Writers of Pro Football Prospectus 2008

Most Recent FO Features

RiversPhi04.jpg

» 2014 KUBIAK vs. ADP: The Underrated

Going too low in your fantasy draft: veteran quarterbacks, running backs who do more with their hands than their feet, and Houston's (only) two good receivers.

16 Jan 2012

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.
  • 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.

This really means something, of course, when it comes to the Giants, whose season has followed a crazy up-down-up trend similar to four years ago. If we look only at DVOA over the last four weeks, the Giants would be number one. But the Giants were lousy between Week 10 and Week 15. Those games are all accounted for in the current weighted DVOA ratings. History has shown us that most of the time, it's better to look at two or three months of games. And by history, I mean pretty much most of the last 20 years except for the 2007 Giants and 2008 Cardinals. So, honestly, who the hell knows. The Giants confuse me.

The playoff odds report is also updated through through this week's games.

* * * * *

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.


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 NO 32.6% 1 14-4 48.9% 1 16.6% 30 0.4% 17
2 NE 32.4% 4 14-3 40.1% 2 15.5% 26 7.7% 3
3 PIT 26.0% 3 12-5 15.7% 9 -7.0% 5 3.3% 9
4 PHI 24.4% 5 8-8 16.8% 8 -6.9% 6 0.6% 16
5 GB 22.3% 2 15-2 34.7% 3 16.0% 28 3.6% 8
6 HOU 20.2% 6 11-7 8.5% 12 -10.5% 2 1.2% 13
7 SF 19.7% 7 14-3 3.2% 16 -8.7% 4 7.8% 2
8 DET 16.1% 8 10-7 20.1% 6 1.2% 14 -2.8% 27
9 BAL 14.5% 10 13-4 8.3% 13 -9.6% 3 -3.4% 29
10 NYG 13.7% 13 11-7 17.5% 7 4.6% 17 0.9% 15
11 ATL 12.4% 9 10-7 8.9% 11 -5.7% 10 -2.2% 26
12 SEA 11.0% 12 7-9 3.0% 18 -5.9% 9 2.1% 11
13 SD 10.4% 15 8-8 26.0% 5 16.5% 29 0.9% 14
14 TEN 9.9% 16 9-7 1.4% 19 3.3% 15 11.9% 1
15 MIA 9.1% 11 6-10 -2.9% 20 -6.4% 8 5.6% 5
16 NYJ 6.9% 14 8-8 -3.7% 21 -6.9% 7 3.7% 7
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 CAR 2.1% 17 6-10 26.8% 4 21.7% 31 -3.1% 28
18 DAL -2.7% 18 8-8 10.8% 10 12.9% 23 -0.6% 19
19 CIN -5.0% 20 9-8 3.1% 17 10.8% 22 2.7% 10
20 WAS -6.0% 22 5-11 3.2% 15 7.4% 19 -1.8% 24
21 CHI -6.6% 19 8-8 -25.7% 32 -13.2% 1 6.0% 4
22 CLE -9.5% 24 4-12 -4.7% 22 9.3% 20 4.6% 6
23 OAK -10.0% 25 8-8 6.4% 14 15.1% 24 -1.2% 20
24 KC -10.8% 23 7-9 -14.6% 26 -5.5% 11 -1.7% 23
25 ARI -12.8% 26 8-8 -15.6% 28 -1.0% 13 1.9% 12
26 DEN -16.0% 21 9-9 -5.9% 23 10.2% 21 0.1% 18
27 JAC -17.7% 27 5-11 -17.9% 29 -1.7% 12 -1.5% 22
28 BUF -22.9% 28 6-10 -6.3% 24 15.4% 25 -1.3% 21
29 IND -25.0% 29 2-14 -14.6% 27 6.9% 18 -3.4% 30
30 MIN -32.0% 30 3-13 -11.2% 25 15.6% 27 -5.2% 31
31 STL -32.5% 31 2-14 -21.8% 31 4.6% 16 -6.2% 32
32 TB -45.2% 32 4-12 -19.9% 30 23.1% 32 -2.1% 25

Here are the one-game DVOA ratings for the second round of the playoffs. Yes, Houston did in fact come out with a slightly higher DVOA than Baltimore. Fumbles are such a bummer.


DVOA (with opponent adjustments)
TEAM TOT OFF DEF ST
SF 55% 17% -16% 22%
NO 21% 44% 14% -8%
NE 106% 68% -21% 17%
DEN -102% -54% 42% -6%
BAL 23% -24% -36% 11%
HOU 29% -12% -43% -2%
NYG 63% 31% -35% -3%
GB -11% 1% 19% 7%
VOAf (no opponent adjustments)
TEAM TOT OFF DEF ST
SF 23% 20% 20% 22%
NO 9% 33% 16% -8%
NE 124% 73% -34% 17%
DEN -112% -41% 66% -6%
BAL 8% -33% -30% 11%
HOU 10% -27% -39% -2%
NYG 38% 40% -1% -3%
GB -15% 6% 27% 7%

Based on DVOA, this actually isn't the worst game the Packers played all year. It's close. They had -17.5% DVOA for their Week 11 narrow win over Tampa Bay, and -15.5% for their Week 15 loss to Kansas City.

The Patriots' 45-10 domination of Denver wasn't just a run-of-the-mill big win where the Patriots offense played its usual good game and the Broncos offense played its usual bad game. It was actually one of the biggest wins of the season. The Patriots' single-game DVOA was the third highest of the entire season, behind only two games: Houston's Week 7 41-7 win over Tennessee and Baltimore's Week 1 35-7 win over Pittsburgh. Denver's single-game DVOA was the second lowest of the entire season, ahead of only Washington's 23-0 loss to Buffalo in Week 8.

Posted by: Aaron Schatz on 16 Jan 2012

213 comments, Last at 28 Jan 2012, 11:30pm by Dales

Comments

1
by Joseph :: Mon, 01/16/2012 - 1:01pm

Just a question: Is SF's ST DVOA so high because they recovered those fumbles? It's not like they broke long returns or made 60-yd FG's or something.

2
by fiddler (not verified) :: Mon, 01/16/2012 - 1:20pm

Probably has more to do with the 43.4 net punting and six kickoffs resulting in starting field position of NO14, NO20 (touchback), NO20 (touchback), NO21, NO12, and NO14.

17
by Vince Verhei :: Mon, 01/16/2012 - 2:29pm

I meant to mention this in Audibles, but yes, SF's kickoff coverage was outrageously great in that game.

Forcing fumbles will boost a team's DVOA, but recovering won't -- that's mostly luck.

173
by Jon Sullivan (not verified) :: Tue, 01/17/2012 - 3:20pm

To echo what Vince said, if you watch the tape of SF's kick coverage it was other-worldly. It's possible that they missed a few tackles, but I honestly don't remember a sequence where the first defender to get a hand on the return man didn't bring him down.

178
by bravehoptoad :: Tue, 01/17/2012 - 3:59pm

It sure looks like a skill, doesn't it? And yet, according to some, "...kickoff and punt coverage are relatively normally distributed - more specifically, they are not predictive or consistent on a season-to-season basis...."

http://www.advancednflstats.com/2011/12/how-important-is-opponent-starti...

180
by Aaron Brooks Go... :: Tue, 01/17/2012 - 6:15pm

"Not consistent on a season-to-season basis" in no way precludes a given squad from being high performance within a season.

182
by tuluse :: Tue, 01/17/2012 - 6:21pm

I think the Bears and Colts have shown there can be season-to-season consistency each at one end of the spectrum.

188
by bravehoptoad :: Tue, 01/17/2012 - 9:58pm

True, but I get the idea they use this to mean it's entirely non-predictive, and therefore not worth looking at. It's one of the things I like about DVOA, that it gives full credit to special teams, when other intelligent rating systems do not.

189
by Eddo :: Tue, 01/17/2012 - 10:01pm

I'm pretty sure bravehoptoad would agree with you. The person he's quoting (Brian Burke of Advanced NFL Stats) doesn't include special teams at all in his mathematical ratings. That has always struck me as ridiculous.

73
by Reinhard (not verified) :: Mon, 01/16/2012 - 5:09pm

and they CAUSED fumbles. and I dont remember but special teams fumbles might be the kind that are recovered more often by kicking team or something.

3
by Dales :: Mon, 01/16/2012 - 1:27pm

DVOA's inexplicable love of the Eagles is peeking through yet again!

Don't worry, Aaron. The Giants confuse everyone.

5
by PaddyPat :: Mon, 01/16/2012 - 1:40pm

You really think Eagle lovin' is still inexplicable? I thought the model of this year's Jets said a lot about DVOA's emotional bonds with teams. Long drives with lots of third down conversions and small gains make DVOA swoon and coo. Big play offenses are like lousy pick-up lines by comparison--good for a laugh, but hardly enough to touch your heart.

9
by Dales :: Mon, 01/16/2012 - 2:01pm

I actually agree. It is explicable. Reid pushes the buttons that DVOA likes having pushed.

It's still fun to point out, though. Good, dangerous team. But not quite as good or dangerous as a straight ranking by DVOA would indicate.

15
by Independent George :: Mon, 01/16/2012 - 2:19pm

I keep telling you guys, it's the color scheme! If Michigan State were in the NFL, they'd be ranked in the Top-10, too.

104
by Pat (filler) (not verified) :: Mon, 01/16/2012 - 6:35pm

Yeah, those buttons like "scoring points" and "preventing the other team from scoring."

It's nothing specific to DVOA. The Eagles dramatically improved over the length of the season, so their weighted DVOA is obviously going to become better and better the farther we get from the early portion of the season. Heck, they finished with the best point differential in the NFC East.

It's a running joke that "DVOA loves the Eagles" but the disparity between their ranking and their performance has always been explainable as something much more simple. It's not like the Jets were this year where you might've had something weird going on.

116
by Dales :: Mon, 01/16/2012 - 7:29pm

Been two years ago or so since I last looked, but the Eagles had a streak of years going where their wins were lower than their expected wins per DVOA. I would suspect, given their lofty DVOA ranking this year, that this data point would add to the effect.

DVOA is based on things that correlate to wins, when a team is behaving like a normal team. Reid's Eagles always do these things. It is a positive attribute. It is a reason that, as a Giants fan, I hope the Eagles do fire Reid. I think he's clearly a big part of their success.

The thing is, the Eagles aren't behaving sufficiently like an average team for that model. Reid will forget about the running game for a long stretch of time, and he's been blessed to have backs who are incredibly well-suited for that offense (Westbrook, Shady). So they are hitting the model's sweet spot at a rate the model suggests should have a remarkable rate of return, but are getting results that have them about where they are every year- average but scary.

Kind of like the Giants, but without the peaks being quite as high.

187
by Whatev :: Tue, 01/17/2012 - 7:43pm

Well, we still don't have a way for DVOA to penalize teams for awful, awful clock management, right? That seems like one thing that could help explain the difference.

116
by Dales :: Mon, 01/16/2012 - 7:29pm

Been two years ago or so since I last looked, but the Eagles had a streak of years going where their wins were lower than their expected wins per DVOA. I would suspect, given their lofty DVOA ranking this year, that this data point would add to the effect.

DVOA is based on things that correlate to wins, when a team is behaving like a normal team. Reid's Eagles always do these things. It is a positive attribute. It is a reason that, as a Giants fan, I hope the Eagles do fire Reid. I think he's clearly a big part of their success.

The thing is, the Eagles aren't behaving sufficiently like an average team for that model. Reid will forget about the running game for a long stretch of time, and he's been blessed to have backs who are incredibly well-suited for that offense (Westbrook, Shady). So they are hitting the model's sweet spot at a rate the model suggests should have a remarkable rate of return, but are getting results that have them about where they are every year- average but scary.

Kind of like the Giants, but without the peaks being quite as high.

150
by Pat (filler) (not verified) :: Tue, 01/17/2012 - 11:50am

Been two years ago or so since I last looked, but the Eagles had a streak of years going where their wins were lower than their expected wins per DVOA.

First, estimated wins is not the number of games that DVOA thinks they should've won. It's the number of games you would estimate them to win against an average schedule. It's not opponent or fumble luck adjusted. Wins are.

As a note, though, Pythagorean wins *aren't* opponent or luck adjusted. And if you look at that, you'll see the Eagles are much closer, except for a few fluke years.

2011: 9.2 estim. wins, 8 actual, 9.8 pyth. wins
2010: 10.8 estim. wins, 10 actual, 9.4 pyth. wins
2009: 10.9 estim. wins, 11 actual, 10.2 pyth wins
2008: 11.7 estim. wins, 9.5 actual, 11.3 pyth wins
2007: 10.2 estim. wins, 8 actual, 9.1 pyth wins
2006: 11.9 estim. wins, 10 actual, 9.8 pyth wins
2005: 7.7 estim wins, 5 actual, 5.9 pyth wins

So if you compare apples to apples - that is, non-opponent adjusted to non-opponent adjusted, the fluke years were 2008 and 2011. That's it. The other years, estimated wins was higher most likely because of opponent and fumble luck adjustments. And DVOA vs VOA agrees on that.

2005: DVOA -2.8%, VOA -10.4% (diff = 7.6%), 2.7 fewer wins than actual
2006: DVOA 27.2%, VOA 22.7% (diff = 4.3%), 1.9 fewer wins than actual
2007: DVOA 13.7%, VOA 8.0% (diff = 5.7%), 2.2 fewer wins than actual

Hey look! The difference between estimated wins and actual wins is almost directly proportional to the VOA-to-DVOA correction. Who'da thunk it!

And 2008 and 2011 have been discussed like crazy. They were weird. Whatever. All teams have those seasons. There is no magic here. The Eagles underperformed because of a difficult schedule (and *not* just the SCHED ranking) and some bad luck. That's it. If there was a "VOA estimated wins" they'd be smack-dab where they're expected to be.

There is no "DVOA loves the Eagles" magic. There is only "the Eagles love opponent adjustments and corrections for fumble luck" magic.

190
by Dales :: Tue, 01/17/2012 - 10:46pm

Estimated wins, as you said, is based on an average opponent. Fortunately, the DVOA listings had the strength of the past schedule. The analysis held, as the Eagles' opponents generally were not far, on average, from the league average.

DVOA "likes" the Eagles to win more frequently than the Eagles win. I stand by the assertion.

191
by Pat (filler) (not verified) :: Tue, 01/17/2012 - 11:00pm

DVOA is not VOA plus opponent adjustments equal to the SCHED rating.

It corrects for fumble luck, weather, opponent field goal percentage, and opponent adjustments based on how the game was played.

The difference between VOA and DVOA entirely explains the difference between estimated and actual wins. I don't know how I can make that clearer. This "DVOA loves the Eagles" thing just has some insane life of its own.

196
by Dales :: Wed, 01/18/2012 - 8:17am

Pat, are you arguing that a team that plays a schedule that have an average opponent DVOA of zero (when the baselines are such that the league average is zero), that estimated wins is not measuring what a team "should" have won against that schedule?

Because if you are not arguing that, then you missed my point.

And, just in case you are planning on going there, I completely get that there is a difference between playing a schedule where every team is perfectly average and playing a perfectly average schedule.

197
by nat :: Wed, 01/18/2012 - 10:26am

As a disinterested observer, I have to say that you seem to be willfully missing the point about VOA.

VOA measures the value over average that actually happened. You don't need to and shouldn't use DVOA - even with strength of schedule backed out - to do what you are trying to do. Neither DVOA nor [DVOA - SoS] is the right stat to track actual in-game results in a season. That's what VOA is for.

So, if you want to argue about FO stats "loving" the Eagles, you need to start with VOA and wins, and to compare the Eagles' VOA-to-wins correlation with other teams in the league.

Simply put: You're doing it wrong.

204
by Pat (filler) (not verified) :: Thu, 01/19/2012 - 1:33am

The slight caveat there is that VOA isn't intended to correlate to wins, it's intended to correlate to points. Estimated wins is intended to correlate to wins, but it's not intended to "predict" wins, it's more "this is the number of wins a team like this usually gets."

So you would actually want to check the VOA-to-points correlation. The connection between points and wins you don't need to check because the Eagles are already in line with their Pythagorean projection.

