Writers of Pro Football Prospectus 2008

Most Recent FO Features

WayneReg01-1.jpg

» Audibles at the Line: Week 7

Week 7 features big comebacks for Buffalo and Detroit, big routs in Denver and Indianapolis, and big fat cojones for the St. Louis Rams special teams.

10 Jan 2006

Playoff Power Rankings

by Aaron Schatz

For the first time this year, we are continuing with DVOA into the playoffs, so that we can continue to do the FOXSports.com power rankings. Those rankings with extended commentary are found here.

The headline on FOXSports.com is "PATRIOTS INADEQUATE" and states, "While the Pats get respect from their competition, they won't get it from Aaron Schatz, who says that New England's postseason win streak is overhyped."

It needs to be known that I didn't write this. I don't write the headlines, and I don't write the excerpts. This is a monumental misunderstanding of what I wrote in today's Patriots comment. I have emailed FOX and asked them to change the excerpt to something about the Panthers. I also asked that I be allowed to issue a correction. I never, ever, ever said that the Patriots are inadequate or that they could not beat Denver. All I said was that there is no point in talking about a 10-game postseason win streak because the 3 Super Bowls are impressive enough on their own and the win streak doesn't add anything to the accomplishment.

Patriots fans who read this discussion thread and are active on message boards, please post this apology to Patriots fans. I'm going to ask FOX that I be allowed to write headlines and excerpts for my articles from now on, rather than having a web producer write the headlines. I'm really angry about this.

Here's what you should know about the DVOA table below:

  • These numbers are weighted DVOA. That means that Weeks 1-4 are not included, while Weeks 5-10 are somewhat discounted.
  • The exception is the shaded column marked TOTAL DVOA, which represents DVOA with all 18 weeks equal.
  • However, unlike the standard DVOA numbers on this website, these numbers do not include certain portions of Week 17 games outlined in last week's FOXSports.com commentary. (These plays are not accounted for in the TOTAL DVOA column either.)
  • Although only the playoff teams are listed, all 32 teams have been ranked, and the ranks for offense, defense, and special teams represent this. Teams which did not play in the wild card round are treated as if they had a bye week.

The commentary also includes some discussion of similarity scores for both Eli Manning and Chris Simms. I've printed the stats for the similar players below.

One more note: If you are wondering what happened to some of the notations on our discussion threads like number of comments and time last commented, we think these are big reason for our recent slowdowns, so we've removed them for now in an effort to keep the site running as smoothly as possible.

* * * * *

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>

I'm not going to bother to run the whole DVOA explanation; if you are new to the website, you can read about it here. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.


TEAM WEI.
DVOA
LAST
WEEK
W-L TOTAL
DVOA
RANK WEI OFF
DVOA
OFF.
RANK
WEI DEF
DVOA
DEF.
RANK
WEI S.T.
DVOA
S.T.
RANK
1 IND 39.5% 2 14-2 38.1% 1 33.7% 1 -9.9% 9 -4.1% 29
2 DEN 38.1% 1 13-3 33.9% 2 28.5% 2 -8.6% 10 1.1% 13
4 PIT 30.4% 7 12-5 30.3% 3 12.5% 8 -16.8% 4 1.1% 14
5 CAR 29.0% 11 12-5 20.1% 10 0.5% 13 -22.4% 3 6.1% 5
6 SEA 28.5% 4 13-3 29.5% 4 23.4% 5 -3.1% 15 2.0% 12
7 WAS 26.1% 5 11-6 22.6% 8 -0.6% 14 -23.8% 2 2.9% 8
9 CIN 16.0% 6 11-6 27.9% 5 23.8% 4 8.0% 25 0.1% 17
10 NE 15.3% 13 11-6 9.5% 13 14.2% 7 1.5% 21 2.7% 9
11 CHI 13.9% 12 11-5 13.6% 12 -12.5% 24 -27.5% 1 -1.1% 23
12 NYG 13.2% 9 11-6 17.8% 11 1.2% 11 -11.1% 6 0.9% 16
14 JAC 9.4% 10 12-5 19.7% 9 5.3% 9 -3.1% 16 1.0% 15
15 TB 3.8% 15 11-6 5.1% 14 -4.6% 16 -8.4% 11 0.0% 18

The three (non-playoff teams) missing from this table are Kansas City at #3, San Diego at #8, and Minnesota at #13.

Here are the one-game DVOA ratings for the first round of the playoffs. Yes, DVOA believes that Tampa Bay outplayed Washington. (This is true even without the opponent adjustments.)


TEAM TOT OFF DEF ST
CAR 177.7% 34.8% -126.2% 16.7%
NYG -142.6% -104.7% 29.0% -9.0%
PIT 102.4% 54.3% -45.9% 2.2%
CIN -49.3% 0.3% 43.6% -6.1%
JAC -48.8% -48.3% -8.7% -9.3%
NE 66.0% 22.1% -36.6% 7.3%
WAS 0.1% -48.2% -49.5% -1.1%
TB 31.0% -36.6% -61.4% 6.3%

Here are the most similar players to Eli Manning over two seasons (since 1978). I limited the list to players 25 years old or younger, with at least a 700 similarity score in year one and a 750 similarity score in year two. Players are in order by harmonic mean of the two similarity scores.


Name Year 2 Team Age Exp. G COMP ATT PaYD PaTD INT COMP
PCT
YD/
ATT
ruYD SIM
Y
SIM
Y-1
Eli Manning 2005 NYG 24 2 16 294 557 3762 24 17 52.8% 6.75 80 x x
Jay Schroeder 1986 WAS 25 2 16 276 541 4109 22 22 51.0% 7.60 47 881 851
Jim Everett 1987 RAM 24 2 15 221 412 2815 14 18 53.6% 6.83 113 760 883
Ken O'Brien 1985 NYJ 25 2 16 297 488 3888 25 8 60.9% 7.97 58 799 829
Bernie Kosar 1986 CLE 23 2 16 310 531 3854 17 10 58.4% 7.26 19 828 795
Boomer Esiason 1985 CIN 24 2 15 251 431 3443 27 12 58.2% 7.99 79 800 811
Aaron Brooks 2001 NO 25 2 16 312 558 3832 26 22 55.9% 6.87 358 833 747
Steve Deberg 1979 SF 25 2 16 347 578 3652 17 21 60.0% 6.32 10 820 716

Now here are the most similar players to Chris Simms over a single season (since 1978). This one I limited to players with either two or three years of experience, which removed Bernie Kosar and Tim Rattay. I also removed second seasons by the same player, as the list originally had two Neil O'Donnell seasons and two Charlie Batch seasons. By the way, can you guess who shows up on the list? Ah, genetics.


Name Year Team Age Exp. G COMP ATT PaYD PaTD INT COMP
PCT
YD/
ATT
ruYD SIM
Y
Chris Simms 2005 RB 25 3 11 191 313 2035 10 7 61.0% 6.50 31

x

Neil O'Donnell 1992 PIT 26 2 12 185 313 2283 13 9 59.1% 7.29 5 896
Charlie Batch 1999 DET 25 2 11 151 270 1957 13 7 55.9% 7.25 87 887
Phil Simms 1981 NYG 27 3 10 172 316 2031 11 9 54.4% 6.43 42 862
Steve Deberg 1980 SF 26 3 11 186 321 1998 12 17 57.9% 6.22 4 860
Jack Trudeau 1987 IND 25 2 14 179 321 2222 8 8 55.9% 6.93 10 858
Patrick Ramsey 2003 WAS 24 2 11 180 338 2169 14 9 53.3% 6.42 62 858
David Carr 2003 HOU 24 2 12 167 295 2013 9 13 56.6% 6.82 151 856
Drew Brees 2003 SD 24 3 11 205 356 2108 11 15 57.6% 5.92 84 853

1987 players in both tables have their seasons pro-rated for the strike.

Posted by: Aaron Schatz on 10 Jan 2006

183 comments, Last at 18 Jan 2006, 2:24am by Christian X

Comments

1
by Becephalus (not verified) :: Tue, 01/10/2006 - 11:47pm

Wow first time I have had the first post and I have nothing to say. Err um keep up the good work "Phillipe Columbia".

And other than the CAR game I was 3 for 3 vs spread and over with help of trusty DVOA :)

2
by Matthew Furtek (not verified) :: Tue, 01/10/2006 - 11:53pm

This just in... NFL Network's "Sounds of the Game" just showed visual evidence of Sean Taylor hawcking spitting at Pittman.

Sean Taylor you just had the highest punk-DVOA game of the season....

3
by Catfish (not verified) :: Wed, 01/11/2006 - 12:26am

"Also, Carolina is on the east coast, so the Panthers get extra credit."

Oh god, someone is going to take that comment seriously. Gatts, please let us know how many trolls send hate mail because they can't recognize sarcasm.

4
by mikeabbott (not verified) :: Wed, 01/11/2006 - 12:32am

I have been very impressed with DVOA for years,(PS: thank you, thank you, thank you everybody for keeping this site the least trollish of any I care about).
I have just seen a season where a team I watch carefully has changed drastically in ability over the course of a season.
I know you have a system that weights more recent results higher but how has that been tested?
I think you have the data and I'm certain you have the abilty to apply the test.
What I want to know is how well do results age?
how well does DVOA in game X predict DVOA in game X+N[1..]?

5
by David (not verified) :: Wed, 01/11/2006 - 12:38am

how well does DVOA in game X predict DVOA in game X+N[1..]?

I'm not sure of the numbers, but that question is the reason for just about every tweak Aaron & co. make to the formulas that calculate DVOA. The goal is for the stat to be as close to perfectly predictive as thye can get it. So it's fairly close, I'd imagine.

6
by admin :: Wed, 01/11/2006 - 12:47am

No, it isn't... but that's because this is the NFL. :) There's a ton of randomness. It's as close as I know how to get it.

