2014 - 2:51pm
As much as I love Manning and understand the ridiculousness of blaming a week 17 loss on him when he played five snaps, I almost want to do it because he counts that as his consecutive starts to a career record. It's got to cut both ways, Peyton--you count the game for your durability vanity then you count it for your record as a starter.
IMO, neither should really count--I guess I can see why you'd count it for durability because he was ABLE to play, but that always bugged me when an argument was made to include those games for one stat but not for others. I'd treat those games like a walk in baseball--just pretend it never happened.
#68 by Bobman // Jan 14, 2014 - 2:47pm
@ Ryan D re home/road splits. If HFA is worth about 17% in team DVOA (as I read here earlier this year), I wonder how much of that is QB specific.
i.e. in terms of team wins, Brady (and Rodgers, Cutler, Roethlisberger, etc) get a team boost for HFA so in the whole universe of cold weather games, those guys would be expected to win more than dome/sunbelt QBs playing in cold games because half are likely at home. Say those guys play 30 freezing games over their careers and half are at home, while Manning (Rivers, Schaub, Ryan) play just ten in their careers, and ALL are on the road. All other things being equal, I'd certainly expect the road guys to lose a bit more (team stat). The sample size is probably not big enough to be meaningful to drill down to individual effects on QBs, but I bet it's there because those cold games are on the road.
(That being said, Manning used to explain the Colts superior road performance--a few years they were better on the road than at home--because the crowd noise forces him and the O to focus much more on the audibles. Was he just trying to psych out the opponent crowds in an interview? Maybe.)
#22 by theslothook // Jan 13, 2014 - 9:49pm
It's hard to read just how much circumstances affected the broncos gameplan. This game reminded me of their approach against KC the first time. They were nervous about their tackles, but more importantly, about manning's fragile ankle. The goal, therefore, became all about going for quick short passes whenever possible, largely at the expense of their typical style of offense(Which actually has a lot of down the field plays).
VS SD,I think they were nervous about the wind-as was SD I think. And since the run was persistently working, they stayed with it. I doubt they will adopt a similar strategy via NE and will this time come out firing with their passing game.
#23 by PaddyPat // Jan 13, 2014 - 9:51pm
Two questions.
First of all, there are no opponent adjustments on Special Teams? I understand that some special teams situations are rather non-interactive, like kicking, but many of them are not.
Second, I was wondering during the Seattle-New Orleans game, before the first missed kick. If a team gets 8 yards on 3rd and 13 and thereby makes it to reasonable field-goal range, does DVOA view that as a semi-successful play? How does that work?
#71 by Vincent Verhei // Jan 14, 2014 - 3:55pm
Second, I was wondering during the Seattle-New Orleans game, before the first missed kick. If a team gets 8 yards on 3rd and 13 and thereby makes it to reasonable field-goal range, does DVOA view that as a semi-successful play? How does that work?
It's not as black-and-white as "successful plays good, unsuccessful plays bad." Each team gets a certain amount of success points per play, then that number gets compared to what the average offense did in similar situations. So teams that gain big yards on third-and-long, even when they "fail," still get a small amount of credit. (Even on their own side of the field, 10 yards of field position is not insignificant.) So that play was a positive for New Orleans, but not as positive as a first down would have been.
#37 by anotherpatsfan // Jan 14, 2014 - 1:01am
I assume the DVOA for the divisional games is calculated using the whole season DVOA numbers and not the weighted DVOA numbers in the table above. SD's whole season defensive DVOA through week 17 was about plus 17, while Indy's was about plus 1, so Denver's VOA gets a much bigger DVOA adjustment from playing SD than does NE's VOA from playing Indy.
#73 by Vincent Verhei // Jan 14, 2014 - 4:08pm
Just for fun, here are the VOA ratings for NO-SEA in the first three quarters (ignoring special teams for now):
NO: -28% offense, -16% defense, -12% overall
SEA: -10% offense, -33% defense, 24% overall
So yes, Seattle crushed them for three quarters, then got wiped in the fourth. I guess that makes me feel better, but man, that fourth quarter was no fun to watch.
#88 by Supadome // Jan 15, 2014 - 1:50pm
Watching the 4th quarter was both fun and not fun for us Saints fans. You probably have this data, but I just pulled Brees' passing stats by quarter. Note, the PBP lists the Colston catch and illegal pass as an incomplete pass to Colston, which is one way to write it I guess.
1: 3/3, 19 yards
2: 2/8, 15 yards
3: 5/8, 62 yards
4: 12/21, 175 yards
Wilson was 7/9 for 68 in the first half, and 2/9 for 35 in the second.