Or, you could just look at the big VOA-to-DVOA correction for Philly in those years, then remember that Philly ended up 'surprising' everyone late in the year by making the playoffs or beating good teams, and realize that hey, look, maybe that DVOA made sense, and they really did get unlucky.

Plenty of other teams have years where their VOA is way below their DVOA, and their wins are - correspondingly - way below their estimated wins. Philly happened to have a few of those years in a row. Other teams have too - Cleveland from 2008-2010 had fewer wins than estimated wins, and (surprise!) lower DVOA than VOA. It's just no one really cares when DVOA has Cleveland ranked 25th instead of 31st.

205
by nat :: Thu, 01/19/2012 - 9:57am

Your caveat is correct. VOA is designed to reward plays that improve the net average value of the next score. The result is that it's well correlated (see the FAQ) with wins, because a good way to win is to score more points than your opponent.

You make a good point about the Pythagorean projection, too. If that's already consistent with their records, then they don't have a "can't put a drive together when it matters" problem.

If their VOA is in line with their points, then they don't have a "can't make the important play" problem.

203
by Pat (filler) (not verified) :: Thu, 01/19/2012 - 1:15am

Pat, are you arguing that a team that plays a schedule that have an average opponent DVOA of zero (when the baselines are such that the league average is zero), that estimated wins is not measuring what a team "should" have won against that schedule?

Yes, that's right. That is not what it's measuring. Just because the average opponent DVOA is zero doesn't mean that their wins would have matched their estimated wins.

1) Estimated wins uses DVOA, not VOA - which means it's got fumble luck correction and the other VOA-to-DVOA corrections other than opponent adjustments (opposing field goal percentage, opposing kickoff length, weather). If a team gets screwed by fumble luck (or the others), their DVOA will be much higher than their VOA, as will their estimated wins. Oh, and they'll probably lose those games where they lost a bunch of fumbles. You may remember Philly fans (and Aaron) pointing out that it seemed for a while like a team could've just lined up for an 80 yard field goal and made it against the Eagles. It was silly. And it cost them games.

2) The opponent adjustments are not based on the team's total DVOA! They're based on what happens in the game. If team 1 has a run defense of -1000%, and team 2 never runs against them, they don't get a boost to their performance because team 1 could've theoretically stopped a run that they didn't do. You can't look at the SCHED rating. You have to look at the VOA-to-DVOA difference.

My point is really, really simple. Look at the Eagles' VOA in those years (except the fluke years I mentioned). It's always lower than their DVOA. It's usually where people would expect them to be, ranking-wise. The reason the Eagles' DVOA is so high has absolutely nothing to do with their playcalling, the style of drives, or whatever other crazy theory someone's come up with. It's entirely due to the opponent and luck adjustments.

The VOA-to-DVOA correction entirely explains their estimated wins being higher than their actual wins. This is expected. Therefore inventing some other thing is entirely unnecessary.

206
by Dales :: Fri, 01/20/2012 - 8:40am

Well, I hope a few days does not mean that you won't come back. Been busy.

I think we are talking past each other. So let me try it a different way.

Let's take three teams. Baltimore, Philly, and Pittsburgh. And let's use the last 6 years (stipulate the number of years for the time being).

Team / Avg DVOA / Avg VOA / Avg Opponent DVOA
Bal / 20.7% / 19.0% / 1.9%
Pit / 22.9% / 22.7% / 1.5%
Phi / 23.1% / 21.5% / 1.3%

I think these are good comps. All are close to each other in DVOA, in VOA, and in Avg Opponent DVOA for the 6 year period. And, as a nicety, since the claim was made that the VOA to DVOA adjustment explains all of the difference between estimated and actual wins, having Bal having nearly the same adjustment (but slightly greater) is a bonus.

Let's throw in wins and estimated wins.

Team / Avg DVOA / Avg VOA / Avg Opponent DVOA / Est Wins / Actual Wins
Bal / 20.7% / 19.0% / 1.9% / 64.8 / 62
Pit / 22.9% / 22.7% / 1.5% / 63.9 / 63
Phi / 23.1% / 21.5% / 1.3% / 64.7 / 56.5

Looks to me like Philly was about 1 game per year worse than similar teams by DVOA, VOA and opponent's DVOA.

Just for giggles, let's do the same chart, just for the NFC East, sorted by actual wins.

Team / Avg DVOA / Avg VOA / Avg Opponent DVOA / Est Wins / Actual Wins
NYG / 11.7% / 7.8% / 2.7% / 56.6 / 57
Phi / 23.1% / 21.5% / 1.3% / 64.7 / 56.5
Dal / 10.1% / 10.1% / 1.6% / 54.9 / 56
Was / -3.7% / -7.7% / 2.5% / 44.7 / 37

If one looks at NYG/PHI/DAL in the metrics of actual wins and average opponent's DVOA, they are pretty good comps for each other. The Eagles' much higher DVOA is what leads to the much higher estimated wins. But it is not the gap between DVOA and VOA that led to the difference between estimated and actual wins; the Giants VOA is even lower relative to DVOA than the Eagles'. And their opponents have had an average DVOA that is even higher.

(Note- I am open to the idea that the Giants are just a screwed up team and should not be used to draw any conclusions.)

I assert that the opponent component here is essentially a wash, especially over several years when the Avg Opponent DVOA is within a point or so. I suppose one could say that "just because you face a team that is a 20% DVOA quality team does not mean they were when you played them" either due to matchup problems or due to variance. The former strikes me as indicating that your perceived strength is too high if you can consistently be exploited to where you don't win as often as you "should". The latter would just be another form of luck.

Which gets to my main point. At some threshold, one has to start wondering if it really is luck or if it is something else.

Back to the whole 6 years thing. There is one reason I used that-- cherry picked the end point! Each year since then the Eagles either lagged their estimated wins or barely met them. Not surprising in some ways, since they've had the 3rd highest estimated wins over that time (NE has the most with 72.7, which they exceeded at 76, and Baltimore at 64.8 which they fell short of at 62).

But the other teams that have won fewer than 90% of their estimated wins over that time are Cle (79%), Det (86%), Buf (87%), TB (89%), Was (83%), Jac (88%) and Mia (80%). Those all have been mediocre to terrible teams over most of that time. The Eagles are the only good team like this. Fluke? Or something about the team?

As it happens, prior to these 6 years, the Eagles were quite different. They tended to exceed their expected wins every year, as consistently as they under perform it now. It seemed to change when Owens was suspended and then disappeared and McNabb got injured.

Could it be luck? Certainly. But explanations such as "DVOA higher than VOA" don't cut it for me, since other teams with that attribute do not have the same consequence. Opponents don't cut it for me, since teams with very similar schedules don't seem to have the same consequence.

They may in fact just have been a fluky and unlucky team over the last few seasons. But it sure won't shock me if next year the Eagles end up ranked higher in DVOA than their record would suggest, or if the above trends continue.

Keep an open mind to the thought that Reid has developed some tendencies that work better in this particular statistical model than they do on the football field. It is something that I wondered about, because it fits both my perception and what I see in the data.

That is a thought that is not at all at odds with the fact that Reid consistently puts together a damn fine football team- and a very dangerous one at that.

211
by Pat (filler) (not verified) :: Tue, 01/24/2012 - 1:54pm

First: I don't know why you keep leaving off Pythagorean wins. The point of including Pythagorean wins is that Pythagorean wins have nothing to do with DVOA. If a team's underperforming their Pythagorean wins, it's not a "DVOA loves that team", it's a "that team is losing for weird reasons."

BAL: 63 Pyth. wins (62 actual)
PIT: 64.7 Pyth. wins (63 actual)
PHI: 59.6 Pyth. wins (56.5 actual)

Half the gap between BAL/PIT and PHI's "estimated wins" and "actual wins" is explained just by the fact that they scored and prevented points better than you would have expected from their record. This has nothing to do with what DVOA "likes" or "doesn't like."

That's pretty much the entire point. People boil this down to "DVOA likes the Eagles" but probably a better argument would've been "Philly really seemed to not know how to win close games." You can either believe there was a reason for that, or believe it was luck. Whichever.

212
by tuluse :: Tue, 01/24/2012 - 1:58pm

Without looking really closely at the game logs, it could also mean the Eagles are really good at beating up on weak teams, but struggle more than one would expect against stronger teams.

Given their success in the playoffs this seems unlikely.

213
by Dales :: Sat, 01/28/2012 - 11:30pm

" I don't know why you keep leaving off Pythagorean wins."

I have no problem with them being included. The Eagles' performance in that metric is completely consistent with the fact that they aren't as good as the teams that are their best DVOA comps are.

4
by PaddyPat :: Mon, 01/16/2012 - 1:38pm

Woohoo! Pats defense is again better than GB's!! I love how much the weighted Special teams moves around. Tracking those numbers, you can really see how just a few plays on Special Teams can totally transform a game.

6
by PaddyPat :: Mon, 01/16/2012 - 1:43pm

"Fumbles are such a bummer."

I tried to comment on this in the Sunday Games' thread. It seems like fumbles have been a huge part of the outcomes of these playoff games. 4th downs and fumbles, which is just interesting because I think of picks as being more salient in the regular season. Fumble luck has played huge this postseason. If Baltimore, for example, had not fallen on so many fumbles, that Houston-Baltimore game might have turned out differently.

7
by Jim Z. (not verified) :: Mon, 01/16/2012 - 1:53pm

I refuse to believe that the Giants are anything more than a fluke.

How can the entire history of the NFL be correct - in that teams with high DVOA invariably are the most successful in the playoffs and in the end - except with regard to ONE particular team (the 2006-2011 era Manning Giants).

I've watched the Giants play every game this year, and in the majority of their games in the regular season, their defense was absolutely putrid. All of the sudden, they "turn it on" for the playoffs? How is this possible? Why had this not happened at any other time in the history of the NFL playoffs? Something like 95% of the Super Bowl winners in DVOA history have finished in the top-5 DVOA, and yet the 2007 Giants, and if they make it, the 2011 Giants, will have finished outside the top-10!

I refuse to accept that the NFL is decided by randomness and "hot streaks", because in every other instance, or nearly every other instance, in the history of the playoffs, it has *not* been decided by "hot teams". In the 90s, the Cowboys, 49ers, Packers, and so on were all dominant regular season teams that rightfully won their Super Bowls. In this past decade, the Steelers Super Bowl teams were dominant defensive teams that rightfully won their Super Bowls. The Patriots were dominant, rightful champions. As was nearly every other team except for the 2007 Giants. And if the 2011 Giants do not lose to the 49ers next week, they will join the 2007 Giants as undeserving Super Bowl participants flouting decades of historical data.

8
by Jim Z. (not verified) :: Mon, 01/16/2012 - 1:55pm

I mean, what's the point of watching the regular season if it doesn't matter? If all that matters is a few games in January?

Because, the regular season *should* matter, and in 95% of seasons, it *does* matter and it does reveal the likely Super Bowl winner. Except in the case of the 2007 Giants, and the 2011 Giants if they prevail next week and after that.

I mean, the Giants were OUTSCORED this year. They have a NEGATIVE point differential. That's insane. This team absolutely does not deserve to be even playing in the Super Bowl. It makes a mockery of the regular season.

12
by Dales :: Mon, 01/16/2012 - 2:05pm

You should try being a fan of the team. Their peaks and valleys and swoons you can see coming a mile away are indigestion-inducing.

A lot of this year's defensive struggles can be attributed to injuries, some of which have gotten better, coupled with what the team called communication issues, which seem to have been addressed.

The run this year makes a lot more sense than the run they put together in 2007.

16
by Scott P. (not verified) :: Mon, 01/16/2012 - 2:20pm

It almost happened in 2003. Carolina was 17th in Weighted DVOA that year.

135
by dryheat :: Tue, 01/17/2012 - 9:32am

I would say that if they defeat Atlanta, Green Bay, and San Francisco in successive weeks, the Giants absolutely do deserve to be playing in the Super Bowl.

Good teams, and well-coached teams, tend to get better over the course of the season. In hindsight, the 2007 Giants (who greatly improved over the course of the season) looked less fluky after they ran of a 7-0 record in 2008.

167
by dmstorm22 :: Tue, 01/17/2012 - 1:30pm

They didn't start 7-0, as they lost a bizarre game early to Cleveland 35-14. They did start 11-1 however.

10
by Anonymous1111111 (not verified) :: Mon, 01/16/2012 - 2:01pm

That is clearly the dumbest thing I have ever heard. While statistics are important, they generally present probabilities as to what is likely to happen - NOT what will happen. Accordingly, the fact that a team has a tough schedule, loses a number of close games, ends up without a top ten DVOA, and then ends up beating those same teams that it lost to by a few points in the regular season (Week 17 against the Pats in '07) in the playoffs, does not make it undeserving. Injuries and health are critical and may not show up in statistics. In other words, without injuries the Giants may be a top 10 DVOA team, which they are finally playing as (with Osi back, Tuck healthy, Boley back, and a solid filler at MLB).

114
by Ferguson1015 :: Mon, 01/16/2012 - 7:25pm

Without injuries many teams could be a top 10 DVOA team. Injuries happen, even to teams that made the top 10 this year.

11
by Aaron Schatz :: Mon, 01/16/2012 - 2:04pm

... and the 2008 Arizona Cardinals. Don't forget them. It's not just the Giants.

13
by PaddyPat :: Mon, 01/16/2012 - 2:18pm

I've been hitting this point for years now. Whether or not the top seed wins a super bowl isn't really the point. The Divisional round of the playoffs is supposed to protect the best teams from the phenomenon of every given Sunday. Because Football can't play series (due to injuries) we all know that a team like the Packers can, under the right circumstances, lost to anyone, so we nestle them with a home game and a bye to protect them. It used to work. Between 1978 (the first year with the bye) and 2003, the last year with nonenforcement of the 5-yard contact rule, there was an average of .923 upsets out of 4 games in the divisional round of the playoffs annually. There were 4 nutty years with 2 upsets and many many years with none. Most of the upsets concerned strong teams, ie. 12-11 wins beating marginally better teams--12 beats 13 wins, etc. Since 2004, which, incidentally was a year with no upsets, the number of annual divisional round upsets has skyrocketed to 1.625! Even if we take out the insane 2008 season, there's still a correlation in the data that is 92 percent unlikely to have occurred by chance. You can't argue small sample size, the correlation is powerful! It's not about free agency either. The trend still shows in the data if you put the break around 1995, but it drops off to more like 68 percent power with .882 upsets followed by 1.294. The game of football has changed in this passing era, and what it means is simply that the good teams are less stably good, and the not quite as good teams are hardly as lacking in bite. You see it everywhere in the divisional round: 9-7 beats 15-1, 9-7 beats 12-4, 9-6-1 beats 12-4, 11-5 beats 14-2, 11-5 beats 13-3, etc, 10-6 beats 13-3, etc. The regular season is not as important as it used to be!

18
by Karma Coma :: Mon, 01/16/2012 - 2:39pm

What happens to those "upset" totals if you use Pythagorean wins or DVOA instead of raw wins or, even worse, playoff seed?

20
by PaddyPat :: Mon, 01/16/2012 - 2:47pm

This is a good observation and I'm not sure, because we don't have Footballoutsiders data for most of these seasons. But, it's still not really the point. Regardless of their pythagorean wins, the Home teams in the divisional round always won. They weren't stronger teams either--10-6 teams with a bye and homefield beating other 10-6 teams. I think the point here is that a bye and homefield used to be a bigger advantage than it is now. My understanding here is that performance was just more reliable game to game so that if you took a pretty good team and gave it a home game and bye, it would beat most other good teams, hence, the teams that won the most games in the regular season, regardless of pythagorean projections and actual DVOA-type efficiency, would get to the Conference Championships with great reliability because they were always "rewarded" with the byes. That's the incentive for winning in the regular season--wrap up homefield and a bye, and your chances of making it to the Super Bowl are really good. It's not true any more. The top seeds are winning at a rate that's maybe comparable to regular season homefield advantage.

192
by Scott C :: Wed, 01/18/2012 - 1:40am

The first home team to lose this season was Green Bay.

This year:

Home Team W-L = 7-1

Two more games, of both go to the road team, it will still be 7-4.

Small sample sizes do funny things. To get a reasonable measure of the rate that teams with byes win their first game, you would need at least 10 years of data, and at least 15 if you segregate by teams that have at least +2 regular season wins and are the home team.