When I change the weighted formula, I'm always trying to make it do a better job of predicting the team's exact DVOA in the next game, so this is, I hope, the best we've got. By the way, if any of you math and stats folks have ideas of the best way to test the best weighted DVOA formula, something perhaps I haven't done, e-mail me directly. (Don't post here, it will bore people I'm sure.)

7
by Just Another Grammar Nazi (not verified) :: Wed, 01/11/2006 - 12:48am

For the first time this year, we are continuing with DVOA into the playoffs...

Yeah, I was really wondering why you didn't continue with DVOA into the playoffs back in week 12. Oh well, better late than never, I always say.

8
by steve (not verified) :: Wed, 01/11/2006 - 1:03am

The goal is for the stat to be as close to perfectly predictive as they can get it. So it’s fairly close, I’d imagine,

Injuries effect DVOA pretty drastically. Carson Palmer going down pretty much eliminated any chance Cin had to overtake Pit.

Say, doncha think NE is a bit underrated, since the Patriots' players have been replaced by robots.

Then, DVOA calculates playing, not not playing, that is, resting at home eating pizza. Shouldn't Seattle, Indy, etc. adjust downward for inactivity (or upward for recovery of injuries to important players?)

9
by William (not verified) :: Wed, 01/11/2006 - 1:06am

I agree that McGinest's sack record is not paid enough attention, but saying the Pats record of 10 straight wins in the postseason is "overhyped" is, I think, not justifiable. Such a record, in a short span, is very impressive--whether they made the playoffs in 2002 or not. If this was a record spanning, say, 20, 30 years, you would have a case.

10
by Rollo (not verified) :: Wed, 01/11/2006 - 1:13am

Astute observation on the Jags. Their schedule next year figures to be considerably harder. However, their offense improved over the year to a top 10 group according to DVOA. Leftwich was having a great season and should get better, and Wilford, Greg Jones, and Barnes can only improve with more seasoning. I agree the biggest improvement the Jags could get would be a stud TE, as Brady offers next to nothing in the passing game, which is where the real room for growth is if Matt Jones seriously improves. With upgrades at RT and TE and much improved play from Jones and Wilford, I think Jacksonville's offense could pass the defense as the team's strength next year.

Defensively, the Jags return most starters. Resigning Wright is somewhat important given the lack of a #2 backup corner with experience on the roster, while Ayodele is more expendable (see Winborn, Jamie). I wouldn't be at all surprised to see the team pick an OLB in the first two rounds.

11
by DJAnyReason (not verified) :: Wed, 01/11/2006 - 1:22am

I have to disagree with #9 - its like when people would cite the Yankees recent World Series win streak, ignoring the small matter of 1997.

If you want relatively simplistic (well, in analysis, not execution) way to find weights for a weighted DVOA, just do a multi-variable linear regression where you have a column for last season, and each other column is the team DVOA for n weeks ago. If we're in week 2, you put 0's in every collumn for n>1. Use as many rows as you have data for.

Excel (or a TI-83) can calculate it in a snap.

12
by Kal (not verified) :: Wed, 01/11/2006 - 1:27am

Exactly. Saying the streak is there is ignoring the fact that the Pats didn't actually go to the playoffs one year. That's a lot worse in my mind than losing a playoff game.

But no one cares.

At least it's not as hyped as that stupid consecutive winning streak last season. Gods that was stupid. It doesn't matter! Sigh.

13
by admin :: Wed, 01/11/2006 - 1:37am

Hey, the 21-game winning streak was pretty cool. Anyway, I've already gotten a ton of hate mail on this one, but the point is that when you have these two things ...

3 Super Bowl titles
10 straight postseason wins

... the first one is amazing but the second one really adds nothing. To win the Super Bowl, you have to have a postseason win streak, because if you lose, you are out. Would the 10 straight postseason wins be impressive if there had been, say, a player's strike after the wild card games one year, cancelling the Super Bowl? Would 10 straight postseason wins be impressive if a team did it with five seasons in between each trip to the postseason?

The 21-game win streak is different because, unlike with the ten-game postseason win streak, there was no place in between where other teams were playing meaningful games but the Patriots stayed home because they weren't good enough.

Oh, and I tried the method in comment 11, but the problem is that it gives you weights that bounce up and down each week, which makes no sense logically. Why would games 9 weeks ago be worth much less than games 8 weeks ago or 10 weeks ago, for example? So instead I've done something similar but manually, with trial-and-error, with the rule that one week's weight always has to be equal to or lower than the following week.

14
by MAW (not verified) :: Wed, 01/11/2006 - 1:40am

I'm curious if the ranking for a non-playoff team was drastically altered by the week 18 action. Maybe for next week you can include all 32 teams, but leave out commentary for the 24 teams that didn't play this weekend.

15
by thad (not verified) :: Wed, 01/11/2006 - 1:41am

re 12
Really?
You didn't think that was amazing?
Juzt out of curiosity, why?

16
by Kal (not verified) :: Wed, 01/11/2006 - 2:24am

What I hated was comparing it in any way to the Dolphins perfect season, which is what so many sports folks seemed to do. Winning all your games in one season - when you have injuries, playoffs, etc - and winning a bunch of games over two seasons are just so monumentally different things. Was it impressive? Kinda. It wasn't as impressive to me as winning back-to-back superbowls, but it was impressive sorta kinda.

What I don't like about it is that it doesn't really do anything about how good a team is, because, well, they are two different teams. To me, it'd be like using DVOA from a prior season in the weighted average for this season.

Honestly, I thought that the 13-0 start of the Colts and Broncos was more impressive - again, because it was done in a single season. My biggest problem with it was the massive amount of hype it got on ESPN, like it was this unbelievable thing.

Ah well. It's almost as bad to me as pointing to a QB's win-loss record as an indicator of their talent. Though that one pisses me off a heck of a lot more.

On the linear regression question - Aaron, while it doesn't make sense from that perspective, does it produce more accurate results?

17
by admin :: Wed, 01/11/2006 - 2:30am

Dunno. Would have to test the 2000-2004 regression on 2005's numbers. Off-season project. I have a feeling that you would need like 50 seasons worth of data to have the regression make sense, a few seasons is just going to have too much noise.

18
by Jeremiah (not verified) :: Wed, 01/11/2006 - 2:32am

How did the Colts and Broncos switch places for weighted DVOA when neither played?

19
by Jeremiah (not verified) :: Wed, 01/11/2006 - 2:33am

Oh, my question was answered on the FoxSports page.

20
by Chris M (not verified) :: Wed, 01/11/2006 - 2:54am

Aaron,

What do you think those similar seasons for Simms and Manning indicate for each player? What can we expect their next seasons to look like?

21
by Walt Pohl (not verified) :: Wed, 01/11/2006 - 2:59am

I wonder one thing about similarity scores: what do they look like for a quarterback that we know turns out to be a success? Peyton Manning or Donovan McNabb, say.

22
by Chris M (not verified) :: Wed, 01/11/2006 - 3:07am

To clarify - you said that we can't use the other QBs' stats to directly project Eli, re: completion percentage. So, where did they improve?

23
by Jon (not verified) :: Wed, 01/11/2006 - 4:19am

I know you were probably exaggerating for effect, but the Packers didn't exactly "lay down like dogs" against the Jets in 2002.

1. The Packers, playing for home-field advantage, were down only 14-10 at halftime before the Jets pulled away to win 42-17.

2. Darren Sharper was injured, greatly weakening the Packers' secondary.

3. The game was at the Meadowlands; the Packers were only 4-4 on the road, but 8-0 at home.

4. Chad Pennington was not the twice-injured Pennington of today, but rather the NFL's best QB according to DPAR despite only starting 12 games.

The Patriots were in control of their playoff destiny going into Week 15, but blew it by losing 24-7 at Tennessee and 30-17 to the Jets.

Otherwise, good article.

24
by Uptight (not verified) :: Wed, 01/11/2006 - 5:37am

#7 - Though the name you use is offensive (at least use a small "n"), your point was definitely LOL.

25
by POUM (not verified) :: Wed, 01/11/2006 - 8:34am

#24 - 'At least use a small "n".'
Your post was also LOL considering JAGN's nazism is of the grammatical variety and the use of capitals in names is one of the principle tenets of grammatical national socialism. If he were to follow your advice and use a small "n" I think he'd be duty bound to crack a cyanide capsule soon after posting.

26
by senser81 (not verified) :: Wed, 01/11/2006 - 9:51am

Interesting that Steve DeBerg resembles both Eli Manning and Chris Simms. I figure that DeBerg played so long and for so many teams, he probably resembles every other QB in NFL history.

27
by admin :: Wed, 01/11/2006 - 10:13am

Given the torrents of vicious, angry hate mail that I've getting from Pats fans, it is nice to see that comment number 23 thinks that I'm a biased Pats homer. You can't make anyone happy, can you?

Believe me, I know the Pats blew their playoff destiny that year. Remember, Football Outsiders exists because I wanted a way to prove that the 2002 Pats were a better team than the local media said they were.

28
by spenceKarl (not verified) :: Wed, 01/11/2006 - 10:13am

What were the weights used for weeks 5-10?

29
by spenceKarl (not verified) :: Wed, 01/11/2006 - 10:17am

I enjoyed your Pats commentary this week. By the tone, it was pretty clear how much angry New England mail you must have received.

30
by admin :: Wed, 01/11/2006 - 10:25am

Well, I found out what the problem is. This is the headline and excerpt on the FOX playoffs page:

PATRIOTS INADEQUATE
While the Pats get respect from their competition, they won't get it from Aaron Schatz, who says that New England's postseason win streak is overhyped.

I didn't write this.

I don't write the headlines, and I don't write the excerpts. This is such a monumental misunderstanding of what I wrote that I understand why Patriots fans are angry. Except they are not angry at me, they are angry at a FOXSports.com headline writer.

I have emailed FOX and asked them to change the excerpt to something about the Panthers. I also asked that I be allowed to issue a correction. I never, ever, ever said that the Patriots are inadequate or that they could not beat Denver. All I said was that there is no point in talking about a 10-game postseason win streak because the 3 Super Bowls are impressive enough on their own and the win streak doesn't add anything to the accomplishment.

I'm embarassed and very angry. Patriots fans who read this discussion thread and are active on message boards, please post this apology to Patriots fans. I'm going to ask FOX that I be allowed to write headlines and excerpts from now on.

31
by spenceKarl (not verified) :: Wed, 01/11/2006 - 10:32am

That sucks. And they even got a discouraged looking photo of Brady to fuel the fire.

32
by Becephalus (not verified) :: Wed, 01/11/2006 - 11:00am

lol Aaron I understand your dislike of being misrepresented, but I think its kind of humorous. Headline should be: PATS HOMER BASHES FAVORITE CLUB BECAUSE HE MISTAKENLY BELIEVES FOXBOROUGH IS IN OREGON.

:)

I mean seriously who cares about the opinions of people who get their blood worked up over such things. NE will either perform well on Sunday or they won't and nothign you say is going to change that. People identify so much with their teams its scary, they seem to see and percieved slight aaginst them as a personal assault.

33
by A Pats Fan (not verified) :: Wed, 01/11/2006 - 11:01am

Best of luck re: the title, but that's normally an editorial decision IIRC - to get high clickership.

See 'Post, New York' for examples of good headlines :)

But, sorry, that sucks. I'd have hoped NE fans would be smart enough to read the actual article, not just the headline...

34
by admin :: Wed, 01/11/2006 - 11:12am

OK, they changed it. Here's what it now says:

"AFC teams hold the top three spots in the latest power rankings. However, the team nobody wants to face finds itself buried behind a team that lost last week. The Pats get respect from their competition. Aaron Schatz gives the champs their due too, but says not to make too much of their playoff win streak."

Am I wrong to think that it is still a problem to say that the Pats are buried behind a team that lost last week? (The Bengals) People who see this still won't understand that the ratings are stat-based and include the entire season (well, back to Week 5). I don't think that the Bengals would beat the Patriots if they played tomorrow, but then again DVOA does not know that Carson Palmer tore his ACL. Should I have them re-write this again?

And seriously, how on earth do you get "Patriots inadequate" out of that comment I wrote? Did anyone else get the impression that I was saying that the Patriots are inadequate? Honestly, I thought it was a throwaway comment. The important comment to me was the one about Carolina having the best game of the year, I thought that would clearly get the headline.

As far as the New York Post, that's not the way I do things. I don't like sensationalist headline writing. I like puns. I'm willing to fight to have my pieces led by intelligent headlines. I started FO to have smarter football talk, not more sensational. Believe me, DVOA is controversial enough as it is, it doesn't need more controversy.

35
by spenceKarl (not verified) :: Wed, 01/11/2006 - 11:28am

It seems that the headline writer doesn't understand what you're trying to do either. The headline makes no effort to point out that the FoxSports Power Rankings are objective. But I'm sure like A Pats Fan said, part of the writer's job is to draw people's attention to the site; And what better way than to paint a picture of you holding a giant DISRESPECT card?

Maybe if they wrote something that says the Patriots statistically aren't getting the respect their competition is giving them. Maybe that would look better. But to link the disrepect thing with your name, implies that the rankings are purely subjective.

36
by B (not verified) :: Wed, 01/11/2006 - 11:32am

Kal #16: The 2004 Steelers had a 15 game single-season winning streak (Well, 14 regular season plus one playoff victory), wouldn't that be more impressive than the Colts or Broncos 13 game mark?

37
by White Rose Duelist (not verified) :: Wed, 01/11/2006 - 11:32am

Aaron's headlines is clearly ranked too low because he doesn't respect the Streak. FOX Sports headline writers is way better than this. Teh pats will loose Sunday because of teh FO msg board curse and it will be all Aron's fault,

38
by Rich Conley (not verified) :: Wed, 01/11/2006 - 11:38am

lol 37

re 35 (SpenKarl)

DVOA is not objective. I love DVOA, but any system like this, you have to pick and choose between certain variables to track. It could be 100% accurate, and it would still be subjective. (Just me nitpicking)

Aaron, this isnt really related, but is there any metric anywhere of how teams do (in general) while playing with a backup quarterback?

I'm curious, because I was having an argument with someone about Ron Mexico, and they brought up the fact that over the last years, ATL wins a lot higher % of games when hes playing. Its my belief that teams in general are worse when a backup QB is in, and it has nothing to do with Mexico being good.

39
by JMM (not verified) :: Wed, 01/11/2006 - 11:44am

RE #9 "Carson Palmer going down pretty much eliminated any chance Cin had to overtake Pit."

Cinci had a 10 point lead over Pitt 2x. There was no need for them to overtake Pitt. See TMQ's discussion. I'd even go further then TQM and ask where Rudi Johnson was from the middle of the 2nd quarter through the rest of the game.

40
by spenceKarl (not verified) :: Wed, 01/11/2006 - 11:53am

RE: 38 (Rich)

Well the reason why I consider it to be objective is because DVOA is based on evidence rather than opinion and emotion. That's all I was saying. That the headline seemed to imply that Aaron's ranking were purely subjective.

41
by admin :: Wed, 01/11/2006 - 11:57am

Rich,

Every time you nit pick like that, it provides fuel to all the people on the Internet and all the trolls who come to these message boards to say that I am biased in favor of one team or hate their team and so forth. When I say "not subjective" I mean that I don't favor one team ahead of another team. If you really think that I designed this system to favor a specific team, than make the accusation. If all you are saying is that I had to make a subjective decision whether to design a system that predicted wins in the current season or a system that predicted wins in the future, you are just flooding my email with hate mail and crap, because the majority of people don't understand the subtle nuance of your statement. If you really like this site and want us to write more articles and do more research instead of spending more time deleting hate mails, stop.

42
by Dan Riley (not verified) :: Wed, 01/11/2006 - 11:59am

Oh, I cannot believe the comedy I just woke up to today...and right in the middle of a playoff week. Aaron running around like one of those scientists in a 50's space invaders movie begging the pitch fork crowd to be reasonble as they trample over him and the camera comes in for a close up of his abacus being crushed under foot. We love FO because it is the nexus between rational and irrational thought (and I'm not just talking football). Those monkeys at Fox and WEEI and ESPN want to draw eyeballs and stir controversy and get people talking. While Aaron wants everybody to cooly admire his calculations and let the numbers guide them to their logical conclusion ("Hmm, Pats Total DVOA 9.5%; Broncos 33.9. Guess I'll just go see Brokeback Mountain Saturday night.")

43
by Scott de B. (not verified) :: Wed, 01/11/2006 - 12:03pm

While agree that a 10-game playoff streask doesn't have any added significance because those wins all come in a row, nevertheless the fact that Brady has a 10-0 record is pretty incredible. He could go 4-7 for the rest of his career and have a better postseason record than Elway. He could go 4-4 and have a better record than Bradshaw. He could go 6-6 and have a better record than Montana.

44
by admin :: Wed, 01/11/2006 - 12:04pm

And while I'm clearly ticked off and at the end of my rope (the only break I've had in the last year was not exactly a vacation, as some of you know):

Are Patriots fans really saying that the Pats' ten-game postseason winning streak is more impressive than the Dallas Cowboys also winning three Super Bowls in four years, but in the middle year going 12-4 and losing the NFC Championship game instead of going 9-7 and missing the playoffs? The Cowboys won their first playoff game in 1996 too, so if 1994 had never happened, they would have ten straight. They get punished for going 12-4? That's just stupid.

45
by Dan Riley (not verified) :: Wed, 01/11/2006 - 12:13pm

Speaking of being rational, in looking over the Broncos season, I'm thinking BB might want to order up a copy of that Ravens game. Any Bronco fans want to explain why that might be a waste of Bill's time.

46
by Rich Conley (not verified) :: Wed, 01/11/2006 - 12:16pm

I agree that the 10 game postseason winning streak is a product of winning the 3 superbowls, and is not by itself really that impressive.

But heres the question... If the Pats were infact better in 2002 than they were in 2001 (and I believe they were).. Had they won that last game and made the playoffs, how would they have done?

47
by Levente, Hungary (not verified) :: Wed, 01/11/2006 - 12:16pm

Aaron,

Please accept my sympathy.
This Fox headline really silly (to put it lightly). The news of the day (and the key point in your article) is definitely not the bit about the Pats' streak, I don't get why they have to stick to that.

48
by senser81 (not verified) :: Wed, 01/11/2006 - 12:19pm

re: post #41

Wow. A bit overdramatic, would you say?

Re: post #38

Agree on both your points. I've heard that same 'Ron Mexico' argument. I think it proves that guys like Kurt Kittner suck...not that guys like Vick are actually elite QBs.

49
by Oswlek (not verified) :: Wed, 01/11/2006 - 12:20pm

Wasn't Washington the team that had terrible luck with fumble recoveries?

If so, what a reversal of fortune. Not only did they recover every offensive and defensive fumble, but they recovered a fumble by a guy who had just recovered a fumble.

That was an amazingly lucky break, btw. Had the lineman just fallen down without dropping the ball it is probable that the Skins get no points there.

50
by Oswlek (not verified) :: Wed, 01/11/2006 - 12:29pm

Rich #46,

The total season numbers may suggest that the 2002 Pats were better than the 2001 team, but they were certainly not at the end of the year.

Even if we ignore the fact that Brady's throwing shoulder was dislocated, their #1 & #3 WR were hurt, the RBs were terrible, the DL/LBs couldn't pressure the QB and the secondary got old and slow overnight. Oh, and they were 31st in the league in ypg and ypc.