I believe that the home team and bye are a little weaker advantages than 25 years ago, but the advantage is still there.

34
by BSR :: Mon, 01/16/2012 - 3:31pm

I think you hit upon the answer when you mention the "nonenforcement of the 5-yard contact rule". I think what has slowly happened since this change in 2004 is that refs call the rule more closely in the regular but have now going back to the nonenforecment in the playoffs. The inconsistency between regular season and playoffs is staggering. Watching all the games this past weekend and last, I noticed far fewer defensive holding/interference penalties in comparison to the regular season. Baltimore has zero penalties in their game. This difference in penalty enforcement is suddenly making mediocre teams during the regular season more competitive as their pass rush is given more time to disrupt passing.

49
by PaddyPat :: Mon, 01/16/2012 - 4:13pm

Wow. Really interesting point. I hadn't even thought of this. Wonder if this means the Pats are more prepared than GB and NO--the Tight Ends have less trouble fighting through holding than smaller receivers--I mean, it's pretty hard to mug Gronkowski! But just... wow! Good idea!

193
by Rhys :: Wed, 01/18/2012 - 3:47am

In Foxboro Stadium, the Tight End mugs you!

99
by MJK :: Mon, 01/16/2012 - 6:30pm

It also could be a simple matter that the rule change made long passes more likely to be successful (or draw a spot-foul PI call). Long passes, and DPI even moreso, are a great equalizer between a good team and a bad team. Consider...

Good team drives down the field three times at will, mixing efficient passing with effective, hard running, and scores a TD and two FG's. Bad team plays terrible, getting stuffed on their runs and throwing wildly inaccurate passes, but completes one long pass where safety falls down and gives up a TD, and has another bogus DPI call in their favor in the end-zone, and gets a second easy TD.

Bad team now leads 14-13.

102
by Will Allen :: Mon, 01/16/2012 - 6:35pm

Again, I think if you go back to 1978, and don't mention the biggest change to NFL rosters since that time, which is the salary cap, you are missing the biggest part of the story. The best roster managers now simply cannot stockpile talent in the way it used to be done. I think the full effect of the cap took place over time, and by the mid 90s or so, we were simply in a different league. I think the rule changes may further emphasize that trend, in that it makes the roster construction so centered on the qb, until we get to the point where 14-2 or 15-1 teams, while great at passing, have deficiencies that such teams in the past did not have.

Put it this way; even with the current rules, and adjusting for 20 years of general atheletic improvement in human beings, the '89 Niners beat this year's Packers fairly comfortably four times out of five.

14
by JIPanick :: Mon, 01/16/2012 - 2:18pm

...and the '01 Pats, and the '05 Steelers, and the '03 Panthers, and the '06 Colts (although they get a little bit of a pass for being a good team having a bad year), you get the idea.

It's beyond disgusting. I want good champions, not penny-ante wildcards that make fluky runs.

21
by Jim Z. (not verified) :: Mon, 01/16/2012 - 2:47pm

The 05 Steelers at least were a top-3 DVOA team the year prior, and probably would have been the top DVOA team in 2005 barring injury, and yet, still finished #4 in total DVOA at the end of the season.

The 06 Colts finished the season #7 in total DVOA, so they were at least top-10, AND they finished as the #1 team in offensive DVOA. They were a dominant team with a mediocre defense.

So really, you're looking at the '03 Panthers, who at least didn't win the Super Bowl, and the '01 Pats, the '08 Cardinals, and the '07 Giants as the flukes of the Super Bowl era in recent memory.

24
by PaddyPat :: Mon, 01/16/2012 - 2:57pm

The comment about the Steelers is true, but weren't we all really expecting 2005 to be the Colts' year? They finally got the monkey off their backs with Belichick, knotted up the 1 seed, won 14 games, put it all together, etc. etc. And 2006, surely that was finally Marty's year, or failing that, if he's just ultimately a playoff incompetent, that was Steve McNair's big year with the Ravens--Billick finally had an offense and the Colts couldn't stop anyone on the ground. Those were the great teams those years. 2007, well it was the Pats vs. the Cowboys right? Great shoot 'em up showdown, or Pats-Packers rematch! 2008 was the blisteringly tough year when all the best teams played defense--Tennessee, Pittsburgh, New York, and Carolina. I was looking forward to something strange like Tennessee-Carolina, but New York Giants-Pittsburgh Steelers was a good match up too. I could go on.

27
by Jim Z. (not verified) :: Mon, 01/16/2012 - 3:08pm

In 2005, both the Steelers and the Colts played well throughout the regular season. The 2005 Steelers upsetting the Colts is not a strange outcome, as the Steelers were simply slightly less dominant in the regular season.

In 2006, the Chargers and Colts were both great teams, and the Colts were the #1 offense in the football, so an upset of the Chargers by the Colts was not unfathomable.

The 2007/2011 Giants simply do not belong in the same class as the 2005 Steelers or 2006 Colts. They were downright *awful* in the regular season (especially on defense), the 2011 Giants even more so than the 2007 team.

32
by Aaron Brooks Go... :: Mon, 01/16/2012 - 3:22pm

Wasn't 2006 a classic case of good team gets a foundation player back from injury and becomes great? That defense was stellar for the 3.5 games every season where Bob Sanders was alive.

53
by Anonymous1 (not verified) :: Mon, 01/16/2012 - 4:24pm

You would think so, but Indy was Sanders was equally terrible in the season.

What made Indy's defense look great is changeup on DL responsibility (crashing down with the DEs rather than simply running upfield) and two offenses that seemed unwilling to adjust. NE was taking it to them until their RBs started dropping like flies and Chicago was a joke offensively.

It is similar to this year's Patriots, who could make it to the SB without having to face a single competent passing offense, and then getting to avoid the best passers in the NFC as well.

29
by JIPanick :: Mon, 01/16/2012 - 3:15pm

#4 DVOA or not, they were not on the same level as the Colts or Broncos.

#7 DVOA is pretty dang weak for a Super Bowl champion.

Let me put it this way: were I a betting man, I'd bet that between 1989 and 2002, aside from he '01 Pats, no team ranked meaningfully* lower than #2 in DVOA won a Super Bowl. Both the '05 Steelers and the '06 Colts would have been the worst Super Bowl champion of the decade, easily, had they won ten years earlier. Heck, so would the '03 Pats.

*This is to exclude teams like the '97 Broncos, who were technically 3rd but were only 0.6% out of first.

56
by Independent George :: Mon, 01/16/2012 - 4:32pm

The '05 Steelers had Tommy Maddox & Charlie Batch start 4 games, and still ended up #4 DVOA. If I recall correctly, most of us had made that mental adjustment and considered the Colts game an upset, but favored them over the Broncos and the Seahawks.

The '06 Colts were weak on account of that run defense, but I was also remember everyone not on the "Peyton can't win the big one" bandwagon thought that if anyone could overcome that defense, it was Peyton. As it turned out, he didn't have to overcome anything.

58
by Jonadan :: Mon, 01/16/2012 - 4:35pm

Here's the record by DVOA for anybody who didn't check already and is curious. If there's data for 1989-1991, I can't find it (premium maybe?)

'92 (XXVII) Cowboys (#1) over Bills (#7, #2 AFC)
'93 (XXVIII) Cowboys (#2, #2 NFC) over Bills (#12, #7 AFC)
'94 (XXIX) 49ers (#3, #2 NFC) over Chargers (#7, #4 AFC)
'95 (XXX) Cowboys (#2, #2 NFC) over Steelers (#4, #2 AFC)
'96 (XXXI) Packers (#1) over Patriots (#8, #3 AFC)
'97 (XXXII) Broncos (#3, #2 AFC) over Packers (#1)
'98 (XXXIII) Broncos (#1) over Falcons (#7, #3 NFC)
'99 (XXXIV) Rams (#1) over Titans (#5, #4 AFC)
'00 (XXXV) Ravens (#3, #3 AFC) over Giants (#10, #3 NFC)
'01 (XXXVI) Patriots (#12, #6 AFC) over Rams (#1)
'02 (XXXVII) Buccaneers (#1) over Raiders (#2)
'03 (XXXVIII) Patriots (#4, #4 AFC) over Panthers (#17, #8 NFC)
'04 (XXXIX) Patriots (#2, #2 AFC) over Eagles (#7, #1 NFC)
'05 (XL) Steelers (#4, #3 AFC) over Seahawks (#3, #1 NFC)
'06 (XLI) Colts (#7, #5 AFC) over Bears (#6, #2 NFC)
'07 (XLII) Giants (#16, #8 NFC) over Patriots (#1)
'08 (XLIII) Steelers (#2, #1 AFC) over Cardinals (#21, #12 NFC)
'09 (XLIV) Saints (#6, #4 NFC) over Colts (#8, #3 AFC)
'10 (XLV) Packers (#3, #1 NFC) over Steelers (#2, #2 AFC)
---
'11 (Highest hypothetical for XLVI): Patriots (#4, #2 AFC) vs 49ers (#6, #3 NFC)

Observations:
- JIPanick would be correct. Betting with the whole internet at your disposal means you're playing with house money, though, so no great credit (sorry).
- More as a case of trivia, the two best teams from each conference by DVOA have only faced each other in the Super Bowl once: in 2002, Bucs over Raiders.
- The following Super Bowls were upsets by DVOA: '97 Broncos over Packers; '01 Patriots over Rams; '05 Steelers over Seahawks; '06 Colts over Bears; '07 Giants over Patriots; '10 Packers over Steelers.
- The difference doesn't seem so much getting to the Super Bowl (though even there the upsets have increased) as winning the thing as an underdog. The data would suggest that there was a significant change (though not in 2001): I would posit the culprit is probably the 2002 realignment, mostly due to the decrease in division games (from a minimum of 8 to an even 6). Why did that matter? Let's demonstrate: the worst division winner in 2001 was 5-3 in division; the next year they were 4-2 – I'd think of it as one (or two) less games that "matter". I'm not sure there's a way to correct that, to be honest, with 32 teams (we'll have to wait for 36?), though the obvious suggestion would be to actually seed by record.

---
"When you absolutely don't know what to do any more, then it's time to panic." - Johann van der Wiel

78
by JIPanick :: Mon, 01/16/2012 - 5:18pm

There is no data for 89-91, hence the "bet". The only place I think I might miss it is if the '90 Giants are significantly behind both the Bills and the 49ers - I'd be shocked if SF '89 and WAS '91 aren't tops.

86
by PaddyPat :: Mon, 01/16/2012 - 5:51pm

Thoughtful post, but you're missing the point. The discussion here is about upsets in the divisional round of the playoffs. Any home team losing is a big deal considering the bye and the homefield. So there are probably two problems at work here. First, the home teams in the divisional round seem less likely to be the best teams via DVOA of late. Second, the best teams are not doing the job of winning. Let's go through your same discussion looking at the Divisional Round, although obviously, we have a truncated data set without the years 1978-91:

1992:
35.4, 13-3 at home beat 33.7, 9-7.
31.3, 14-2 at home beat 18.2, 9-7.
10.2, 11-5 at home beat 12.0, 11-5.
16.9, 11-5 at home lost to 17.1, 11-5.

1993:
25.6, 10-6 at home beat 13.4, 11-5.
25.0, 12-4 at home beat 10.9, 9-7.
9.4, 12-4 at home beat 10.4, 10-6.
17.1, 12-4 at home lost to 21.5, 11-5.

1994:
29.2, 13-3 at home beat -2.8, 9-7.
33.7, 12-4 at home beat 23.1, 9-7.
30.7, 12-4 at home beat 10.8, 11-5.
11.8, 11-5 at home beat 19.0, 10-6.

1995:
33.5, 12-4 at home beat -2.6, 10-6.
43.3, 11-5 at home lost to 16.3, 11-5.
21.8, 11-5 at home beat 1.4, 10-6.
28.6, 13-3 at home lost to -8.3, 9-7.

1996:
41.3, 13-3 at home beat 29.7, 12-4.
20.5, 12-4 at home beat 24.0, 10-6.
27.7, 13-3 at home lost to -0.6, 9-7.
10.4, 11-5 at home beat 18.5, 10-6.

1997:
30.1, 13-3 at home beat 9.2, 10-6.
28.5, 13-3 at home beat -4.1, 9-7.
29.7, 13-3 at home lost to 29.5, 12-4.
28.3, 11-5 at home beat 12.3, 10-6.

1998:
22.9, 15-1 at home beat -17.6, 9-7.
19.0, 14-2 at home beat 25.8, 12-4.
31.2, 14-2 at home beat 24.1, 10-6.
28.7, 12-4 at home beat 8.0, 11-5.

1999:
34.3, 13-3 at home beat 11.0, 10-6.
11.7, 11-5 at home beat 14.0, 10-6.
26.5, 14-2 at home beat 6.7, 9-7.
3.9, 13-3 at home lost to 16.7, 13-3.

2000:
12.4, 12-4 at home beat 17.7, 11-5.
-5.3, 11-5 at home beat 1.5, 10-6.
33.2, 13-3 at home lost to 24.1, 12-4.
22.2, 12-4 at home beat 25.0, 11-5.

2001:
28.9, 14-2 at home beat 19.2, 12-4.
16.5, 13-3 at home lost to 28.4, 11-5.
18.4, 13-3 at home beat 7.8, 10-6.
7.7, 11-5 at home beat 20.1, 10-6.

2002:
23.5, 12-4 at home beat 10.8, 9-6-1.
33.4, 12-4 at home beat 14.0, 10-6.
29.6, 11-5 at home beat 16.0, 9-7.
9.7, 11-5 at home beat 9.0, 10-5-1.

2003:
20.0, 12-4 at home beat 21.8, 10-6.
2.3, 12-4 at home lost to 0.8, 11-5.
21.8, 14-2 at home beat 23.0, 12-4.
30.3, 13-3 at home lost to 22.0, 12-4.

2004:
23.2, 13-3 at home beat -7.7, 8-8.
-5.0, 11-5 at home beat -26.5, 8-8.
38.1, 15-1 at home beat 27.5, 10-6.
35.9, 12-4 at home beat 28.1, 12-4.

2005:
29.7, 13-3 at home beat 22.3, 10-6.
0.8, 11-5 at home lost to 18.6, 11-5.
33.6, 14-2 at home lost to 27.9, 11-5.
31.1, 13-3 at home beat 9.1, 10-6.

2006:
23.2 13-3 at home beat -12.9, 9-7.
6.6, 10-6 at home beat 27.2, 10-6.
29.1, 14-2 at home lost to 27.1, 12-4.
28.8, 13-3 at home lost to 17.3, 12-4.

2007:
22.9, 13-3 at home lost to 1.1, 10-6.
19.9, 13-3 at home beat 14.2, 10-6.
53.1, 16-0 at home beat 23.7, 11-5.
30.7, 13-3 at home lost to 18.4, 11-5.

2008:
27.0, 12-4 at home lost to 33.5, 9-6-1.
19.0, 12-4 at home lost to -3.2, 9-7.
26.9, 13-3 at home lost to 29.1, 11-5.
29.6, 12-4 at home beat 17.5, 8-8.

2009:
23.5, 13-3 at home beat 11.4, 10-6.
18.4, 12-4 at home beat 25.5, 11-5.
17.9, 14-2 at home beat 30.5, 9-7.
13.6, 13-3 at home lost to 16.9, 9-7.

2010:
14.9, 13-3 at home lost to 23.4, 10-6.
2.0, 11-5 at home beat -24.0, 7-9.
45.4, 14-2 at home lost to 18.3, 11-5.
36.8, 12-4 at home beat 22.5, 12-4.

2011:
28.3, 15-1 at home lost to 9.0, 9-7.
19.0, 13-3 at home beat 23.3, 13-3.
22.5, 13-3 at home beat -11.0, 8-8.
17.2, 12-4 at home beat 19.4, 10-6.

Okay, that's a lot of data. I think there are some interesting trends. First, the top seeds were generally the stronger DVOAs before 2004. I imagine that trend would continue if you traced things back farther toward 1978. That's not as often the case now. A lot of times the divisional round is actually a matchup between a stronger road team and a home impostor. There are also more straight-up upsets in both directions—weak home team defeats stronger challenger and weak challenger beats strong home team. Moreover, and I have no idea if this is statistically significant, you get the sense that there are more out and out bad teams making it to the divisional round!