Had they made it in they might have won one game (only because it would have been at home) but IMHO, they would have been the worst team in the league to make the playoffs.

51
by ScottR (not verified) :: Wed, 01/11/2006 - 12:30pm

Re 46,50:

And considering Tampa's defense that season (-32.4%) was the most dominant of any team's in the annals of DVOA (1998-2005). And that their total DVOA was second best in the annals ('99 Rams 42.2%), I don't think the Bucs would have lost to any combination of AFC teams in the playoffs.

52
by B (not verified) :: Wed, 01/11/2006 - 12:31pm

46: In order to make the playoffs, the Patriots had to beat the Jets in week 16. Assuming they do that, the most likely occurance would be a victory over the Colts at home and a loss to the Raiders at home (Revenge for the tuck game, I guess). This gives Brady an 11-1 record in the playoffs, and is that any less impressive than 10-0?

53
by kleph (not verified) :: Wed, 01/11/2006 - 12:31pm

while i understand the frustration the errant headline is causing (as well as the strain on the inbox) there is a positive aspect to this. it certainly will get more folks looking in at what the fuss is about and, i believe, more than a few will see what is going on here and stick around. we have already seen a number of folks show up this season, initially highly critical of the assessment of their team and later understanding what the FO approach is trying to accomplish. the trick is surviving the tsunami of trolls until that happens.

54
by Andrew (not verified) :: Wed, 01/11/2006 - 12:33pm

Aaron:

Winning 10 playoff games in a row over a 5 year span is one of the most amazing things ever done in football. Even with missing the playoffs one year, winning 10 postseason games in a row is just incredible.

The longest postseason streak any team ever had previous to this was 7 wins in a row by five teams - Steelers 1974-76, Packers 1965-67, 49ers 1988-90, Cowboys 1992-94, Broncos 1997-98.

If the Patriots simply win one more game, they will also break all these streaks in terms of wins over consecutive seasons without missing the playoffs.

55
by Starshatterer (not verified) :: Wed, 01/11/2006 - 12:36pm

Are Patriots fans really saying that the Pats’ ten-game postseason winning streak is more impressive than the Dallas Cowboys also winning three Super Bowls in four years, but in the middle year going 12-4 and losing the NFC Championship game instead of going 9-7 and missing the playoffs?
Well, yeah, some Patriots fans are saying just that. They're thinking like fans -- which is to say, not thinking very much at all.

What the fans who send you hate mail see, is every critic (and there are quite a few) who says that the 90's Boys, the 80's Niners, the 70's Steelers, the 60's Packers, and so on, all were better. That whatever the Patriots have done recently, isn't as much as they think. (The earlier FoxSports headlines helped those who confused you with a carping critic, remain so confused.) The less reflective of them lash out. I can't blame you for resenting them, but I feel their point from time to time.

As a Pats fan who occasionally thinks, the 10-game playoff streak is nice, like a kiss on the cheek from Mom. The three Superbowl wins are the weekend in Vegas with a supermodel.

What I *really* want is a 13-game playoff streak.

56
by admin :: Wed, 01/11/2006 - 12:37pm

If the Patriots simply win one more game, they will also break all these streaks in terms of wins over consecutive seasons without missing the playoffs.

Aha! See, THAT's impressive. That's different than just winning Super Bowls, because they also had to make it to the playoffs every single season.

But right. If the Pats had been 11-1, it would not have been any less impressive than 10-0. Which is the whole point. If 11-1 is no better than 10-0, than there's nothing special about the "0" in 10-0. What's special is THREE. As in THREE Super Bowl championships. Why isn't that "enough respect" for Patriots fans?

Starshatterer's metaphor is apt. And hey, I'd like the 13-game streak too. But biased emotional homerism is not my schtick. That's Bill Simmons' schtick. Objective stats analysis is my schtick.

57
by Catholic Samurai (not verified) :: Wed, 01/11/2006 - 12:42pm

RE #42:

If you find a football fan that would rather go watch Brokeback Mountain then watch Divisional Playoff football, give me their #. They need to be put in the Smithsonian as an oddity.

58
by mactbone (not verified) :: Wed, 01/11/2006 - 12:42pm

The problem is that any team that wins the Super Bowl had to win at least three games in a row. Actually, some of the older teams had to win fewer games in a row - do we penalize them for being dominant in an era with fewer postseason games?

Any team that wins a Super Bowl has won three games. Any team that goes to the playoffs will either win the Super Bowl and extend their streak or lose in the playoffs and end their streak. The Buffalo Bills went to four straight Super Bowls and lost the big game. They only had a winning straek of three at any given time. Therefore going to four Super Bowls is less impressive than winning three Super Bowls (plus one game) while missing the playoffs any year that the team does not win the Super Bowl. Those three Super Bowls could come 20 years apart yet we're expected to believe that it's impressive to win the Super Bowl any time the team is in the playoffs no matter how far apart those appearances come?

59
by Joe (not verified) :: Wed, 01/11/2006 - 12:48pm

The problem you're dealing with is that most NFL fanatics are just that - fanatic. They don't see past wins and losses - the only thing that matters is who wins the game, regardless of injuries, luck, or any other factor. If your team wins, any and all arguments, no matter how ridiculous or contrived, are suddenly valid - because at the end of the day your team is still playing.

QB had a bad game but the team still won? It was because the coach was limiting his chances on purpose while he was waiting for the other team to turn the ball over.

Team had a 10 game "win streak" interrupted by a 9-7 season? That's because they only step up their play when IT'S ALL ON THE LINE!

Aaron, I think there are many people who appreciate the fresh look on football analysis here, but a site like Fox or E$PN isn't targeting those people. The average consumer of football media is only concerned if their team wins or loses, not about the underlying factors that contribute to a win or loss.

And the average football fan will read that paragraph and say, of course you stat-geek, the only thing that DOES matter is wins and losses!

60
by Oswlek (not verified) :: Wed, 01/11/2006 - 12:51pm

Mactbone #58,

I don't know if your analogy is appropriate. The reason th 10-0 record is being pushed so heavily is because it is the same coach/QB/other significant players combo that has done all the winning. As I said before, I agree that had the Pats been fortunate (or maybe unfortunate?) enough to be a playoff team in 2002 we wouldn't be hearing about the streak now.

I'm not sure even the media is ignorant enough to push a playoff winning streak if no significant holdovers from the prior teams are intact.

61
by Clod (not verified) :: Wed, 01/11/2006 - 12:57pm

How about this for changing the subject.

"They could really use a tight end in the draft, perhaps Maryland's Vernon Davis or Oregon's Tim Day"

Funny that Day is the replacement for Wrighster, who was supposed to be a good TE and had a nice career at Oregon. Although I must say that the "spread" offense put in this year at Oregon doesn't seem to need a quality TE very often. Either that or Day regressed considerably.

62
by Andrew (not verified) :: Wed, 01/11/2006 - 1:07pm

Aaron #44:

If all you can really ask for in a season is a chance to win a title by playing in the Championship Game, then the Cowboys achievement from 1992-1995 is definitely more impressive than the Patriots, because they went to 4 consecutive championship games (and won 3 of them).

By this measure - consecutive appearances in a title game in the Super Bowl era, we should also give some love in order from best to worst to the Raiders (1973-1977 - 5 in a row!), Cowboys (1992-1995), Bills (1990-1993), Cowboys (1970-1972), Raiders (1967-1970), Eagles (2001-2004).

There is also something to be said for the Browns 6 consecutive title appearances from 1950-1955, and the Dolphins 3 consecutive Super Bowl appearances 1971-1973.

63
by dryheat (not verified) :: Wed, 01/11/2006 - 1:09pm

[i]The three Superbowl wins are the weekend in Vegas with a supermodel.[/i]

Actually, the three Superbowl wins cost me a LOT less, and I enjoyed them just as much. Now if it were only TWO SuperBowls....well, then it becomes a tougher call.

64
by Mikey (not verified) :: Wed, 01/11/2006 - 1:10pm

What's driving Indy's weak special teams DVOA?

As heavy road underdogs - and deservedly so - I feel that my Steelers need to absolutely dominate special teams to win this game, so I'm curious about any insight into the Colts ST.

As for the FOX headline, I thought it was hilarious. Sometimes you just have to laugh. Next week's headline will be "SCHATZ: BRADY CAN'T GET IT UP" and then in much smaller print the sub-head will read (The Patriots' DVOA, That Is!)

65
by Harry (not verified) :: Wed, 01/11/2006 - 1:16pm

Aaron,

Who are these "Patriot fans" of which you speak? Sounds like a few bandwagon jumpers who read Fox and got riled up. I think most Patriot fans actually agree with you. Give these so-called "fans" a simple test - if they can't answer these questions they aren't real Pats fans:

a) which coach had a better record with the Patriots - Mike Holovak or Bill Parcells?

b) Where did John Hannah play college ball?

c) What was Clive Rush's innovative defensive scheme called?

d) In terms of pure talent which Patriot team was the best all time? (hint - it ends in 76)

e) What was the snowplow game?

f) Who was the best LB during the 80s - Andre Tippet or LT?

g)Which teams, in order, did the Patriots beat to get to Superbowl XX?

66
by CaffeineMan (not verified) :: Wed, 01/11/2006 - 1:18pm

Good work, as usual Aaron.

Hey look! With the JAX win last weekend, the Patriots' weighted defense DVOA moves from 11.3% (26th) all the way up to 1.5% (21st). ALMOST AVERAGE BABY! Now that's exciting. Forget about the streak.

67
by ScottR (not verified) :: Wed, 01/11/2006 - 1:20pm

RE: 64 (Mikey)

What's driving Indy’s weak special teams DVOA?

From a quick look at the Special Teams page: Kickoffs, kick returns, and punt returns. Their punting doesn't looks so hot either.