108
by PaddyPat :: Mon, 01/16/2012 - 6:49pm

The number of road teams with stronger DVOA (throwing out differences of +/- 3 DVOA points) in the divisional round per year since 2004 is a less compelling correlation than it initially appears. There were 0.833 out of 4 games between 1992 and 2003 and then 1.0 out of 4 between 2004 and 2011. That's somewhat marginal in statistical power, maybe 25-30 percent likely to have occurred by chance. Even so, it's interesting. I could probably come up with more measures to demonstrate that seeding and overall team strength are much less stable than they used to be.

19
by Quixzlizx (not verified) :: Mon, 01/16/2012 - 2:44pm

Perhaps college football would be a more worthy sport to follow for those of you who can't bear the concept of playoffs.

22
by PaddyPat :: Mon, 01/16/2012 - 2:48pm

My thoughts here are not complaints. I'm a little unnerved by what's happened to the sport and am trying to adjust. I think football is more like poker now. The best players win marginally more than the jokers, but every tournament is likely to have a jokers all the way to the final rounds.

23
by Jim Z. (not verified) :: Mon, 01/16/2012 - 2:51pm

The NFL playoffs were fine until 2007 (you could argue maybe 2001).

Again, 95% of Super Bowl winners, and even participants, in the history of the game were legitimate dominant teams in the regular season, by any measure (DVOA or traditional).

The 80s/90s playoffs were filled with legit., deserving winners such as the Cowboys dynasties, the 49ers dynasties, the Packers dynasties, the Broncos dynasties... Even in this past decade, the Steelers were a legit. defensive powerhouse throughout the regular season, as were the Patriots.

I want to see those teams in the Super Bowl, not some middling team that took the regular season off and inexplicably gets into the Super Bowl. It offends me as a statistician and as a football fan.

25
by Quixzlizx (not verified) :: Mon, 01/16/2012 - 3:00pm

You liked the NFL playoffs when they simply confirmed that the best regular-season teams were indeed the best teams overall, and now you don't like them when the results diverge from the regular season. My original comment stands, since you only enjoyed the use of a playoff system when it affirmed a college-football style of end-season rankings. If you only desire the usage of playoff games when they confirm the end of regular-season rankings, why have a playoff at all? Perhaps we should just hand the Lombardi trophy to the DVOA winner instead.

26
by Jim Z. (not verified) :: Mon, 01/16/2012 - 3:05pm

I don't mind upsets - if the team that is pulling off the upset is at least a worthy playoff contender, i.e. not a 9-7 team that was outscored in the regular season.

For example, the 1998 Falcons upsetting the 1998 Vikings is a perfectly acceptable playoff outcome. That's why the playoffs were instituted - to determine, via a single game, the "better" team. This only works if the "better" team is simply a a less dominant team. If it is an outmatched, mediocre team inexplicably outplaying superior opponents in multiple consecutive weeks, then I find myself disliking the format, or at least the inscrutable machinations responsible for the unsual outcomes.

36
by Jonadan :: Mon, 01/16/2012 - 3:43pm

It's always seemed to me like the justification for a playoff is, at its most basic, the fact that everybody can't play everybody, thus some sort of "unbalanced schedule" is unavoidable. You see this in the original World Series, Super Bowl, Stanley Cup as well I think though I don't really follow hockey, where the game was basically an inter-league championship. Conversely, the EPL and other European soccer leagues don't have playoffs as such, because the entire format of the league demands a full round robin.

Of course, playoffs also make for good drama, good TV, high ratings, so even when not strictly demanded by the season format (see: MLB's slowly expanding postseason), they're probably worth it to the league.

Anyway, I can't help thinking there's a fine line between (to use the college football argument) including everyone who belongs and letting in teams that don't. The last few years have demonstrated one problem with the NFL's "automatic bid" for division champions – the bit where it's possible to win a division with a losing record. But the Giants did make the playoffs – with the NFC's 6th best record, and of the teams left out in the AFC only the Titans equalled the 9-7 mark and I'm absolutely certain the Giants are better than the Titans – so if they can make the run, I don't see how we call them "unacceptable". The games seem to agree: so far they've humiliated the Falcons and thrashed the Packers.

If what you're trying to say is that the playoff field needs to be smaller, then say that; but you seem to be blaming the Giants for not losing like they were supposed to, which seems odd.

---
"When you absolutely don't know what to do any more, then it's time to panic." - Johann van der Wiel

37
by Andrew Potter :: Mon, 01/16/2012 - 3:49pm

Even in European soccer, the top competition on the continent - the UEFA Champions League - is a playoff for precisely the reason you suggest, that it's not possible for every team to play every other so an elimination tournament is necessary.

43
by Jim Z. (not verified) :: Mon, 01/16/2012 - 4:02pm

It would be wholly unacceptable to see the Giants even in the Super Bowl, let alone winning it, due to the simple fact that they were *terrible* in teh regular season. Records aside, did you notice their 2011 season DVOA? The fact that they were *outscored* in the regular season? Let me repeat that - they were *outscored* in the regular season. Never, in the history of the NFL, has a team that was *outscored* in the regular season won a Super Bowl.

The 2011 Giants should be considered a travesty of a Super Bowl contender if you care at all about regular season statistics, which is why we are even here at FootballOutsiders.com, right?

45
by tuluse :: Mon, 01/16/2012 - 4:11pm

I'm here because I want to understand the game better. Statistics is a good start, but it's the end all be all.

Also, the Giants won their division and were 12th by DVOA, that's a legitimate playoff contender by most measures.

48
by Dales :: Mon, 01/16/2012 - 4:13pm

The Giants were not "terrible" in the regular season. They played the hardest schedule in the NFL and had a winning record. They played the best teams in the NFL's regular season and won one (vs. the Pats), lost competitively in two (Green Bay and San Fran), and got skunked in one (NO). That is not optimal. It is not good. But it is far, far, far from terrible.

The fact is (and this is something I was consistently saying before the Giants' went on the run at the end of 2007) that the Giants have had enough talent to be a dominant, top three team every single year. They have been hampered partly by injury, but mostly by playing stupidly. More times than not, it matters not how the Giants' opponent plays nor who the Giants' opponent is. All that matters is how the Giants play. (Admittedly, how an opponent played would impact how the Giants played, but I am talking about something much more than that. The Giants when rolling can beat anyone and when the Giants are in a funk they'll lose to anyone, regardless of how the other team plays).

The thing to get upset about is not that this bad team is going on another run. It is why this pretty damned good team manages to screw up so many games each year.

85
by Jim Z. (not verified) :: Mon, 01/16/2012 - 5:51pm

The Giants have less talent on their roster, injury or not, than the Eagles or Cowboys in their own division.

The Giants are a less talented team, on paper, even without injuries, than most of the teams in the NFL.

They have a few good pass rushers, a solid QB, two good WRs, and that's basically it.

89
by Dales :: Mon, 01/16/2012 - 5:57pm

You be trolling.

Eli: top 5 QB
WRs: top 3 set. Top 5 at worse
DL: best in NFL
RBs: serviceable. No worse than middle of the NFL.
OL: trending down, but not horrible
DBs: no worse than middle of the pack
LBs: terrible but better than the Eagles
K,P: better than average

Easily a top-10 talent team. And if focusing only on starters, probably top 5.

107
by Jim Z. (not verified) :: Mon, 01/16/2012 - 6:47pm

Eli is a top-10, but not a top-5 QB. I'd take Rivers, Brees, Brady, Rodgers, Romo, Roethlisberger, and even Stafford over Eli in a heartbeat.

WRs are a top-5 unit as a whole, sure.

DL is hardly the "best" in the NFL. In adjusted sack rate, they weren't even in the top-5. They disappear in some games. I watched several Giants games this year in which the pass rushers weren't even a factor. Pierre Paul and Umenyiora are good pass rushers that are liablities in the run game. Tuck is an elite all-around LDE that battles injuries. The DTs are JAGs. That's it. It's certainly a top-5 unit, but probably not the best in the NFL.

RBs are mediocre. They were last in the NFL in rushing this year. Some of that is OL, but the RBs are probably below-average at this point.

OL is relying on two fossils at OT (Deihl and McKenzie), a mediocre free agent C (Baas), and some spare parts at OG. The two young players (Beatty and Petrus) are underhwleming. It's a bottom-10 unit and will get worse unless they draft quality replacements.

The DBs are terrible outside of Webster and Phillips, who are above-average but not elite.

The LBs are terrible.

The Special Teams are middle of the pack.

If you think this is remotely a top-5 team you are delusional. I'd take many rosters over this one. I could do a positional breakdown of the Giants vs many other teams and end up giving the edge to the other team.

118
by Dales :: Mon, 01/16/2012 - 7:33pm

I see what the disconnect here really is.

You have bought so much into the concept of stats as a true distillation of what's happening that when the model breaks down, you give nearly all of the credence to the model and think what everyone's eyes could see must be wrong.

That's fine.

120
by Anonymous one (not verified) :: Mon, 01/16/2012 - 7:42pm

"Pierre Paul and Umenyiora are good pass rushers that are liablities in the run game."

Osi may or may not be a liability in the run game. JPP is demonstrably not, as evidenced by his defensive metrics. I refer you to an obscure site called "Football Outsiders" for evidence (see, e.g., defeats if you don't like solo tackles as a stat). Maybe you are mixing up JPP with J. Babin? And calling JPP and Osi "good pass rushers" is a bit silly as well. They are both among the 10-15 best pass rushers in the league. It would be interesting to see the Giants adjusted sack rate for games where Osi was playing versus those where he (or he and Tuck) were injured.

146
by Mark S. (not verified) :: Tue, 01/17/2012 - 11:32am

"Eli is a top-10, but not a top-5 QB. I'd take Rivers, Brees, Brady, Rodgers, Romo, Roethlisberger, and even Stafford over Eli in a heartbeat."

You are badly underrating Eli. He plays in an outdoor cold weather stadium, behind a poor offensive line (subpar last year and atrocious this year), and has gotten zero help from the running game the last two seasons (worst in the league in 2011!). He's now done it with several different groups of young/inexperienced receivers and TEs, and he absolutely 100% carried the Giants this year while bradshaw was out and the defense was riddled with injuries.

Yes, I am a Giants fan, but I think it's pretty objective at this point - Eli is really, really good.

148
by JIPanick :: Tue, 01/17/2012 - 11:39am

"Not better than Romo, Roethlisberger, or Rivers" is hardly arguing Eli isn't really, really good.

The simple fact of the matter is that, while Eli is a pretty good QB, only the injuries to Schuab and Manning the Greater made Eli a top quartile QB this year. Top-10 but not Top-5 is right on target.

152
by ChrisFromNJ :: Tue, 01/17/2012 - 12:03pm

Roethlisberger, Romo, and Rivers are all quarterbacks that were better than Eli in years past, but worse than him this year- Rivers especially. This year, the only quarterbacks I'd take ahead of Eli would be Rodgers, Brees, and Brady. A healthy Peyton would of course be better as well (though who knows if we'll ever see a healthy Peyton again), and several others (Schaub, Ryan, and Roethlisberger) are in the same echelon.

159
by Mark S. (not verified) :: Tue, 01/17/2012 - 12:37pm

Well, I would argue that he is better than those guys (well, not a healthy Manning the Greater), especially this year.

The Giants had the fourth rated (by DVOA) passing offense this year - behind only GB, NO, and NE. That was with a very weak offensive line (both tackles and the center were aparticularly abominable all year), little to no help from the 20th best run game in the league (FYI, GB was ranked #6, NO was ranked #2, and NE was ranked #4 in DVOA rushing), a serviceable (at best) TE, and 2 very good wideouts who have some real flaws (Nicks: health, Cruz: inexperience/"rookie" mistakes. His third wideout missed the bulk of the season with injuries, and after that there is a huge dropoff on the depth chart.

Additionally, I would throw in arguments that he plays in tougher conditions (outdoors, cold weather) than Brees, Stafford, Romo, Rivers, and Schaub.

Personally, and all told, i would have him fourth behind the Big Three of of Brady, Rodgers, and Brees. But the "simple fact of the matter" is this: Eli was effing awesome this year.

155
by Pat (filler) (not verified) :: Tue, 01/17/2012 - 12:12pm

Yeah, Romo, Roethlisberger over Manning "in a heartbeat" is just silly. Those three have been dancing around each other in DYAR for years now.

Manning, 2008-2011: 1081 DYAR/year
Romo, 2008-2011: 1123 DYAR/year
Roethlisberger, 2008-2011: 1002 DYAR/year

They're basically at the same tier. Below the "Brady/Rodgers/Brees" tier, and just behind Rivers. Including Stafford after effectively one year is just crazy-land. Heck Matt Ryan's got a better resume than Stafford.

You did forget one other thing that Eli is very, very good at. Staying healthy. Manning has not missed a game yet since he started full-time. Let me say that again. Manning has not missed a game yet since he started full-time. With Roethlisberger and Romo, you don't get them for a year. You get them for like, 12-13 games, and then either get a hobbled version of them or Jon Kitna or Byron Leftwich for the rest. With Manning you get him a full year, and that's a big difference. Steelers with a healthy Roethlisberger probably are still playing.

179
by Guest789 :: Tue, 01/17/2012 - 6:06pm

I think you have two distinct tiers.
Super-Elite: Rodgers, Brady, Brees, pre-injury-Peyton
Elite: Romo, Rothlisburger, Eli, Stafford, Schaub, maybe Cutler

Ranking within those tiers is entirely subjective and dependent on the person doing the ranking.

-----

“Treat a man as he is, and he will remain as he is. Treat a man as he could be, and he will become what he should be.”

181
by tuluse :: Tue, 01/17/2012 - 6:20pm

Michael Vick and possibly Cam Newton deserve to be in the 2nd tier.

Also, I'm pleasantly puzzled that non-Bears fans continue to think Cutler is elite or near elite. I think he is but that's because I watch him get battered like fried chicken every week. If one was just casually checking stats they'd probably think to themselves that he isn't producing at a high level any more.

183
by Guest789 :: Tue, 01/17/2012 - 6:24pm

Definitely could make a case for Vick and Newton. Their omission was a function of me listing off the top of my head.
I think we all saw Cutler's value during the Hanie disaster.

-----

“Treat a man as he is, and he will remain as he is. Treat a man as he could be, and he will become what he should be.”

202
by Mark S. (not verified) :: Wed, 01/18/2012 - 5:34pm

IMO, Cutler is the most underrated player in the league. To my eyes, he has more to overcome than any other QB: bad O-line, poor WR corps, cold climate/bad weather stadium, and despite all that he's pretty effective. I would certainly have him in the second tier of QBs.

51
by Aaron Brooks Go... :: Mon, 01/16/2012 - 4:20pm

1957 Lions. +20 in the regular season (were negative through their first 10 games). Beat the (-4) San Francisco 49ers in the divisional game, before destroying +93 Cleveland in the title game.

That team tended to win close and lose big. It happens.

55
by M :: Mon, 01/16/2012 - 4:32pm

That was the Lions last gasp with their core from the great 50's squads. That was not a common occurence every single year, though.

1951 7 4 1 +77
1952 9 3 0 +152
1953 10 2 0 +66 (won close, lost games by 12 & 13 to LA)
1954 9 2 1 +148
1955 3 9 0 -45 (WTF????)
1956 9 3 0 +112
1957 8 4 0 +20

Granted, the Giants have been pretty good in the Eli Manning era. But they seem to have a stinker game (or several) every year where they get utterly destroyed, and their point differential is usually even weaker than their record in most years.

Funny anecdote - in Tom Coughlin's last 3 seasons in Jacksonville, they had a losing record and positive point differential. In circumstantially related news, Eli Manning has a reputation for being a terrific 4th Qtr QB.

69
by Independent George :: Mon, 01/16/2012 - 4:54pm

The 2011 Giants should be considered a travesty of a Super Bowl contender if you care at all about regular season statistics, which is why we are even here at FootballOutsiders.com, right?