68
by CaffeineMan (not verified) :: Wed, 01/11/2006 - 1:20pm

Harry, your quiz, while a lot of fun, could be considered ageist. :D :D

69
by Al 45 (not verified) :: Wed, 01/11/2006 - 1:24pm

if they can’t answer these questions they aren’t real Pats fans:

a) which coach had a better record with the Patriots - Mike Holovak or Bill Parcells?

b) Where did John Hannah play college ball?

c) What was Clive Rush’s innovative defensive scheme called?

d) In terms of pure talent which Patriot team was the best all time? (hint - it ends in 76)

e) What was the snowplow game?

f) Who was the best LB during the 80s - Andre Tippet or LT?

g)Which teams, in order, did the Patriots beat to get to Superbowl XX?

So real Patriots fans have to be at least 35-40 years old? The young ones don't count unless they spent time studying the entire history of the Patriots?

70
by B (not verified) :: Wed, 01/11/2006 - 1:24pm

"What’s driving Indy’s weak special teams DVOA?"
Mike Vanderjet is driving them, and thier lousy punt coverage is helping out a lot, as wekll. In fairness, the Colts don't punt very often, and Hunter Smith is a very good punter. The problem is he routinely out-punts his coverage.

71
by Rich Conley (not verified) :: Wed, 01/11/2006 - 1:28pm

yeah harry, I dont think I was alive (and past the age of reason) for any of those questions.

72
by Harry (not verified) :: Wed, 01/11/2006 - 1:29pm

Just to get Packers fans riled up - the Green Bay 9-0 win streak (the "record" the Pats broke) is just as bogus, since it ignores 1963- 1964 when the Packers did not make it to the NFL Championship game. Of course the Packers did win the 3rd place games in both those years.

73
by hwc (not verified) :: Wed, 01/11/2006 - 1:31pm

Regarding Aaron's little controversy: Don't take it so seriously. The whole point of something like football commentary is to stir up opinion. If you have "Pats fans" looking for your scalp, good. That means that you have stirred up opinion. Be thankful. The alternative is that nobody cares about DVOA analysis.

As for playoff records and the like, I would suggest that one thing we are overlooking (and DVOA ignores) is that playoff games are qualitatively different from regular season NFL games. Not only is the competition at a higher level, but the intensity of preparation and the pressure of the format doesn't even compare to regular season games.

For that reason, I would suggest that the Pats steamrolling through the Colts, Steelers, and Eagles last postseason was far more remarkable than their regular season record. In a very real sense, postseason games separate the men from the boys in the NFL.

Postseason is all about elevating your game to a higher level and being able to execute under extreme pressure. There are many examples of teams that look like solid gold in the regular season and then self-destruct when the heat gets turned up in the kitchen in the playoffs. The teams that we really remember from the history books are those teams that excel in postseason competition -- much like we remember golfers who win the majors, not the journeymen players who rack up wins in the Hartford Open. The fact that Tiger Woods has had years that he didn't win a major does not detract from his record-setting accomplishments.

74
by Starshatterer (not verified) :: Wed, 01/11/2006 - 1:32pm

Harry (#65 )--

a) Holovak retied before I was born.

b) Hannah was out of college before I was in kindergarten.

c) I may have been eating solid food, by the time Clive Rush was fired.

d) Wrong. It's now.

e) An excuse for Miami fans.

f) Even coldhardfootballfacts.com says it's Lawrence Taylor.

g. Jets, Raiders, Dolphins.

Let OMO vouch for the kind of fan I am. You're just old. :-P

75
by Harry (not verified) :: Wed, 01/11/2006 - 1:36pm

#68,#69,#70

Of course the questions are old. That's the point - if you're really a fan you would know something about the history of the team. I wasn't alive for the Holovak era either. But any Red Sox fan knows who Babe Ruth, Ted Williams, Jim Rice and Roger Clemens are. Any Pats fan should know Gino Cappelletti, Mike Haynes, John Hannah and Andre Tippet. I suppose it is true that in general football fans are far less interested in history than baseball fans.

76
by Shannon (not verified) :: Wed, 01/11/2006 - 1:38pm

If they can't answer these questions, they aren't real patriots fans...

How about just "Who was Hugh Millen?"

77
by Rich Conley (not verified) :: Wed, 01/11/2006 - 1:40pm

RE:75

We know who the Babe, Ted Williams, Etc are not because we are sox fans, but because it is so ingrained in the culture of Boston.

Football is not so ingrained here yet to have that effect. I'm sure it is in areas like Green Bay.

As far as I'm concerned, if none of them are playing, or coaching the team, its irrelevant.

78
by Shannon (not verified) :: Wed, 01/11/2006 - 1:44pm

If you watched every down of the 1-15 seasons, you're a fan.

And LT was better, but Tippet could probably kick his @ss with a roundhouse; and if not, he'd bring in Chuck Norris.

79
by CA (not verified) :: Wed, 01/11/2006 - 1:56pm

I know I'm going to get bashed for this post, but aren't almost all streaks in sports overrated due to the fact that there is so much chance involved? During that 10-0 streak, the Patriots have played several games that could have gone either way, most notably the tuck rule game and Super Bowl XXXVIII. I know many of us would like to think that the reason that the Patriots won those close games is that "they have the hearts of champions" or "Brady / Vinatieri / the Patriots in general are clutch" or some other nonsense like that, but, ultimately, a big reason for the 10-0 streak is that the Patriots have been lucky as well as good. What if the officials had made the wrong call on the tuck rule play, something that I think most of us easily could see having happened? The Patriots would be riding "only" a seven game postseason win steak, and they would have won "only" two Super Bowls. But they wouldn't have been any worse a team. The games I mentioned essentially were coin flip games. Am I supposed to be more impressed with the Patriots because it came up heads both times? Don't get me wrong; the Patriots have been hands down the best team in the NFL over the last five years, and they deserve every championship they have won. But almost all streaks are a matter of luck as well as skill. That's something that all regulars at this site should recognize.

80
by Harry (not verified) :: Wed, 01/11/2006 - 1:56pm

Hey, I never said "Andre Tippet" was the correct answer. As long as the question doesn't make you go "who the hell is Andre Tippet?" you get it right.

81
by Tom Kelso (not verified) :: Wed, 01/11/2006 - 1:58pm

From these responses, I guess none of these Patriot "fans" ever got the concept in Diner.

Holovak, Hannah went to Alabama, don't know, 1976, Ron Meyer (from Send Money University) sent out a work release driver on a plow to clear a spot for John Smith to kick a FG to beat the Dolphins, Lawrence Taylor, and Jets, Browns, Dolphins (not sure about the Brownies).

Some of them are obscure, but really, not knowing about John Hannah -- probably the best guard to ever play the game -- means you sit at the kiddie table.

82
by Rich Conley (not verified) :: Wed, 01/11/2006 - 2:01pm

80.

I think anyone who knew that Andre Tippet was like a 4th degree black belt, and was close personal freinds with Chuck Norris, got the question right.

For what its worth, LT may have been a better LB, but AT would have won in a fight.

83
by Ray (not verified) :: Wed, 01/11/2006 - 2:13pm

RE: #82 "AT would have won in a fight."

You sure? Ever watch Ultimate Fighting Championship? The hightly trained martial arts guys always get beat down by the "wrestler/grappler" guys. I'm betting LT would have fought dirty enough to nullify Tippet's superior skill. ;^)

84
by Nuk (not verified) :: Wed, 01/11/2006 - 2:14pm

Can any Pats fans tell me what other Patriots are 10-0 in the playoffs? I'm assuming there are other players that have been on the team for the last 5 years.
I wish the announcers would tell us, just for variety, what a left guard's record as a starter is once in a while.

85
by Rich Conley (not verified) :: Wed, 01/11/2006 - 2:17pm

re83

You can't forget, those are Highly Trained Wrestler/Grappler guys. The whole thing with UFC, is that once a wrestler gets ahold of a martial arts guy, the fight is over. The martial arts skills are almost totally nullified. Most of the wrestler/grappler guys also are trained in martial arts, its just not their specialty.

I do agree that LT would fight dirty.

86
by Ed (not verified) :: Wed, 01/11/2006 - 2:19pm

Aaron, I get your point of not needing to hype the 10 game win streak when 3 super bowls say it all (3 super bowls = 9 playoff wins). But I do think that 10 games of playoff experience (especially among key positions & head coaches) is statistically relevent in predicting next weeks winners. Didn't last weeks losses by QB playoff rookies highlight this? I don't know how you would capture this in your formula; maybe you do already.

87
by Oswlek (not verified) :: Wed, 01/11/2006 - 2:24pm

Nuk #84

Here is my (probably imperfect) list:

Belichick/Mangini/Scarnechia (plus a few lower level coaches)
Brady
Faulk
Light (although he is on IR)
Bruschi (the Golden God)
Vrabel
McGinnest
Seymour
Troy Brown
Chatham
Izzo
Pass
Vinatieri

88
by Harry (not verified) :: Wed, 01/11/2006 - 2:31pm

#81 - nice job Tom - your vintage Joe Kapp game shirt is in the mail. The one you missed is "the Black Power Defense". Clive Rush thought game planning was too hard, and that just having an all African-American defense would be the key to turning around the Pats, even though he didn't have 11 black defensive players on the squad and had to move some from the offense to make this happen. It didn't work needless to say. Possibly proof that Clive Rush was literally insane.

89
by Shannon (not verified) :: Wed, 01/11/2006 - 2:32pm

Have any except Vinateri and Brady? There's been alot of churn and injuries... Does it count if the starter was on the team but injured?

I'd guess Troy Brown or Willie McGinest. And Bruschi is probably 9-0, but was out last game. Maybe Faulk? But I think he missed at least one with injury.

People it's not - Antwoin Smith, Ty Law (j/k), Seymour, any other defensive lineman, anyone in the secondary...

90
by dryheat (not verified) :: Wed, 01/11/2006 - 2:49pm

Lonie Paxton
Don Davis

91
by Oswlek (not verified) :: Wed, 01/11/2006 - 3:08pm

Dryheat #90,

Paxton is a good catch, but Don Davis was a staring LB on the 2001 Rams.

92
by Oswlek (not verified) :: Wed, 01/11/2006 - 3:13pm

Sorry for the double post, I glanced over Shannon's

Shannon #89

I read the question to be who was on the team, not necessarily who played. But if you want to exclude the guys who missed any of the games than the list is:

Brady
Vrabel
McGinnest
T. Brown
Chatham
Izzo
Pass (he may not have played in all of them, but he was not injured)
Vinatieri
Faulk

93
by John Gach (not verified) :: Wed, 01/11/2006 - 3:16pm

[Originally posted as a comment for "Scouts' Take on Playoff Teams" but I think my comments may be more relevant to this discussion, so I'm reposting it in slightly edited form].

I subscribed this year to SI.com’s War Room [Sports Illustrated's online site]. To say the least, their predictions were underwhelming and sometimes zany. The analysis and discussions of the upcoming games were interesting, but not close to FO’s level of sophistication.

Which leads to the question, why is FO’s statistical approach so clearly superior to impressionistic commentary?

1) When one watches games one is trapped by the now of perception.

2) Since much of what happens is redundant (repetitively recurring events that are highly similar), we attend consciously to the anomalous and nonrecurring events. For example, if one were asked to recount as exactly as possible what happened in a football game, one (without recourse to tape or film) would usually take about as long to tell the story as the highlights on NFL prime time. The brain just isn’t designed to remember the redundant.

3) When we watch games live, we see with all our built-in biases and prejudices. Without considerable special training we have little chance of being more accurate than uncorroborated eyewitness testimony in court trials, for we tend to see what we want to see, what we already expected to be the case.

4) FO’s statistical approach allows one partially to eliminate the hindrances outlined in #1-3, for one can now retrospectively comprehend patterns that were inherent in the live action but difficult or impossible to detect while watching live. It's the postgame stat sheet with each play adjusted for relevance, strenght of opponent, and playing conditions. Of course, one still has to interpret the patterns and make inferences based on them (which may be wildly wrong, as was Aaron Schatz’s prediction that the Giants would defeat Carolina). By following the DVOA changes week to week and reading the insightful and enlightening articles posted on FO) one begins to gain an understanding of the structure that underlies the manifest phenomena — to put it differently one begins to be able to see the games (at least in retrospect) somewhat the way a coach does.

Postscript for this note: Anent DVOA's predictive power, one should note that probabilities are all that can be projected. Even aside from teams playing better or worse than expected on the basis of prior performance, there is no way to account for random events or just plain luck -- see the TB-WAS playoff game for an example.

My observations in #1-3 describe a fan or typical game-watcher, not those who do have specialized expertise. I'm sure that a linebacker or offensive line coach sees a game quite differently than I do.

94
by Judy B. (not verified) :: Wed, 01/11/2006 - 3:20pm

Didn't Paxton miss 2003?

95
by Independent George (not verified) :: Wed, 01/11/2006 - 3:23pm

Forget about LT fighting dirty. He'd be hopped up on enough drugs that you'd need a shotgun to stop him.

*sigh* and I'm a Giants fan.

#86 - After last week's playoff games, I remember reading somewhere that "playoff rookie" QBs are something like 33-34 against the veterans (I can't remember if includes Palmer or not). Which isn't to say that experience doesn't matter, but it seems far less important than, say, the quality of the supporting cast. Remember - you get more playoff experience by winning games, and games tend to be won by the better team.

96
by Will Allen (not verified) :: Wed, 01/11/2006 - 3:26pm

If it is any consolation Aaron, headline writing is probably the most consistently ineptly performed task at most media outlets. I've given up on paying attention to any AP headline because they are so bad, and I've nearly come to the conclusion that the AP headline is a contrary indicator. Then again, they may be of the opinion that inaccurate headlines drive readership, therefore they do it deliberately. I think that is likely the case with FOX's headline writer. Your misery is shared.

97
by B (not verified) :: Wed, 01/11/2006 - 3:29pm

92: Don't we have to exclude guys who lost playoff games in previous years, cause they'd no longer be 10-0. That would exclude Bellicheck too, but whatever. Anyways, going by that criteria, it's down to Brady, Chatham, Pass, Faulk. Those four are obviously the best players in the NFL, cause they're a perfect 10-0 in the playoffs.

98
by Bobman (not verified) :: Wed, 01/11/2006 - 3:31pm

Harry,
Are you for real about the Black Power D? (I'm the right age but the wrong team--an old Colts fan, so don't hold not-knowing it against me. Actually, I think of Ron Meyer as Colts coach! When he managed to finagle Dickerson, his old boy from SMU, onto the for a couple weak playoff runs in the late 80s.)

Back to the BPD: when was this? History says it must have been around 1970 or so. The world needs to know more about this. Pretty sure down in D.C., Scalia would use it to toss out any affirmative action cases as "meritless." What insanity, and my perception is that it's especially uncharacteristic of a Boston sports team.

99
by Kal (not verified) :: Wed, 01/11/2006 - 3:31pm

#36: Yes? I thought the Steelers last season were very impressive, especially given that they were playing with a rookie QB. That meant either that their defense was absolutely dominant or they had caught lightning in a bottle.

100
by B (not verified) :: Wed, 01/11/2006 - 3:33pm

Oh, and I forgot about Izzo. He was on the Dolphins from 1996-2000, but it doesn't look like they made the playoffs in that span.

101
by Sophandros (not verified) :: Wed, 01/11/2006 - 3:37pm

The problem you’re dealing with is that most NFL fanatics are just that - fanatic. They don’t see past wins and losses - the only thing that matters is who wins the game, regardless of injuries, luck, or any other factor. If your team wins, any and all arguments, no matter how ridiculous or contrived, are suddenly valid - because at the end of the day your team is still playing.

It's not just NFL fanatics. I've seen SEC fans in general and LSU fans specifically become the biggest USC haters on the planet--even though most rational people know that the 2003 USC team would have beaten LSU by two scores.

102
by Starshatterer (not verified) :: Wed, 01/11/2006 - 3:39pm

not knowing about John Hannah — probably the best guard to ever play the game — means you sit at the kiddie table
I know about Hannah -- as a Patriot.

I try not to pay too much attention to college football. The subject is painful -- I went to Rutgers.

103
by pcs (not verified) :: Wed, 01/11/2006 - 3:42pm

Teh Partiots win streak is clearly ranked too high because the Houston Texans are undefeted in the playoff and never lost a post season gam in they're history. Counting who has the least post season losses is clearly better than this. DOVA sucks and David Car rules and Bo Bice rocks!!!!!!

104
by B (not verified) :: Wed, 01/11/2006 - 3:47pm

And David Patten almost made it. 9-0 with the Patriots, and the Redskins are 1-0 this year, but it looks like he's no longer on the active roster.

105
by Will Allen (not verified) :: Wed, 01/11/2006 - 3:55pm

Great post, Mr. Gach. I'd add that if one really wants to grasp what has occurred in a game, one has to watch it more than once; there is just too much going on to really take it all in while watching it live.

This still doesn't address the fact that most t.v. shots crop the field so that it becomes difficult to really see the game accurately. Toss in play by play guys who are useless, in that they don't give subsitituion information consistently, and even having Tivo and the NFL Ticket (which are great) makes accurately evaluating the game, by simply watching it, difficult for the most impartial observer.

Also, assuming one hasn't figured out a way to get paid for being knowledgeable about the NFL, the reality of the time needed to watch enough games, often enough, becomes an insurmountable barrier to real enlightenment. Heck, most of the people paid to analyze NFL games don't watch enough to really know what they are talking about. All-pro selections and power rankings by people whose inquiry is limited to watching three or four games a week, one time, are next to worthless. The FO method provides an empirical, although imperfect, framework for getting one's arms around a truly immense body of information.

106
by Harry (not verified) :: Wed, 01/11/2006 - 3:55pm

#98

The BPD was in 1969 and it's true, I've seen it in a number of sources. Click on my name to see one of them. Of course it didn't last more than one game.

You're right that an all-black defense sounds pretty uncharacteristic for a Boston sports team of that era, but I think the Pats were a lot better about integrating than the Red Sox or Celtics. As far as I know the Sullivans were pretty open minded for Boston Irish, just very tight with their money.

107
by Ed (not verified) :: Wed, 01/11/2006 - 4:00pm

#95: I was unaware of the 33 -34 record of "playoff rookie" QBs. That would mean last weekend's results where an anomaly. Of course that statistic alone doesn't include other players /coaches exeperience; nor their number of playoff appearences (won or lost). Maybe there is no correlation, or at least not a heavily weighted one. I still think it might be relevent given last weeks outcomes.

108
by Sammy3469 (not verified) :: Wed, 01/11/2006 - 4:06pm

Aaron, I know last weeks Panther's performance was off the charts, but how similar is the improvement over the past 2 games similar to what they did in 2003?

109
by dryheat (not verified) :: Wed, 01/11/2006 - 4:08pm

The Celtics?

You mean the first NBA team with a black player, the first NBA team with more than one black player, the first NBA team to play five black players, and the first NBA team to hire a black coach?

Those Celtics?

Funny how perception is sometimes, because the 1980s Celts had two dominant white players, they became known as a racist organization.

110
by Jon (not verified) :: Wed, 01/11/2006 - 4:27pm

Re: 27

Aaron, I'm sorry for starting this whole run of posts about the Pats' postseason streak. Believe me, I know you support the Patriots, but in no way do I think you're a "biased Pats homer," nor did I try to argue anything to that effect. "Lay down like dogs" just made it sound like the Packers didn't try in the Jets game, which was far from the case.