I care about statistics insofar as they illuminate things which are not easy to understand. I care about statistics because I want to understand what it is I'm seeing, and what it is I'm not seeing. What I love about FO is that it combines statistical analysis with observation, and attempts to combine the two. Your mistake is that you're looking at the Giants mediocre stats from the regular season, and assuming that represents a static measurement of reality, while ignoring the possibility of random variation within those numbers.

Statistical analysis tells me there's no such thing as 'Winner Juice'. It does not mean that statistical models capture all variables, or that individual performance does not change over time.

84
by Jim Z. (not verified) :: Mon, 01/16/2012 - 5:50pm

So what you're saying is that we can ignore statistical measurements during the regular season, throw them out the window, and construct an entirely new reality during the postseason, despite not being able to do this in roughly 95% of most NFL seasons?

And do all this for one particular team, and that one particular team (or set of players and coaches) only? (The 2007/2011 Giants)

136
by armchair journe... :: Tue, 01/17/2012 - 9:47am

hey uh.. whatever happened to those 2008 Giants?

or those 2011 Packers?

your vitriol would carry more than a feather's weight if these "fluke" teams regressed to their apparent mediocrity.

//AJMQB

31
by Snack Flag (not verified) :: Mon, 01/16/2012 - 3:18pm

I don't know how you can look at the Giants and call them a "middling team" - and I'm a Patriots fan. The Giants, more than most other teams, have the potential to be excellent in any given game due to their upper echelon players. As a QB, Eli can go off and have a monster game (though he is prone to some stinkers here and there). He's not his brother or Brady, but he can certainly put points on the board. They have a deep WR corps - Nicks is a top 10 receiver in the mold of Michael Irvin to me (maybe higher if he could stay healthy) and Cruz and Manningham are excellent secondary threats. Most importantly, JPP, Tuck, and Osi combine to create probably the most disruptive line in the game. Every year, practically 2/3s of the league is in search for an effective pass rusher in the draft or FA. The Giants have 3 of them.

The great regular season teams are often deep and consistent, but those aren't requirements for success in the playoffs. I hate to say it because it's such a vague term, but you need playmakers, guys who can turn a single game in one direction or another. The Giants have plenty of them. If anything, this is a validation of their performance in the regular season because they trudged through injuries and a really difficult schedule to win the NFC East. They took GB down to the wire in their first meeting. This Giants team is as worthy as anyone to win the Super Bowl this year, and the playoffs are much more exciting because of it.

35
by Jim Z. (not verified) :: Mon, 01/16/2012 - 3:37pm

The 2011 Giants have, on offense:

- An above-average QB who is just outside or just inside the top-10 QBs, depending on who you talk to.
- Two stellar WRs who are nevertheless inconsistent and prone to drops
- No running game
- Nothing else on offense outside of the two WRs
- An old, ineffective offensive line

On defense, they have:

- A great, but not dominant, defensive line with some nice pass-rushing DEs such as Umenyiora and Pierre-Paul, an excellent, but often injured, LDE in Tuck, and JAGs at the DT spot.
- Zero above-average linebackers
- Two above-average DBs in the secondary - Webster and Phillips - both of whom are not elite.

So to recap:

- They have good, not great offense with a good, but not great, passing attack and a nonexistant running game, and an offensive line that is a liablity
- They have mediocre defense with a good, but not great pass rush, and liabilities at every other position except for one CB spot (Webster) and one S spot (Phillips)

I watched this team play against the Eagles and Redskins late in the season and watched as they were dominated on both lines of scrimmage. I watched this team play against the Saints and Seahawks late in the season and watched their defense give up TDs as if they were a sieve. This is not a worthy Super Bowl contender.

39
by M :: Mon, 01/16/2012 - 3:58pm

Jim - I share your disgust regarding what the playoffs have turned into. I don't mind the 2010 Packers or 2005 Steelers winning it all - those teams fundamentally were probably top-5 squads those years who via bad luck & weird seeding managed to be #6 seeds in their conference. By comparison, teams like the Giants in 2007 & 2011 as well as Arizona in 2008 defy logic, at least from a statistical perspective.

Historically, we have had other instances where teams have proverbially come out of nowhere to go far in the playoffs. The Houston Oilers in 1978-79,1980 Oakland Raiders, and 1987 MN Vikings come to mind. FYI - I don't count the 1985 Patriots as a complete fluke, as they were arguably the 3rd best team in the AFC that year.

The simple fact is that with so few games in a season, there are going to be fluky seasons here and there - especially in an era where it is hard to build a dynasty even with an all-time great QB. I'm beginning to come to the realization that with the expanded playoffs post-1978 as well as the liberalization of passing rules and increased player movement between teams that maybe the dynasties and dominant teams we used to see in the past will no longer come to define the NFL each year.

It used to be that conference championship - especially in the NFC - was legitimately between two teams that were the best in the conference. We didn't watch the playoffs because the outcomes were virtually random, we watched the playoffs to watch great football - blocking, tackling, powerful running, etc. Now the entertainment of football is based upon uncertainty in the outcomes as well as seeing a few players put a team on their back for a few games. The greatness we see in the playoffs is not team-oriented any more, it is player-oriented. As much as it pissed me off that a team that earlier lost a game 47-7 was in the Super Bowl (2008 Cardinals), it was a joy to watch Kurt Warner and Larry Fitzgerald perform out of their minds for 4 straight games.

Yes, I used to watch football to see dominant teams. Now I watch to see special players take over a game, and try to appreciate that aspect more.

132
by Tim R :: Tue, 01/17/2012 - 5:21am

The 2008 Cardinals were really weird. At the start of the season they were a legitamately good team, top 10 DVOA, it was only really after the Giants loss that knocked them out of home field contention but had the division locked up that they really sucked. I'm not sure if this was because they actually weren't very good (their playoff permormance suggests otherwise) or that having nothing to play for made them lose a bit of focus and the difference between being good and bad in the NFL isn't that much.

143
by Eddo :: Tue, 01/17/2012 - 11:14am

Another possible factor, related to the "losing focus" theory, could be in their gameplans. I recall, in the playoffs that year, Whisenhunt and company played very aggressively, increasing their chances of winning, but also of losing badly. I thought it was a good strategy (high-variance employed by an underdog), and I belive some FO writers expressed the same feelings.

It could be, that once the Cardinals had no shot at homefield advantage, the staff decided to play things closer to the vest, not giving other teams tape of their aggressive nature.

I don't really know; I didn't see enough of the Cardinals that year to say for sure, but it seems plausible to me.

41
by Snack Flag (not verified) :: Mon, 01/16/2012 - 4:02pm

So they're mediocre solely due to your subjective thoughts about their talent? DVOA has them as the 7th best offense in the league, and Eli threw for almost 5000 yards this year. In any given game, they can put up a lot of points. Their defense has been battered by injuries all year; they were down to starting their 4th corner at some points in the season. They're not the greatest team ever, but they can certainly beat any team in the league due to their strengths in the passing game.

I watched Green Bay get beaten by Kansas City. I watched Green Bay go down to the wire in the regular season with the same Giants team that you're saying isn't a Super Bowl contender. I don't see how GB is any more worthy, especially when they had one of the worst defenses in the league.

I saw a 10-6 Green Bay team last year get healthy at the right time and make a run to win the Super Bowl. What was so different about that team vs. this Giants team? They didn't even win their division (like the Giants did) and they won only one more game.

Like I said, I don't even like the Giants, but you at least have to respect who they are. I'm not sure why people keep acting like it's some mystery they win in the playoffs. Eli can throw the ball, their receivers can catch it, and the d-line can get in the face of opposing QB. Add excellent coaching to that mix and you have a Super Bowl contender.

46
by M :: Mon, 01/16/2012 - 4:12pm

Bill Barnwell made a similar 2010 GB vs. 2011 NYG comparison. My point is this - I believe this NYG team has a legitimate chance to win the SB given the 4 remaining flawed contenders. This would be the first SB participant ever to have a negative point differential, although 2008 AZ was only +1.

They are playing better football than anyone over the past month - but it violates my sensibilities that the champion could have a negative scoring differential in the regular season.

If fully healthy and engaged in all 16 games, I would put the Giants' roster as one of the best in all of football. That said, they often seem to quit in certain games and get their asses kicked. IMO, dominant teams don't get blown out. That's one reason why I'll never view the 1976 Oakland Raiders as an all-time great team.

59
by RickD :: Mon, 01/16/2012 - 4:38pm


They are playing better football than anyone over the past month...

How quickly we forget. This game was less than a month ago.

http://scores.espn.go.com/nfl/boxscore?gameId=311218019

80
by M :: Mon, 01/16/2012 - 5:37pm

Good point - I forgot that game is within the last 30 days. That said, anyone wanna bet that even INCLUDING that game the Giants may have the best DVOA of any team over the last 5 weeks?

What I really meant is the last 4 weeks of games.

60
by Derek (not verified) :: Mon, 01/16/2012 - 4:39pm

In a season with no truly dominant/great/elite teams why is everyone in such an uproar that a mild underdog who has shown a great deal of variance between playing really well and playing really poorly managed to run off a string of great performances that by any turn of logic they were perfectly CAPABLE of doing?

All you self-proclaimed statisticians seem to have developed a strange abhorrence to anything other than the most probable outcome. Which is ludicrous in a game which has a 16 game season. There is NOTHING aside from small sample size in football.

An excellent example has been that DVOA has been more predictive in the playoffs than weighted DVOA. There are SO many variables that second half performance doesn't say more than first half. And for teams with such a high variance in performance from week to week, a run of four poor games shouldn't engender such confident opinions of a team's ineptitude.

If your favorite baseball team had a four game stretch where they stunk you'd be laughed out of the conversation for expressing such shock and dismay that they when turned it back around.

It's frankly depressing to see what should be a conversation among rational minds wailing about such an exciting weekend of football or complaining about an era which doesn't seem to match the convention the statistics you're using.

This should be EXCITING. It means you haven't got it figured out yet. You haven't figured out how to account for teams with certain play-styles, where injury and individual performance haven't been properly calibrated, where small sample size and randomness still play a large rule. This isn't baseball where the undiscovered country is in calibrating existing measures or fielding results... this means there's a lot of exciting, innovative, work to be done. There are greater obstacles to overcome in football because it's such a dense sport with so little data.

But instead you're all whining because a good team played great for a few weeks in a row. Get over yourselves.

65
by RickD :: Mon, 01/16/2012 - 4:47pm

Don't judge statisticians by Jim Z's comments. (I'm not really a statistician, but I could certainly teach undergrad stats.)

I have no idea what Jim Z. is talking about, but if he thinks his opinion about the Giants has anything to do with statistical training, I would have to beg to differ. As you say, there are no truly elite teams this season. We have three with elite offenses and shitty defenses, and a couple with great defenses and meh offenses. The closest thing to a complete team we had this season was the Texans, and they were done in by injuries.

The Giants are as good as anybody. Or at least, they are as deserving as anybody. They won a division title in one of the toughest divisions in the NFL (where else would the Eagles have finished third?) And sure, they had a very poor stretch in the middle of the season. As others have said elsewhere, that's due to injuries.

The Giants have an excellent pass rush, a competent running game, a good QB, and with the development of Victor Cruz, they have two WRs (along with Nicks) that are as good as any other pair of top WRs in the NFL.

And it's not like they beat the Packers in some kind of fluky fashion. That game wasn't that close. Prior to that, they shut out the Falcons' offense. At this point, arguing that the Giants had no right to be in the tourney is pathetic.

91
by M :: Mon, 01/16/2012 - 6:02pm

There's a certain elegance if at the end of the season the champion is a team whose overall body of work stacks up as one of the best teams over an entire season. That's a personal bias - one that I freely admit. The term "statistics" should not be confused with statistical theory. I have no idea whether Jim Z is a true statistician or not, but in my OPINION I agree with his fundamental premise.

However, I do not think that the Giants' performance this year violates any principles of statistical theory. In fact, I believe that one season was once simulated 10,000 times over on PFR (back when they still did blog posts). In one of those simulations the WORST team (based on the average performance over the 10,000 runs) actually won the Super Bowl. Granted, I'm oversimplifying the results, but the point is that the only violation the 2011 Giants are guilty of is violating my preconception that a Super Bowl winner should be one of the league's best teams in any given year...over the entirety of the regular season and playoffs.

Again, this is a personal bias which I share with Jim Z. But this occurence does not violate statistical theory, and indeed probably supports it.

95
by Dales :: Mon, 01/16/2012 - 6:10pm

What you and Jim Z are missing on is that the disconnect is not with the Giants' statistical record and where they are, but with how good the Giants really are with their statistical record. And it really is the interesting disconnect, but gets missed when you try to talk about how bad they are.

Anyone who has watched them over the past month would be hard pressed to call them a bad team. Anyone who has watched them over the past decade would say that this year is right in line with the maddeningly inconsistent performance of the team.

This is a very good, very dangerous (to use Coughlin's phrase) team that simply is less consistent than they should be. It would be great to figure out why, but it is sophomoric to say they are terrible.

103
by M :: Mon, 01/16/2012 - 6:35pm

I'm having a tough time recalling my own arguments on the matter, but I do not believe I've said the Giants are a bad team. I also do not think they are a good team - for the entirety of the season. Right now, however, they are a REALLY good team, and think they have as good of a chance as anyone left to win it all.

My disappointment lies only in that I like looking back at end-of-season stats for all sports, and if I see 2011 NYG as Super Bowl champions with a negative point differential, it violates my sensibility. Is that rational? No, not really. But it's still my opinion, damn it.

For the record, I live in MN, and watched the 1987 MN Twins voraciously as a child. I was thrilled as hell to see them win the World Series that year. However, do I think that team was one of the best teams in 1987? Hell no - I believe there were at least 7 teams that were much better. I actually treasure the 1991 Twins more, because that was actually a good baseball team - not just a fluke with 3 weeks of good play.

Here's a more realistic question: If the Giants manage to win the Super Bowl with 2 more dominating victories (10+ point margins), are they an all-time best team?
From what I can tell, the last team to accomplish that is the 1996 GB Packers - who truly DID have a dominating regular season - in a year where almost all of their wide receivers were injured for a couple of games.

110
by Will Allen :: Mon, 01/16/2012 - 6:56pm

You somewhat mischaracterize the '87 Twins. They were one of the best team in baseball, IF you only considered the first two starters in the rotation. When either the future Cy Young winner, Viola, or the future HOFer, Blyleven, were starting, they were really, really, good. Where they were lucky was being in a division where only having to major league starters could still result in a divisional title. Once they got into the post season, where getting by with two good starters becomes much more viable, they were formidable. I was in college, and went down to K.C. for the last games of the regular season, just for the hell of it. K.C. basball fans can be very knowledgeable, so over the course of those two games, I talked to some of them about how they saw the Twins prospects against either the Blue Jays or Tigers, then locked in a tight race on the last weekend. I told people that the Twins would be outmatched against the Blue Jays, but they would whip an aging Tigers team, because the third starter would not matter at all. They did, and then caught somewhat of a break to face a Cardinals team with injuries, but the Cards would have had a tough time in any case beting the Twins in the Metrodome.

The '87 Twins caught some breaks, but as long as Viola or Blyleven were starting, they were very good.

153
by M :: Tue, 01/17/2012 - 12:04pm

Also, don't forget that in odd years in that era the AL West team would have home-field advantage in both playoff rounds, whereas in the even years it was the NL West. This post-season scheduling quirk no doubt helped cement the following pattern of WS winners:

1985 KC Royals (AL West)
1986 NY Mets (NL East)
1987 MN Twins (AL West)
1988 LA Dodgers (NL West)
1989 Oakland A's (AL West)
1990 Cincinnati (NL West)
1991 MN Twins (AL West)

The only team to violate the pattern was the 1986 Mets, although they certainly had an adventurous postseason that year.

If memory serves, no team has ever had a home/road split quite like the 1987 Twins. Hence, it makes it easier to understand their success that year. But they were still not a great team that year.

162
by Travis :: Tue, 01/17/2012 - 12:54pm

FWIW, the 1986 Mets were supposed to have home field in both the NLCS and WS. However, an Oilers-Bears game was scheduled in the Astrodome for the Sunday of Game 4 of the NLCS, and the commissioner decided to shift games 1-2-6-7 to Houston to avoid the conflict.