If you were to blame any other team for the Pats not making the playoffs, blame Atlanta, who blew a 16-10 4th quarter lead to Cleveland via a fumble at its own 11, a 64-yard William Green TD run, and running Warrick Dunn up the middle on both 3rd and 4th downs from the Cleveland 1.

111
by Mikey (not verified) :: Wed, 01/11/2006 - 4:28pm

101 -

If most rational people know that USC would have beat LSU by two scores in 2003, then why did the more objective elements of the BCS ranking place LSU #1? Which is more rational, data analysis or anecdotal observation?

USC agreed to a certain method of determining a champion, then when the system didn't work in their favor they blew it off and declared themselves to be the champions anyway. Now that's class!

In a similar vein, I'd like to declare that regardless of the outcome of this weekend's games, the Pittsburgh Steelers are the 2005 Champions of Pro Football. Way to go, guys!

112
by Jon (not verified) :: Wed, 01/11/2006 - 4:32pm

Re: 104

David Patten was on the 1997 Giants, who lost to Minnesota in the first round of the playoffs.

113
by Harry (not verified) :: Wed, 01/11/2006 - 4:52pm

#109, you're right, I watched too many Spike Lee movies in the 80s and let it bias my thinking. I apologize to the Celtics.

114
by Dan Riley (not verified) :: Wed, 01/11/2006 - 5:07pm

The Black Power Defense thing was true, and when you think of it, it really was a racist concept. It was all about putting the fear of Huey Newton and Stokely Carmichael into the minds of the Jets and Bills....not at all about we shall overcome. That being said, the coach who imagined it was more nutcase than racist, and it's a story that belongs in the Pre-Respect Age of the Pats along with the time they pulled a guy out of the stands to play defense and drafted a dead guy. All nicely chronicled by the Late Bard of Boston Football Will McDonough. Google his name if you want to learn more about why Pats fans always have to buy extra large sweaters to fit those chips on our shoulders.

115
by Boston Irish (not verified) :: Wed, 01/11/2006 - 5:10pm

# 106:

"...As far as I know the Sullivans were pretty open minded for Boston Irish..."

I'm glad that the Sullivan's didn't perpetuate any stereotypes.

116
by Jeremy Billones (not verified) :: Wed, 01/11/2006 - 5:15pm

>If most rational people know that USC would have beat LSU by two scores in 2003, then why did the more objective elements of the BCS ranking place LSU #1?

Because the objective elements of the BCS are forbidden from analyzing all the data in making their analysis. (They can only use the W or L result of a game, not the score or score differential. They might not even be able to take into account home field; I don't recall.)

I actually have no opinion on the USC vs
LSU question, other than the fact that
the media has been very unfair to LSU
during and since. But the computer
issue is one I find a little annoying.

117
by Kal (not verified) :: Wed, 01/11/2006 - 5:22pm

Aaron, I think the only way you're going to solve this is to show us all, using DVOA, how dominant MACHINE has been in the last 6 weeks. That will prove to the Pats rabid fanbase that you are indeed no anti-Pat hater.

118
by PatsFan (not verified) :: Wed, 01/11/2006 - 5:24pm

Re: 114

Or read Felger's "Tales from the Patriots Sideline", which is a history of all the pre-2001 hilarity and incompetence that surrounded the team from its founding in 1960.

I still remember Callahan saying back in 2000 or so that if someone were to write a book about the team, it should be called "4th and Forever". And as of then, that really would have been a perfect title for a book about the team.

119
by chris clark (not verified) :: Wed, 01/11/2006 - 5:26pm

RE 37 "subjective"

The definition of "subjective" in 37 is so wide that nothing can escape being subjective. Forming a hypothesis, testing it against the data and analyzing whether the data confirms it or not is the "scientific method". If Aaron were subjective (but still stats based) he would look for statistics the support his preformed conclussions. Yes, one has to select a hypothesis to measure (fumbles recoveries are random), but testing that hypothesis and keeping it only when validated by the statistics is not subjective. Now, if Aaron found a better predictor but discarded it because "it didn't feel right", that would be subjective. However, he includes the Pythagorean wins in his analysis because it has good predictive value. He has also analyzed other factors (the stomps, gut checks article for example) to see if they can provide further enlightenment. That is the antithesis of subjectivity.

120
by spenceKarl (not verified) :: Wed, 01/11/2006 - 5:28pm

So I'm still curious what the weighted values are for weeks 5-10. Were they determined similarly to how the normal weights for WDVOA are determined? And weeks 11-18 are weighted at full strength, is that right?

121
by POUM (not verified) :: Wed, 01/11/2006 - 5:28pm

#79 & #93
Couldn't agree more with both of these excellent posts. As I've no interest in Patriots Trivia 1959-2006 you made this thread worth reading.

122
by John Gach (not verified) :: Wed, 01/11/2006 - 5:53pm

Re: #105. Trenchant observations that supplement exactly the point I was trying to make. As Mr. Allen notes, the very framing of the TV image constrains what it is even possible to see in a televised game. Thus, a televised game is really a mediated event filtered through the judgments of behind-the-scene producers and editors, even though we perceive it as immediate. For example, depending on the framing of the image, it may not even be possible to identify what type of defense is being played during a particular play. Every now & then, though, the networks do come through: the visual & verbal coverage of the disguised NE zone that looked like man-to-man in the play that resulted in Asante Samuel's interception in the Jacksonville game was, I thought, terrific.

I completely agree about the worthlessness of most commentary on the games, whether live or retrospective. We FO regulars would like C-SPAN for NFL; we usually get commentary by Madonna. Typically, commentators are seduced by the last game and ignore longer trends (since that would require actually _studying_ the games, wouldn't it?). For example, consider the upcoming Chicago/Carolina game. After Carolina's complete destruction of the Giants, running at will and stuffing the Giants offense entirely, lots of people seem to think that that game presages a Carolina win in Chicago. But look deeper. Except for Chicago's early 24-7 loss to Cincinnati, nobody has been able to score on them at home. And if you look at the stats for the Cincy game (http://www.nfl.com/gamecenter/recap/NFL_20050925_CIN@CHI) you'll find that it was a personal gift from Kyle Orton, whose five interceptions handed the game to Cincinnnati. Even so, the Bears were (somehow) still only down 10-0 late in the third quarter. Palmer threw three touchdown passes but was otherwise held to quite modest numbers. If one reads Ned Macey's FO commentary on the Panthers-Giants game, one will rightfully be more than a bit skeptical about Carolina's prospects next weekend.

As Aaron Schatz pointed out in his Fox commentary and as I noted in an earlier comment in a different thread, the home team has a massive historical advantage in the divisional playoffs: 41-9 since 1990 and 8-2 in the past five years. Thus, even without knowing which actual teams were playing, one would figure that the home teams are, as a group, a 4-1 favorite. It is highly unlikely that more than one road team will win a game and reasonably likely that none will. Paul Zimmerman at SI.com picks NE for the lone upset (figuring, maybe, that surely one road team has to win and this looks like the best possibility). Personally, I like all the home teams; but if I were to single out a possible upset, I'd be completely contrarian and pick Washington. Yes, they're beat up and injured and barely beat Seattle at home in overtime, but they've played the best football in the league over a six-game span, have a solid and cunning coaching staff, and are being picked as automatic losers by nearly everyone (9 point underdogs as of today). They're a dangerous team, as Carolina would be were they not in a bad matchup in Chicago.

It is possible to get recompensed for having some expertise about the NFL by wagering, but it takes a lot of time, great discipline, and high tolerance for risk to do it well. See the numerous interesting comments (including a few by me, which I hope were interesting) in the discussion of " The Evolving Predictive Powers of DVOA" at http://www.footballoutsiders.com/2005/10/14/ramblings/dvoa-rankings/3003....

Will Allen wrote: "The FO method provides an empirical, although imperfect, framework for getting one’s arms around a truly immense body of information." Exactly. I couldn't agree more. In the two years that I've been a regular reader it has transformed how I see and understand football games.

Though obviously it's a great boon to Mr. Schatz in particular and FO in general, nonetheless the (I imagine) large increase in visitors to the FO site does have its demerits, quite apart from the occasional problem even getting onto the site. The casual and first-time visitors, used as they are to impressionistic and ill-informed commentary, cannot be prepared for the seriousness and objectivity of FO. Truly to "get" what FO is doing, one must spend much time over months studying the stats and reading the articles. A one-time 10 minute visit may just leave one confused (or, so it seems, unhappy that one's favorite team is being dissed rather than praised).

123
by Falco (not verified) :: Wed, 01/11/2006 - 7:21pm

If most rational people know that USC would have beat LSU by two scores in 2003, then why did the more objective elements of the BCS ranking place LSU #1? Which is more rational, data analysis or anecdotal observation?

Sorry, but this is one of my biggest pet peeves. The BCS is absolutely not objective, and does not become objective just because it uses “computer rankings.� To call the BCS computer rankings objective does harm to other rankings that really are striving for objectivity. The BCS is, in fact, the polar opposite of what Aaron and FO do here with DVOA analysis. As stated in #116, the “computers� only use W/L and opponent’s W/L, and do not take into account home field (which unlike the NFL, features certain conferences/big name schools that haven’t played a road non-conf game in years). There is no measurement of “how a team played� (margin of victory/pythagorean points for/against, drive success, DVOA, for example), “when a team played� or “where a team played.� It allows a one-loss 2000 Nebraska to play for the title despite giving up 63 points (6 rushing touchdowns? to Chris Brown) in the last game, and in the only road game all year against a team approaching .500.

If it were attempting to be objective, it would test and revise the formula based upon a body of data, and realize that all of the measurements of how a team was playing do matter in improving accuracy. It does occasionally get tweaked, but not by looking at data, but in reaction to a particular problem that comes along every other year. (Quality Wins bonus after FSU/Miami/Washington controversy, e.g.).