The 1990 Reds would have faced the same problem, but the October Bengals-Oilers game was relocated to Houston, and a December game scheduled for the Astrodome was moved to Riverfront.

169
by Aaron Brooks Go... :: Tue, 01/17/2012 - 2:10pm

The '87 Tigers also lost because, for whatever reason, the Metrodome was a house of horrors for them. No team was happier to see the Twins leave the Diaper behind and start playing outdoors again.

172
by M :: Tue, 01/17/2012 - 3:05pm

Jack Morris was undefeated in the Metrodome before game 1 of that series. Granted, the Twins sucked from 1982-86, which had a lot to do with that fact; still, there was this aura about him that the Twins couldn't beat him when he came back "home".

The 1987 postseason sprung up a period where the Twins more or less had Detroit's number for 5 years even though the teams weren't that far apart talent-wise.

119
by Dales :: Mon, 01/16/2012 - 7:35pm

No way would they be an "all-time best" team then. They would need the regular season dominance as well to get that. It would, however, make them a pretty darn impressive team from a "potential" perspective.

151
by M :: Tue, 01/17/2012 - 11:53am

Granted, I set up a "straw-man" argument, but you easily got the point. Right now the Giants are playing legitimately great football - but they simply haven't been that good for most of the year. Prior to the Jets game I'd describe them as a middling team - a little better than average taking into account their schedule, but definitely not elite.

What would be nice is to gain a decent understanding of how to predict which teams will "go off" once the playoffs start, and which teams will play at their established level based upon the whole season's body of work.

Seriously, I know people thought the Giants could give GB trouble, but did anyone KNOW they would be this good? Would any commenter here have bet $100 that they beat GB by 17 or more? I sense quite a few that would have bet $100 on GB to beat NY by that much prior to Sunday.

154
by PaddyPat :: Tue, 01/17/2012 - 12:09pm

You're mixing arguments here. I, for one, would have readily told you that the Giants had a good chance to outperform their record based on the difficulty of their regular season schedule. I still thought they would lose to Atlanta, but I thought they had a chance, and if they performed well against Atlanta, then sure, I would give them a reasonable chance to beat GB or SF. What's a reasonable chance? Maybe they had a 45 percent chance to beat GB? Football metrics are about probabilities. Betting on probabilities often burns you. Try betting on a 55 percent proposition over and over, like a well-counted Black Jack game. There are times when you lose 8 in a row even though the odds were sound.

175
by M :: Tue, 01/17/2012 - 3:27pm

Not intending on "mixing" arguments, I was only proposing a possible season-ending scenario that illustrates my point: that the NY Giants were not a good team for the duration of the regular season, but that they are NOW. Did anyone see them becoming this good during the playoffs? Seriously, in each of the last 4 games they clearly looked like the better team, and it wasn't a matter of luck in winning each game (unlike 2007, where they looked better than their opponents, but still could have lost to Dallas, GB, or NE).

Furthermore, it seems like it is becoming more and more likely each year that a particular team will "gel" late in the season and go on a run. Some of these teams are actually strong teams statistically that lost too many close games to get a bye in the playoffs. Fitting this category would be the 2005 Steelers and the 2010 Packers. Some of these are teams that while having certain exceptional players who are capable of taking over a game did not look particularly good for most of the season, such as the 2007 Giants or the 2008 Cardinals.

There is one more recent trend that could also be causing a reduction of home-field advantage in the playoffs, and that is the skewing of ticket-purchasing fans toward those who are reasonably well-off. I've heard many players & coaches state that home crowds are much more passive than they used to be, which eliminates a significant advantage for most teams. It could also be that teams travel in much better conditions than they used to, which could also reduce the traditional home-field advantage.

Ultimately, after that last paragraph, maybe it does seem that I'm mixing arguements. Something has definitely changed over the last few years, though - and it doesn't easily lend itself to study via statistical analysis or data mining.

101
by Aaron Brooks Go... :: Mon, 01/16/2012 - 6:34pm

They won a division title in one of the toughest divisions in the NFL (where else would the Eagles have finished third?)

Based on performance, the NFC West.

More seriously, I think they'd also finish 3rd in the NFC North and the AFC North, especially if they got the Good Bears early in the season.

129
by JasonK :: Tue, 01/17/2012 - 12:49am

Technically,the Eagles finished second. Both PHI & DAL had 8 wins,and Philly had the better divisional record.

128
by Joseph :: Tue, 01/17/2012 - 12:19am

"They are playing better football than anyone over the past month."

There's a team down in New Orleans that will call BS on that one.

47
by Jim Z. (not verified) :: Mon, 01/16/2012 - 4:12pm

10-6 Green Bay in 2010 finished the regular season with the #3 overall DVOA and the #2 defensive DVOA. 10-6 Green Bay in 2010 also finished the regular season having given up the 2nd fewest points (15.0 ppg) in the league and having outscored the opposition 388-240, a point differential of +148.

9-7 New York in 2011 finished the regular season with the #12 overall DVOA and the #20 defensive DVOA. 9-7 New York in 2010 also finished the regular season having given up 25.0 ppg, which was 25th in the league. They finished the regular season having been *outscored* by their opponents 394-400 for a point differential of -6.

Despite Green Bay "struggling with injuries" last year, they were, statistically, a dominant team that caught some bad breaks in games that dropped their record to a lower mark than would be expected. New York, this year, was simply a terrible team that won a horrible division almost by default.

50
by Dales :: Mon, 01/16/2012 - 4:16pm

Sure would love to hear your definition of the word "terrible". I don't think it means what you seem to think it means.

77
by Athelas :: Mon, 01/16/2012 - 5:15pm

As you wish.

81
by Dales :: Mon, 01/16/2012 - 5:39pm

Inconceivable!

64
by Snack Flag (not verified) :: Mon, 01/16/2012 - 4:43pm

Yes, GBs point differential was better than the NYGs, but what does that really tell us? That GB beat the crap out of a few, mostly, bad teams while the NYG were stomped on in a couple of games.

Okay - so what? I don't see how that says anything about their viability as championship contenders. It's a really small sample size in the grand scheme of things. If the NYG kept just two of those big losses closer and bettered their point differential, they're somehow a better team? They're still trotting out the same players in the playoffs and future games will not be affected by those past results.

87
by Jim Z. (not verified) :: Mon, 01/16/2012 - 5:54pm

What about the fact that, in the history of the Super Bowl, there has never been a team that played in the game that maintained a negative point differential in the regular season.

Why do you think, over 50+ years, that has been a constant? Maybe point differential is representative of overall quality of play, or quality of the team?

GB did not "beat the crap" out of a few bad teams in 2010. They played well against everyone. NY in 2011 consistently hemorrhaged copious points to nearly everyone they played.

98
by Snack Flag (not verified) :: Mon, 01/16/2012 - 6:24pm

Pro football reference can be helpful:

2010 GB beat:
A 4-12 Buffalo team 34-7
A 6-10 Cowboys team 45-7
A 6-10 Vikings team 31-3
A 6-10 SF team 34-16
And then they demolished a good (10-6) Giants team 45-17 for one of their best wins of the year.
Their other 5 wins we're by a combined 29 points. They had no big losses because, and I'm happy to admit this, they were an excellent team.

This year's Giants lost big to NOLA in the Superdome, had a 5 turnover double digit loss to Seattle, and were beaten twice by double digits against the Redskins. The Seattle loss and the first Redskins loss have just about nothing to do with the current team that's taking the field. Point differential is descriptive, but it won't tell us much more about the Giants than what we already know: sometimes they'll compound mistakes and give away games by more than they should. However, that doesn't take away from the talent that they have on the roster and the fact that they're a difficult matchup for any team in the league due to their strengths.

Please stop beating the drum that this Giants team is "terrible." Football is about exploiting matchups and, as we've seen in the ATL and GB game, the Giants are excellent at that on occasion.

149
by Mark S. (not verified) :: Tue, 01/17/2012 - 11:48am

"So they're mediocre solely due to your subjective thoughts about their talent? DVOA has them as the 7th best offense in the league, and Eli threw for almost 5000 yards this year"

This is the biggest disconnect in Jim's argument. How were they the 7th best offense in the league this year with a terrible line, no running game, the 10th best(or thereabouts) quarterback, and only 2 good, but not great, receiving options? Some of those units/positions must be better than he thinks they are (hint: it's the quarterback).

As for the improvement on defense - where is the g-d mystery? They were ravaged by injuries early and got as healthy as they've been all year in the last few weeks. Your Packers comparison from last year is spot on - I remember last year (after watching them destroy the Giants in Week 16 as they were getting healthy) thinking that was the best team I'd seen all year - they would have won 13 games going away if they'd been healthy.

75
by DGL :: Mon, 01/16/2012 - 5:10pm

I think Jim Z is really a Giants fan who is trying a reverse-FOMBC.

54
by JonFrum :: Mon, 01/16/2012 - 4:31pm

No statistician would ever make your fundamental error.

61
by RickD :: Mon, 01/16/2012 - 4:39pm

Dammit, I wish this comment section had a parent button.

You've intrigued me, but I have little interest trying to trace the correct vertical line back to the post you're responding to.

67
by Jonadan :: Mon, 01/16/2012 - 4:53pm

The original post was JimZ's blanket statement:

The NFL playoffs were fine until 2007 (you could argue maybe 2001).
Again, 95% of Super Bowl winners, and even participants, in the history of the game were legitimate dominant teams in the regular season, by any measure (DVOA or traditional).
The 80s/90s playoffs were filled with legit., deserving winners such as the Cowboys dynasties, the 49ers dynasties, the Packers dynasties, the Broncos dynasties... Even in this past decade, the Steelers were a legit. defensive powerhouse throughout the regular season, as were the Patriots.
I want to see those teams in the Super Bowl, not some middling team that took the regular season off and inexplicably gets into the Super Bowl. It offends me as a statistician and as a football fan.

Seems Jim wants a) defense and b) dynasties (and probably isn't interested in why that doesn't seem to be happening any more).

The "statistician" comment I assume was just pointing out that 4 years (or even 10) is not a great sample size.

The "easy" way to find the parent comment (still a nuisance): put the cursor on the drop-line and scroll back up the page until the line dead-ends. Though I'd assume you figured that out already and just didn't want to put the effort in. ;)

---
"When you absolutely don't know what to do any more, then it's time to panic." - Johann van der Wiel

28
by Otis Taylor89 :: Mon, 01/16/2012 - 3:15pm

Um, did you notice how many injuries the Giants had this year?
Yeah, every team has injuries, but very few this year lost as many key players as the Giants in the middle of the year - and then got several of them healthy. If you are playing with completely green LBs because your experienced LBs are hurt and then the experienced LBs start playing because they are healthy, which team do you think will play better?

131
by pet mack (not verified) :: Tue, 01/17/2012 - 1:36am

You nailed it. The Giants' play is highly correlated with injuries, much more than average DVOA, etc. The spread on Giants games has reflected this: nobody, but nobody expected a GB blowout, despite the G's 9-7 regular season. Any statistic that doesn't include current team roster is missing a huge prior probability.

139
by SFC B (not verified) :: Tue, 01/17/2012 - 10:33am

As a Texans fan, seeing anyone explain another team's "poor" performance as being due to injuries makes me laugh.

145
by Eddo :: Tue, 01/17/2012 - 11:28am

Really? Because injuries certainly explain the Texans lowered quality of play at the end of the year.

With Schaub, they were arguably the best team in football. They were 7-3, their total DVOA was 32.3% (ranked #1), and their offensive DVOA was 24.6% (#4).

After Schaub got hurt, they were 3-3. Their overall DVOA(*)(**) fell to 19.4% (#5), and their offensive DVOA fell to 13.3% (#9).

(*) I don't have Premium access, so I can't say exactly what their DVOA was in the games Schaub didn't play.

(**) I used end-of-regular-season DVOA as "overall", for these purposes.

30
by dmstorm22 :: Mon, 01/16/2012 - 3:17pm

The 2001 New England Patriots were not a dominant team in any way. They scored a total of three offensive touchdowns throughout the playoffs. The 2005 Pittsburgh Steelers were a very good team, but were still a #6 seed, and had some good fortune (Carson Palmer's ACL tear, Nick Harper cutting the wrong way, Vanderjagt s***ting the bed on that kick, Bill Leavy's Super Bowl calls) along the way. The 2008 Cardinals didn't win the Super Bowl, but they came damn close and were worse than the Giants of 2011.

42
by Anonmouse (not verified) :: Mon, 01/16/2012 - 4:02pm

Fact not fiction. It did not sound right to me so I looked it up. 3 offensive TDs (1 per game). Only 2 passing TD's by Brady all playoffs! What a fluke the 2001 Pats were. Tuck rule, punt return and blocked FG return in Pit game, Ty Law pick 6 in SB, what a friggin run that was.

52
by Aaron Brooks Go... :: Mon, 01/16/2012 - 4:24pm

It was slightly more explicable when it was revealed they had cheated.

62
by JonFrum :: Mon, 01/16/2012 - 4:39pm

Good to hear that Football Outsiders has fans in middle school.

68
by Harry (not verified) :: Mon, 01/16/2012 - 4:54pm

Really? Please explain exactly how this "cheating" had any effect during the 2001 playoffs. Did Vinatieri have a special ball?

72
by Independent George :: Mon, 01/16/2012 - 4:58pm

They played versus Oakland, whom the NFL has had a grudge against for 40 years. Duh.

70
by Anonmouse (not verified) :: Mon, 01/16/2012 - 4:54pm

I could understand the "cheating" comment for the 2003 and 2004 seasons but not for the 2001 season. If I remember correctly, it was defensive hand signals the Pats were recording and memorizing. This would explain why they were so good against Steeler blitzes but it doesn't explain the blocked FG return, Troy Browns punt return TDs, and Ty Law's pick 6 in the SB. If the Pats were stealing defensive signals in 2001, it certainly did not help them score on offense.
I also believe that many teams were recording and memorizing defensive hand signals. A few coaches came out and said "everybody does it" right when the Pats got busted. For some reason they all became mum on the topic later. It did probably lead to the communicators on defense which is nice.
Speaking of communicators: Teams used to regularly complain that their communicators would stop working while playing the Patriots. Anybody notice the problems the Broncos were having with their communicators? I'm sure it would have been mentioned if the game was even close.

76
by dmstorm22 :: Mon, 01/16/2012 - 5:11pm

I hate when Spygate gets rehashed, but in his defense, the 2001 AFC Championship Game was one of the games where the NFL seized illegal taped footage from.

113
by Anonymous1 (not verified) :: Mon, 01/16/2012 - 7:24pm

No it wasn't. Goodell specifically said the footage was from 2006 and 2007. The 2001 filmed practice was pure fiction despite everyone in the media wishing otherwise.

121
by dmstorm22 :: Mon, 01/16/2012 - 7:45pm

The taping of the Rams walkthrough was ficition. The taping during the 2001 AFC Title Game against the Steelers was not.

133
by slomojoe (not verified) :: Tue, 01/17/2012 - 8:53am

Even if true, and leaving aside the copious evidence that taping signals was such common practice in the NFL in those years that the commissioner had to issue a reminder in 2007 about the precise locations from which such taping would or would not be allowed, are you actually arguing that taping the AFC title game against Pitt helped the Pats win the same game?

142
by NotJimmy (not verified) :: Tue, 01/17/2012 - 11:10am

For the love of The Fallen Tebow (he bows, eyes closed) - STOP THIS INSANITY NOW!!!!!!!
IT'S OVER!

57
by dmstorm22 :: Mon, 01/16/2012 - 4:33pm

Actually, 1 passing TD by Brady. The other was by Bledsoe in the 2001 Title Game. Brady did get a rushing TD against the Raiders though.

33
by Otis Taylor89 :: Mon, 01/16/2012 - 3:25pm

Um, did you notice how many injuries the Giants had this year?
Yeah, every team has injuries, but very few this year lost as many key players as the Giants in the middle of the year - and then got several of them healthy. If you are playing with completely green LBs because your experienced LBs are hurt and then the experienced LBs start playing because they are healthy, which team do you think will play better?

140
by SFC B (not verified) :: Tue, 01/17/2012 - 10:38am

True. The Texans only got one of their key players back during the season. Their other injured key players got IR'd. They still finished ahead of the Giants in record and DVOA.

Injuries are a crutch.

194
by rich316 :: Wed, 01/18/2012 - 4:59am

Could you explain why you think injuries are a crutch?

38
by Dired :: Mon, 01/16/2012 - 3:57pm

I'm counting the home team being 7-1 in these playoffs so far. All other data aside, that certainly implies some kind of value to playing well in the regular season.

44
by M :: Mon, 01/16/2012 - 4:04pm

One aspect not brought up is tragedy affecting dominant teams. Isn't it possible that the 2005 Colts and the 2011 Packers were sufficiently by the respective deaths of Dungy's son and Philbin's son that they did not bring their A-game in the playoffs?

Granted, Dungy's son died during the regular season (if my memory is correct), but even after a month the emotional wound from coping with a suicide still hurts very, very badly.

We aren't talking about Michael Jordan inhuman hard-ass competitors here - from all that I've heard, Dungy & Philbin actually have feelings.

63
by JonFrum :: Mon, 01/16/2012 - 4:41pm

You think the players care about a coach's son dying?

74
by Dales :: Mon, 01/16/2012 - 5:10pm

Yes

83
by NYMike :: Mon, 01/16/2012 - 5:48pm

You can listen to Aaron Rodgers' radio program from last week on ESPN Milwaukee 540 (it's available on the internet). It's a 45 minute program and they talk about Michael Philbon for the first ten minutes. I don't know how much of a factor it was in the game, but they were clearly effect as people. Not an excuse, but a factor.

94
by Eddo :: Mon, 01/16/2012 - 6:10pm

They certainly care about their coach, and his well-being.

Plus, you have to figure that Philbin wasn't able, both logistically and mentally, to prepare at the same high level he had been all season.

100
by Kurt :: Mon, 01/16/2012 - 6:33pm

This is dumber and lamer than LaughingAtTheIgnorance's posts in the audibles thread, and won't get 1/10th the criticism. OMG THE DATA!!!

186
by GK (not verified) :: Tue, 01/17/2012 - 7:29pm

Bingo.

Jim Z's horrendous postings are being treated as intelligent and rational, whether the responding poster believes Jim to be correct or incorrect, when his rambling deserve nothing of the sort.

I think my favorite part about all of this is DVOA had the Giants as the 12th best team in the NFL regular season (!!), but Jim casts this inconvenient factoid aside because the negative points differential fact fits his argument so much better (REGULAR SEASON MEANINGLESS! INJUSTICE! ROAR!)

200
by commissionerleaf :: Wed, 01/18/2012 - 2:01pm

On any given Sunday...

40
by CincySaint (not verified) :: Mon, 01/16/2012 - 4:00pm

I'm assuming the Saints stayed on top of the offensive ratings due to their good performance against a tough defense.

But I'd rather be the #7 Niners and get to play the Giants at home for the NFCCG.

66
by Jonadan :: Mon, 01/16/2012 - 4:47pm

As an aside, if the league is getting more passing-dependent (which the stats all suggest it is), and passing is higher variance (which conventional wisdom at least suggests is true, that conventional wisdom probably being a large part of the "3 yards and a cloud of dust" meme), wouldn't it seem that logically results might also become higher-variance? I could see this both in weather (bad rain, high winds) or if the QB just has a bad day or receivers unanimously decide to DROP ALL THE THINGS!

Combine looser passing rules/more protections (known) leading to a better passing environment (hypothetical but probable), with the increase in division-winner-strength variance due to fewer divisional games (as of 2002) and it seems like the "hot runs" are easily explained as a phenomenon just looking at team records, although I don't know how to account for DVOA.

---
"When you absolutely don't know what to do any more, then it's time to panic." - Johann van der Wiel

71
by CeeBee (not verified) :: Mon, 01/16/2012 - 4:55pm

Folks here need to calm down about the Giants. I'm a damn JETS fan, and I HATE every single Giants player that's ever lived.

Aaron and company keep reinforcing that their advanced statistics are showing what is LIKELY to happen. Not what will ACTUALLY happen.

The Giants have the most games lost by starters in the entire league this year. Now they're actually healthy and playing like an inconsistent team is capable of playing - lights out some weeks, dud the next.

Can they keep it up? I sure as hell hope not, but let's not get mad at them for bucking what is "supposed" to happen. It's just less PROBABLE.

79
by JIPanick :: Mon, 01/16/2012 - 5:24pm

I'm not annoyed because a team DVOA dislikes is threatening to win the Super Bowl.

I'm annoyed because a bad team is threatening to win the Super Bowl.

I'm not mad at the Giants, I'm just annoyed that they are still alive. I suppose if there was one target for my ire it would be Green Bay for picking the worst possible time to disintegrate.

82
by Dales :: Mon, 01/16/2012 - 5:41pm

Make the case that the Giants are a bad team. Not a mediocre one, not an around-average one, but a bad one.

Would love to hear it.

88
by Jim Z. (not verified) :: Mon, 01/16/2012 - 5:56pm

They were outscored in the regular season. A negative point differential. NEVER in the history of the Super Bowl has a team with a negative point differential even made the game.

That should tell you something right there.

90
by Dales :: Mon, 01/16/2012 - 6:01pm

So, if I am reading you correctly, the definition of terrible you are using is "not of sufficient quality to have made a Super Bowl in previous years".

The real definition of terrible is:

ter·ri·ble/ˈterəbəl/Adjective: Extremely and shockingly or distressingly bad or serious: "a terrible crime".
Causing or likely to cause terror; sinister: "a terrible smile".

I think you need to recalibrate yourself.

93
by CeeBee (not verified) :: Mon, 01/16/2012 - 6:04pm

The Giants are FAR from terrible, come on. Even this site has them 10th - above average offense, and average Defense and Specials.

207
by Rick22 (not verified) :: Sun, 01/22/2012 - 11:58pm

Giants aren't terrible. They are mediocre, at best.
But mediocre teams shouldn't be in the Super Bowl. It's happened before, lamentably, and it's happened again with these Giants.

But this is distressing not because they are a bad team, it's distressing because you have to wonder what the point of the season is if mediocre teams get into the game? Arizona acquitted itself well, and quite possibly these Giants will, too. But they are the worst team, by almost any measure, to make the Super Bowl. Somebody has to be the worst of all the teams that have made it....prior to this year it was probably the Rams.

96
by M :: Mon, 01/16/2012 - 6:15pm

Jim - not sure if you're reading my comments, but I'm annoyed as well, although not to the extent of the 2007 Giants' run. This team is at least playing really good football now, whereas that team barely won every game past the wildcard round.

That said, it was bound to happen some time that a team with a negative point differential would get to the super bowl and probably even win it. It's happened in baseball (1987 MN Twins), and that was back in the days when only 4 teams made the postseason every year.

I was actually really hoping that the Giants would completely wipe out Dallas in that final game such that this didn't happen, because this particular edition of the Giants seemed to have all of the elements of a team that could surprise in the playoffs if they put it all together.

Another team that gelled late and went on a run - the 1988 SF 49ers. They were 6-5 and ended up winning the SB. Their only loss was a meaningless finale where they got blown out by the LA Rams - back in the days when losing your last game didn't imply a devastating loss of momentum. But their regular season results and stats still look pretty good 23 years later, at least.

If NY wins the SB this year, in 30 years some geek like me is going to look back and wonder WTF? when looking at their game-by-game results. Shit happens, I guess.

122
by dmstorm22 :: Mon, 01/16/2012 - 7:48pm

They were outscored by 6 points. That is seven points worse than the 2008 Cardinals, and 27 and 28 points worse than the 2003 Panthers/2007 Giants. It is not like teams with average point differentials haven't made it before.

It wasn't like the Giants had the point differential of the 2004 Rams, or 2010 Seahawks. -6 isn't that big of a deal..

158
by Pat (filler) (not verified) :: Tue, 01/17/2012 - 12:26pm

Oh, for crying out loud. The Cardinals had a +1 point differential. The Giants this year have a -6 point differential.

That's *7 points*. 1 touchdown. I think you're making way too much out of a single touchdown, or heck, two missed field goals. Statistically the difference between those two teams is nonexistent. Especially when the Cardinals played in a complete garbage division (they were the only team with a positive point differential) and the Giants played in a much better one.

164
by Will Allen :: Tue, 01/17/2012 - 12:58pm

Boy, if there is anything I'd like people to reflect upon, when using metrics to discuss player or team performance, even advanced metrics (which I think has been a wondeful advancment in understanding the game), it is that we really fool ourselves if we think these metrics offer the precision to make extremely fine delineations with accuracy.

Pat, have you read the rest of the thread? The discussion of the statistical signifigance of having number seeds, or teams with the best record, eliminated early at a greater rate today than in the past didn't feel right to me, although I I certainly didn't check the numbers or math. Perhaps it is because I just don't see a 3% probability as being terribly unlikely, whereas most people do. Any thoughts?

184
by Pat (filler) (not verified) :: Tue, 01/17/2012 - 6:25pm

I don't really pay attention to people reading too much into a sample size of like, 5 or 6 years. It's completely insignificant. Especially because using seeding as a proxy for team strength is ludicrous, since the divisions are arbitrary. Then add in the fact that overall league parity varies a ton from year to year... you'd need *hundreds of years* to get enough data. This is all just silly blips.

I dunno why people are so surprised about high seeds losing. Aaron's said all year that this year is more compressed than other years, so why is everyone so surprised that you get more upsets in a compressed year?

Anyway, again, the seeding is just silly-land. Last year featured DVOA's #2 PIT vs. #3 GB in the Super Bowl. Green Bay was a wild card because they lost a 3-point game to Chicago. They win that game and they're a normal divisional winner who wins the Super Bowl. Whoop de doo.

170
by Kurt :: Tue, 01/17/2012 - 2:22pm

In fairness to the Cardinals their out-of-division schedule that year was tremdendously strong, with eight games against winning teams (including four division winners), one against a 8-8 team and one against a 7-9 teams.

111
by JIPanick :: Mon, 01/16/2012 - 7:11pm

Negative point differential was basically my case.

126
by ChrisFromNJ :: Mon, 01/16/2012 - 8:27pm

If the Giants won the Super Bowl, they'd be one of the worst teams to win it all... but they're not a "bad" team by any stretch of the imagination. Bad would be the Denver Broncos winning it all. This Giants team is horribly inconsistent and averages out to mediocre, but they're better than the 2007 team that won it all, and they were hit by more injuries than any other contender this year. If every team had suffered the same level of injuries, the Giants would still be underdogs but not by *nearly* this much.

176
by M :: Tue, 01/17/2012 - 3:39pm

Re-reading this comment today, this post is the most rational and succinct summarization of the various points people are bringing up. I only wish that my various points could be elegantly summarized.

I don't like seeing champions "crowned" due to luck, that's all. I don't care if the ultimate cause is random variance, key injuries to an opponent, or officiating screw-ups - I want to see the best team win at least a reasonable amount of the time (50%?).

If it becomes a free-for-all in the playoffs each year, why not restructure the whole damn league? Instead of a regular season and playoffs, have round-robin or have double-elimination like the College World Series. Or maybe use the regular season to "seed" the teams, but let everyone go into the playoffs. You'd only have to add one more week, honestly and eliminate byes in the playoffs. Hell, wouldn't that make more sense than adding regular season games?

177
by M :: Tue, 01/17/2012 - 3:41pm

Besides, in that outcome maybe Peyton Manning WOULD have actually come back to play this year, since everyone would technically have a chance to win it all.

92
by CeeBee (not verified) :: Mon, 01/16/2012 - 6:03pm

The Giants are FAR from terrible, come on. Even this site has them 10th - above average offense, and average Defense and Specials.

97
by MJK :: Mon, 01/16/2012 - 6:23pm

To everyone complaining about the Giants...relax.

To even GET to the SB, they have to beat a very good (and criminally underrated) Niners team. I would argue that the Niners are tied for being the third best team in the league this year (with NO, after GB and BAL). And they stand between the Giants and the SB...at home.

Then, to WIN the SB, the Giants will have to beat either NE or BAL, who are probably the fifth and second best teams in the league this year, respectively (yes, I'm a Pats fan, and I think BAL and SF are better teams than NE this year...).

So the chance of them being a bad team upset winner is relatively small...less than 25% if you think their chance of winning either game is at most a coin toss.

And if they do win...well, they will have done so by beating Atlanta at home, Green Bay and San Francisco on the road, and New England or Baltimore on a neutral field. That will be four wins against excellent teams in tough situations. Which would imply that maybe they're not as bad as we thought they were...

105
by Jim Z. (not verified) :: Mon, 01/16/2012 - 6:36pm

If they manage to do all that, then I will be personally offended that they simply decided to take the regular season off.

They looked *so* bad against the Saints and the Eagles. I mean, wretched. If that same team goes on to beat Atlanta, Green Bay, San Francisco, and Baltimore or New England in succession, then clearly they were either the biggest fluke in the history of the NFL or the worst underachievers in the history of the NFL.

106
by Snack Flag (not verified) :: Mon, 01/16/2012 - 6:40pm

Did you see how bad the Patriots looked against the Steelers? Or the Giants? Or the Bills? Teams have off games. It happens. Especially in a place like the Superdome where the Saints pretty much had their way with every team.

109
by Jim Z. (not verified) :: Mon, 01/16/2012 - 6:51pm

Losing to the Saints - fine.

But how do you explain how putrid they were at home, with the divsion on the line, against the Eagles and the Redskins?

Or how about against the Seahawks? Or barely winning the game against the Cardinals? Or barely skating away with a win against the Dolphins at home?

112
by Snack Flag (not verified) :: Mon, 01/16/2012 - 7:14pm

They also beat the Cowboys twice with their season on the line. And the Jets. And they kept one of the 3 best offenses in the game (the Patriots) to 20 points. There was good that went along with the bad.

They're an uneven team that had ups and downs. They had the most games lost to injury of any team this year, and many of those were focused in one area: the secondary. I don't know where your bizarre hatred for this team stems from, but I think you'll enjoy the rest of these playoffs much more if you just concede the fact that the Giants are a dangerous team who struggled a few more times than other good teams this year. If you're a "statistician" you'd understand that we're working with a small sample size and the Giants would have a much different resume had a few things went a different way.

137
by armchair journe... :: Tue, 01/17/2012 - 10:00am

Wasn't there a recent post that noted the 2011 Giants had the 3rd worst AGL since FO has been tracking it? Only the 1-15 Dolphins and the 0-16 Lions had more injuries. I could be misremembering, of course.

As they say.. Thar's your problem.

115
by MJK :: Mon, 01/16/2012 - 7:29pm

For that matter, the Patriots have yet to beat a team with a winning record. Their defeat of the Broncos puts the Broncos at 9-9 (0.500) on the season. They only PLAYED two teams with winning records (the Steelers and the Giants) and lost to both of them. Badly, to the Steelers (the Giants game was close).

To be fair to the Giants, they played a harder schedule than a team like the Pats.

123
by raorao (not verified) :: Mon, 01/16/2012 - 7:56pm

Totally agree; I'm surprised this hasn't a bigger story this year. The patriots are arguably more of a fraud than the giants right now. No wins against winning teams, and close wins against terrible cowboys, dolphins, and redskins squads.

Obviously, New England can't be considered a fraud solely based on DVOA ratings -- but I think they are being somewhat overvalued by long-drives-with-lots-of-conversions syndrome. This weekend will be interesting.

125
by Anonymous2 (not verified) :: Mon, 01/16/2012 - 8:13pm

Another one with a bizarre definition of terrible?

The Cowboys finished 8-8. No definition of terrible encompasses an 8-8 team.

In fact, they are another team that, had the Pats lost to them, would have counted as a loss to a quality opponent. Since the Pats won that game, the Cowboys are "terrible" instead of quality?

127
by JIPanick :: Mon, 01/16/2012 - 11:56pm

The Cowboys aggregate body of work was average, true, but at the end of the year they turned into Tony Romo and the 23 stooges.

124
by Anonymous2 (not verified) :: Mon, 01/16/2012 - 8:08pm

That's true about the so-called quality wins, but there's a bit of a scheduling fluke involved here, IMO. The Pats played, and won, 7 games against teams that finished exactly .500. If they had lost, it would have counted against them as a "quality loss", but since they won, it ends up a non-quality win opportunity. In fact, the Giants themselves would have been an example of this if the Pats had won.

I have no idea how often this kind of thing happens, I just thought it interesting that there were so many games where the Pats beating the team was the primary factor in them not being a quality team. Obviously you'd have to go back and re-research all the quality win business to see if it still correlated as closely to playoff success, but shouldn't this sort of thing be looked at independently of the results of the game involving the team whose schedule is being analyzed?

130
by PaddyPat :: Tue, 01/17/2012 - 1:05am

I think this is a critical point in assessing the Pats' wins this year. I remember in the preseason looking at the NE schedule and thinking yeesh, this is a tough slate of games this year. The Pats basically "housed" a lot of these teams. They broke the heart and soul of the Jets, who were really a pretty tough team this year. They busted the Cowboys at a critical juncture. They lost a nailbiter to the Giants during a weak stretch when it seemed like they were recalibrating their offense away from the no huddle to help a struggling defense with time of possession issues. They were legitimately bested by the Steelers, I'll give Pittsburgh that. Other wins though--dominant over the Raiders, who looked like a playoff team at that juncture Campbell playing well and Darren McFadden running wild, spanking of San Diego, which had a good offense this year, regardless of Rivers' stats. What more do you want to see from them? The AFC didn't produce many dominant teams this year. It produced a whole slew of tough, middle-of-the-road contenders, many of which choked down the stretch. A lot of them were playing tougher when the Pats saw them. Moreover, the NFC East was a rough and tough division this year. All four teams played hard. That the Giants won the division is as much because the Pats beat the other three teams as it is about anything else. If Philly beat NE instead of New York, Philly would be going to the postseason. etc. etc. I'm tired of hearing about NE as impostors. It seems ridiculous.

134
by someotherrich (not verified) :: Tue, 01/17/2012 - 9:18am

Yes. The complaint comes down to "Patriots lost to Steelers. Losers"

Beating the Giants would not have given them the all critical "win over team with winning record" (or "win over playoff team", either, most likely)

However, losing one game to the Jets would have probably done both.

Alas the Patriots only could play the teams they were assigned.

144
by 'nonymous (not verified) :: Tue, 01/17/2012 - 11:16am

Yeah! Way to go, Steelers! That was an awesome win-- my favorite game of the season. We were solid on offense and amazing on defense. Yeah!

(Pardon my celebration. It's just nice to see the Steelers showing up positively in the comments one week after a bad playoff exit. To all the fans with teams still in it, I hope you enjoy some tough, well-played football. I'm pulling for the Ravens now--sure they're rivals, but there's divisional loyalty for me. I'm also happy for the 49ers and Giants. As a Steelers fan, I'm afraid I'm obligated to cheer against the Patriots; no offense intended. I'm hoping the Ravens can study the Steelers-Pats game and pull off a similar victory.)

160
by MJK :: Tue, 01/17/2012 - 12:39pm

OK, I'll temper the point I made earlier. I've been looking over the win probability graphs from Advanced NFL stats. The Pats' schedule was really weird this year. They played almost every game against "average" teams...teams that finish in the 6-10 to 10-6 range. Most of these teams, they stomped on their opponents...the game was never really in doubt in the fourth quarter.

In just one game (Pittsburgh) were the Patriots outplayed and basically out of it by the 4th quarter.

In four games the game was close enough so that the probable winner flipped at least once in the 4th quarter. These "close games" could have gone either way. The Pats were 2-2 in these games (losing to Buffalo and the Giants, beating Miami and the Cowboys).

In every other game, you could basically switch the game off and go out for a beer at the start of the 4th quarter, unless you really wanted to see Dan Orlovsky heaving desperation passes.

So on that hand, you could argue that the Patriots did the best they could with their schedule.

But on the other hand, you could point out that, unlike a team like Baltimore or San Francisco or even Green Bay (for all the talk of Green Bay's easy schedule), the Patriots only faced one of the elite teams this year (Pittsburgh, Baltimore, SF, Green Bay, New Orleans, Houston, arguably Atlanta or Detroit, Chicago pre Cutler-injury)...and lost badly.

So you have ample evidence that the Pats are good enough to win most of the time against an average NFL team. We have very little data that they are good enough to beat an elite team.

This is relevant, because (as Marty Schottenheimer can attest), what it takes to beat an elite team is often very different than what it takes to beat an average team.

161
by dryheat :: Tue, 01/17/2012 - 12:52pm

I think you're getting bogged down in record as an indicator of teams' strength. The Pats beat the Jets twice and Philly, who were both Top -10 DVOA. But since you appear to be more concerned with the NE offense vs. the remaining teams, the Jets weighted defense was #3, the Dolphins #8, and the Eagles #7. That's 5 wins against very good defenses. Baltimore and San Francisco are #5 and 6 respectively.

Again, the schedule that the Patriots played was the one they were assigned. And if they beat the Giants, Titans, Bengals, etc, those wins wouldn't be impressive either.

163
by Travis :: Tue, 01/17/2012 - 12:56pm

The Patriots beat the Eagles with Vince Young starting at QB. Then again, the Giants lost to the Eagles in the same situation.

166
by dryheat :: Tue, 01/17/2012 - 1:12pm

All true, but I was focusing on the Pats O vs. Defense, since with the remaining teams in it, that's where the concern seems to be.

168
by Travis :: Tue, 01/17/2012 - 1:37pm

Fair enough. FWIW, Asomugha didn't start against the Pats and only played a few snaps at dime CB, and Rodgers-Cromartie was out, leading to Brandon Hughes making his only NFL start and Joselio Hanson covering Wes Welker.

I wouldn't read too much into that one game, but then again I don't think the Patriots' potential problems will happen on offense anyway.

171
by Aaron Brooks Go... :: Tue, 01/17/2012 - 2:32pm

Philly was top-10, but only after a late season surge after their parts gelled. They were 15th at the end of week 13, before that streak. NE played them early.

NE, to date, has played 4 top-10 DVOA teams, including 1 game against a top-3 team. (One of the top-10 being an Eagles team who was certainly not top-10 at the time)
Detroit, by comparison, played 7 games against top-10 teams, including 4 games (3 road) against the #1 and the #3, not including two games against Chicago, who was top-10 when Cutler was injured.

I'm not sure season performance really demonstrates that NE is any better than Detroit. Both certainly were good at smacking the crap out of Denver.

174
by dryheat :: Tue, 01/17/2012 - 3:23pm

Well wait a minute...this entire line of thought started with the premise that the Patriots hadn't beaten a good team all year because the teams they beat all ended the year at .500 or below, regardless of their record or strength when they played. The Jets (twice), Raiders, Cowboys, Broncos, Dolphins (once), were all playing extremely well before the Patriots game.

Now, I don't subscribe to the theory that losing to the Patriots caused all these teams such mental anguish that it affected them for the rest of the season. If the argument was that the Team X played a harder schedule than the Patriots, I'd agree (although there are very few teams I'd call anything more than "very good" out there this year). In addition, they got to face an unusually large number of backup quarterbacks. Last year the Patriots played (I think) one of, if not the hardest schedule in the league and got the #1 seed and we know how much that helped them in the playoffs.

I agree with your penultimate sentance. There's no performance-based proof that the Patriots were better than Detroit, or New Orleans, or Houston. Clearly the reverse isn't true either.

185
by Aaron Brooks Go... :: Tue, 01/17/2012 - 6:26pm

I was addressing the follow-up comment to that premise, which was "But but but DVOA!"

Even using DVOA, the Pat's schedule was underwhelming.

208
by Rick22 (not verified) :: Mon, 01/23/2012 - 12:09am

This is exactly my problem with the Giants.
Why bother playing the regular season if you can be mediocre all year and just turn it on for the playoffs? Why not just open the season and have playoffs?

I don't think the Giants are awful. Just mediocre. But mediocre shouldn't be in the Super Bowl.

The only interesting thing about this Giants team is Eli, to be honest. Well, I do appreciate their ability to make sacks, too. But their defense isn't all that talented overall. Eli has carried this team, and quite improbably when you consider his ability to be absolutely awful, at times.

Hey, they beat 3 good teams to make the Super Bowl. The Atlanta game was a definitive win, from my perspective. Though Atlanta rarely wins playoff games.
Green Bay was interesting primarily from the standpoint that they kept turning the ball over. If not for the turnovers, which is really more opportunistic than talent, the game was very different than the eventual outcome.
Then tonight's game - a rookie flubs 2 punt returns - and that's 10 points, 3 of which decided the game (by the way, I can't stand the overtime rules - make it like college, for God's sake!).
In other words, as another website pointed out (I think it was NFL Stats), 25% of all games are decided by luck. Looks like the Giants are finally getting their fair share of the luck.

138
by Goathead (not verified) :: Tue, 01/17/2012 - 10:04am

People need to move on from the idea that the 2007 Giants were a bad team based on their DVOA, and just got lucky. This was a team that had all of the previous years' picks playing in the playoffs. At the start of the year, they were a lousy team, but as the year went on a bunch of young players stepped up, including Domenic Hixon, Ahmed Bradshaw, Steve Smith, Kevin Boss, Aaron Ross, Kenny Phillips. Justin Tuck became a force that year. Eli improved steadily through the season. The team that went into the playoffs was NOT the same team that was horrendous for the 1st 2.5 games. In all of their playoff games they flat out outplayed the higher ranked opposition, and anyone who really watched them the last month of the season could have seen this coming.

Anyone who doubts this can look at what the same group did the next year, they were one of the top few teams all year, went into the playoffs as the 1 seed. Up until Plaxico "Plaxico'd" himself they looked like the best team in football.

This years team, on the other hand, is very hard to evaluate. In the games they should have won easily they either lost or squeaked out wins (see Wash, Seattle, Rams, Cards, Dolphins), but they beat Dallas twice, the Pats, and now GB. They have trailed in all but 1 game this season. I've never seen a team this inconsistent, but all evidence I've seen points to them having a below average D, a very good passing offense, and no running game. A good team that can pass protect should beat them unless they turn the ball over a bunch and drop a lot of passes.

147
by DRohan :: Tue, 01/17/2012 - 11:34am

Excellent point. I think all too often we think of teams being static over a year's period (or even longer). Improvement or regression are dismissed. It might be helpful to view the weekly DVOA chart for them. Any idea where/how to find it?

141
by jhqnz (not verified) :: Tue, 01/17/2012 - 11:02am

My knowledge of statistics is lacking so I may not be able to justify the Giants but poster Jim Z sounds like an irate cowboys fan. When looking at the Giants you have to see them for who they really are. A great team affected by injuries. When the season started they lost 7 defensive starters. They had to start rookies at LB for the first half of the season and who are we kidding. Statistical analysis does't take into account that they started a 4th string cb and their starting mlb wasn't on the team until the 2nd half of the season. Statistics are useless unless you take into account the most important factors. The simplification of defensive schemes have somewhat stifled the egregious secondary mistakes. Perhaps this points to coaching issues more than anything else. The giants have moved to more man (funny when you consider the secondary personnel plays better in man, except Aaron Ross who gets beat almost every play) schemes and with the return of their healthy and dangerous d-line they are a much tougher team. Especially when you look at the strengths of the teams they have beat. The giants are built to RUSH the passer and in a league where more and more teams are focusing on passing to score this is a critical skill to have. Bad officiating has hurt the giants also. The officials tried to give away the Jets game, as well as both GB games. Some things in life don't need a scientific explanation, just common sense.

156
by Joshua Northey (not verified) :: Tue, 01/17/2012 - 12:16pm

All this hand wrining over the NYG seems absolutely bizzare to me. I just don't understand it.

The teams play a 16 game season! You have ALMOST NO IDEA what the true distribution of ability is after such a small sample.

Calling the NYG a "terrible" team is just ridiculous. The evidence for them like most teams points to them being average (because we have so little data). That an average team with some injuries could put it together and perform well is completely within the realm of normal outcomes in the modern NFL.

In the past there were fewer upsets because the overall quality distribution between the teams was much more lopsided. Now the teams are a lot more closely packed, so the outcomes (which were already pretty random) and now hugely random.

When the NYG play the packers it is like taking a D100 and rolling it for both teams, and lets say we give GB a +25 "talent adjustment" based on their regular season performance.

To freak out when the NYG roll and 86 and GB rolls a 10 and say "How could this happen?" is to just completely misunderstand how random the NFL is.

209
by Rick22 (not verified) :: Mon, 01/23/2012 - 12:19am

No, that's not it at all.
There is some level of parity, which makes for unusual outcomes, AT TIMES.

But if the NFL were purely random, everyone would end up 8-8, or relatively close to 8-8.

Fact is, the NFL has dominant teams and management structures who consistently put together good teams, while other teams occasionally pop onto the radar for a variety of reasons. The ones who pop on have the opportunity to do good things from time to time, but if we judge consistency as part of the value of a great team, then 9-7 should not be written off as 'a great team who suffered because of randomness'.

By the same token you can't just say it's cool because a mediocre teams suddenly got hot at 'the right time'.
If either was true - then as I stated earlier, why play the season? Why not just have playoffs?

For those of you commenting that the Giants' poor play coincided with injuries, I believe much of this was dealt with on FO or other sites. This isn't precisely true. Yes, they did have injuries - and key ones - but so did many other teams. If key injuries were a problem, why did the Eagles, who were hit more deeply by injuries in the second game, beat the Giants? Injuries play a role, but not THE role. It's an excuse, really. Every team in the NFL had injuries...why do the Giants' injuries mean they lose, but being healthier means they win?

No, the real issue is coaching focus.
And if the coaching is suddenly focused, it still comes back to the question of mediocrity. Why should a mediocre team, whose coaching staff suddenly becomes focused, benefit just because they backed into the playoffs?

210
by tuluse :: Mon, 01/23/2012 - 2:22am

"But if the NFL were purely random, everyone would end up 8-8, or relatively close to 8-8."

This isn't remotely true. Go flip a coin 16 times, right down the results. Repeat this 32 times. I think you'll find a fair number come out 12-4 or 13-3.

157
by dk240t :: Tue, 01/17/2012 - 12:22pm

Amazing season for the Texans, but really, what an amazing season of what could have been for the Texans!

Forget all the time Andre Johnson missed, or starting guard Brisiel missed with a broken leg (which he played on for half a game), or the punter going down and having to be replaced by an 80 year old, or the safety Manning breaking a leg and coming back, or Mario on IR, or numerous other injuries: how is this team not the Super Bowl favorite if just Schaub was there? They still were the best team on the field in the divisional matchup, with an overmatched 5th round rookie QB. And in a season with a total scheme change on defense, and no offseason and multiple rookies and FA aquisitions joining the team - how good could this defense be with another year of experience? They have definitely improved as the season has gone on, yet their weighted average is all the way up to 2nd!

195
by rich316 :: Wed, 01/18/2012 - 5:06am

JJ Watt looks like a young Justin Smith. Pure beast mode.

198
by NYMike :: Wed, 01/18/2012 - 11:27am

Watt skipped his senior year. If he's still a Badger, I think Wisconsin plays for the National Championship.

199
by Aaron Brooks Go... :: Wed, 01/18/2012 - 1:58pm

What does Watt provide that beats MSU? MSU matches up well with Wisconsin, and in ways that adding Watt does not solve. MSU beat the Badgers last year, too, with Watt.

I think maybe Wisconsin goes 12-1 instead of 11-2, but they still are getting edged out by the ESPN Conference 3rd-place finisher.

201
by Will Allen :: Wed, 01/18/2012 - 3:24pm

I dunno, maybe he sacks Cousins, so Cousins doesn't throw a tocuhdown pass? You really can't envision how a college team might have won a close game by having a pass rusher with first round talent in the game?

165
by Will Allen :: Tue, 01/17/2012 - 1:09pm

I really, really, really, would have liked to see this team with a healthy Schaub all season. I tend to focus more on the NFC anyways, but there just isn't enough time to watch all the teams I have interest in, especially since I'm not willing to shell out $350 for the Sunday Ticket. By the time I was strongly motivated to move the Texans to the top tier of my viewing preference, Schaub and Leinert had gone down and I knew that a team with that qb situation had a limited horizon, no matter what else happened. The few times I did see them play, however, they were great to watch.