Assuredly, if the goal were accuracy, the current “computer rankings�fail because the best information is not used. Any ranking that will lower a team merely because of “who� they play that week (before the contest has been played and we actually see how they played, a novel concept) is not logical. However, if the goal is to get the traditional power conferences in the top spots of the rankings and get the big money payouts of the BCS bowls, it is a resounding success. Teams from power conferences pay smaller schools to come in and fatten their home record. The other schools in the conference then get to use these improved non-conf records in their “opponents’ opponents’ winning percentage,� because they all play each other.

As for USC-LSU, I had USC as slightly better, but the problem was the BCS having Oklahoma #1 after the loss to K-State, which forced a choice between the SEC and Pac-10 opponent's opponent's records, er, LSU and USC.

124
by Kal (not verified) :: Wed, 01/11/2006 - 7:25pm

Another problem with the BCS rankings is that only a third of it is done via computers. The rest is done via the national rankings, and those aren't objective in any remote way.

125
by squintsp (not verified) :: Wed, 01/11/2006 - 8:19pm

Re: 123 & 124

While I agree with Kal, I think Falco is confused about the meaning of objective. Falco, you can disagree with the set of facts that the computer formulas choose to use, but to the extent that they use objective facts (e.g., W/L, opponents W/L), they indeed are objective. If they varied the facts that they took into account based on which team they were ranking, then they would be subjective. Or if they took into account factors that are difficult, if not impossible, to define or observe (e.g., swagger, momentum and (with a tip o' the cap to De-rek Je-ter) intangibles), they would also be subjective.

Peace.

126
by B (not verified) :: Wed, 01/11/2006 - 10:09pm

Is it me, or did this thread somehow turn into an Ayn Rand discussion?

127
by Dan Riley (not verified) :: Wed, 01/11/2006 - 10:36pm

Great one, B...and that's extactly why the neanderthals pretty much steer clear of this place and why Aaron should stop angsting about it and just be happy he's not running Pete Priscopo's place. Jeesh.

128
by Paul (not verified) :: Wed, 01/11/2006 - 10:59pm

Hmmm, I'm wondering...if I go to the Patriots home page, will I find a thread about the playoffs?

129
by DeepThreat (not verified) :: Wed, 01/11/2006 - 11:23pm

10-0 in the postseason is great. You are beating the best in the league every week. Now that that is settled, try and convince me how the 2002 "Big Nickel" Patriots were better than the 2001 team...

130
by shonk (not verified) :: Wed, 01/11/2006 - 11:27pm

Is it me, or did this thread somehow turn into an Ayn Rand discussion?

If it had everybody would be gratuitously capitalizing "objective".

131
by young curmudgeon (not verified) :: Thu, 01/12/2006 - 12:54am

Aaron, it was almost inevitable that signing on with Fox would eventually lead to some kind of sensationalistic misrepresentation--anyone that was really "fair and balanced" wouldn't have to keep reminding us that they are. It was probably a good business decision, but going from a nice little "mostly among friends" web site where rational discussion is the order of the day at least 85% of the time to part of a big time organization in the entertainment business, where hype is king and any publicity is good publicity is going to pose some problems and this incident seems to be one of them. In all honesty, I'm amazed by the level of civility and thoughtfulness that comes through on most of the comments I read here.

132
by Larry R. (not verified) :: Thu, 01/12/2006 - 1:58am

RE: Also, Carolina is on the east coast, so the Panthers get extra credit.
Aaron, you might be wondering why you haven’t had any Seattle comments this time, Probably because all of us people here in Seattle are drinking our Latte’s (coffee for you east coast people) and LAFFING at these power rankings......lol

133
by Larry R. (not verified) :: Thu, 01/12/2006 - 2:43am

Opps sorry "laughing"

134
by kyle(tcn) (not verified) :: Thu, 01/12/2006 - 4:25am

i just read all the comments and laugh and laugh.

i'm still so giddy over DVOA and FO advice telling me to pick up Steve Smith that i don't care how it works out on everything else. i'm the sort of maniac willing to be so enamored with a single success.

too bad the season is winding down. its been a fun one.

135
by Joe Dimino (not verified) :: Thu, 01/12/2006 - 6:07am

#116 and #123, final 2003 Sagarin predictor ratings, which consider points scored and HFA:

1. Oklahoma 96.84
2. LSU 96.33
3. USC 95.07

Just saying, I think whether they would have won by 2 scores is reasonably in doubt even with objective measures that take into account the score, and not just W/L.

LSU won the won that doesn't take score into account over USC also, 98.21 to 95.56.

LSU was 3-0 vs. the final top 10, and 6-1 vs. the top 30. USC was 1-0 and 4-1.

I think it would have been a great game, no way I'd pick anyone to win it by two scores. Put a gun to my head, and I'm taking LSU, that D was incredible.

136
by Kibbles (not verified) :: Thu, 01/12/2006 - 8:50am

Re #110: If you were to blame any other team for the Pats not making the playoffs, blame Atlanta, who blew a 16-10 4th quarter lead to Cleveland via a fumble at its own 11, a 64-yard William Green TD run, and running Warrick Dunn up the middle on both 3rd and 4th downs from the Cleveland 1.

Atlanta didn't keep New England out of the playoffs. New England was doomed the second the Jets lost, because the Jets had the three-way tiebreaker to win the AFC East, which relegated New England to the Wildcard pool, and New England was behind both Cleveland *AND* Denver in wildcard tiebreakers. Even if Cleveland had lost, Denver would have been the conference's 6th seed. New England's sorry tiebreaker situation meant that they HAD to win the AFC East or they were out of the playoffs, and to win the AFC East they HAD to finish with a better record than the Jets.

I only know this because I'm a Denver fan, and there was speculation in Denver after the game that Dan Reeves had lost to Cleveland on purpose just to keep Denver out of the postseason. I never believed a word of it, and had a good laugh at the expense of all the conspiracy theorists, but I guess it stuck in my head.

That was a really wild season. The playoff scenarios going into week 17 were something like 3 pages long. Three teams were tied at 9-7 for the AFC East division title, and then two more teams were tied at 9-7 (tied with the three aforementioned AFC East teams) battling for the final wildcard.

Re #135: I agree, I feel like the majority of this "two scores" business is simply revisionist history. If LSU had won the national championship last year and been called the greatest team to ever play the game of college football this year (yeah, I suppose waiting until AFTER the national championship to make that call was unnecessary after all), then I think everyone would be going on about how LSU would have obliterated USC in 2003.

That said, I'm still cracking up at the media sensationalism that surrounds USC. Apparently they're the best team in the history of college football, despite the fact that they're only the second best team this season. The need for the media to make everything the biggest, the best, the most superlative is ridiculous, and pretty insulting. I'm intelligent enough to appreciate the tremendous accomplishment that USC has managed without you bludgeoning me over the head with how it's the single greatest accomplishment in the history of the sport. I can appreciate an accomplishment that's the THIRD greatest in history quite as easily as I can appreciate an accomplishment that's the SINGLE greatest in history.

In many respects, I feel like that's the same thing that's going on with the Pats now. Yes, their accomplishment is truly astounding. It will STILL be astounding if we acknowledge that it's not the greatest accomplishment in the history of professional football. Yes, the Patriots winning streak is impressive, regardless of any variables or qualitative judgements which we attach to it.

I almost feel like mainstream media is responsible for the "drop in IQ" that accompanies becoming a Patriot fan. It's like they encourage Patriots fans not to think, to simply trust in Belichick, or Brady, or the streak, or "swagger", or whatever. I truly believe that a substantial portion of the Patriot fanbase seems to be operating under the assumption that the Patriots will never lose another playoff game again. Ever.

I mean, they're going up against a team that is clearly better than they are this week. That team has homefield advantage, the bye week, and history on their side. This is a team that has ALWAYS been a tough matchup for the Pats (Brady and Belichick are 1-3 against Denver, with the one win coming by 4 points against a team led by DANNY KANELL). If the Patriots don't have a chance of losing this week (and many fans don't just assert that they are LIKELY to win, they assert that there's no chance of any other outcome), then when will the Patriots EVER LOSE AGAIN?

The Patriots have had a tremendous streak, but as someone mentioned earlier, any such sports streaks exist in very large part due to luck. Likewise, the Patriots have been the BETTER TEAM in 8 of those 10 games. New England has only been an underdog TWICE during the streak- In the AFCC and SB in 2001. They've never played on the road in the divisional round. And trust me, however good Brady and Belichick might be, they are fully capable of losing a football game in the playoffs. Just because they haven't yet doesn't mean that they can't.

137
by dryheat (not verified) :: Thu, 01/12/2006 - 10:11am

The Patriots were actually a 3-pt underdog at home vs. the Colts in last years Divisional Playoff.

138
by Dan Riley (not verified) :: Thu, 01/12/2006 - 10:38am

Shorter Kibbles: What goes up must come down.

139
by Dan Riley (not verified) :: Thu, 01/12/2006 - 10:44am

By the way, Kibs, I asked this earlier in the thread...and in a deeply sincere way, no Snark intended...I'd like a Broncos' fan to offer a take on the Denver game against the Ravens. I thought that would make a good tape for BB to study (though yesterday I read that he had the Pats studying the Broncos v Eagles game...that crazy Belichick).

140
by Al 45 (not verified) :: Thu, 01/12/2006 - 11:27am

I mean, they’re going up against a team that is clearly better than they are this week.

Uh, no... they're going up against a team that was 'clearly' better than them on Oct. 16. This isn't the same Patriots team.

Everyone wants to dismiss the Patriots performance over the past 6-7 weeks because it was against 'bad' or 'weak' NFL teams. Yet, if you all worship at the altar of DVOA, and DVOA adjusts for strength of schedule, it's quite clear that what the Patriots did was far above what would be expected of an average team against those 'weak' opponents.

There's no 'magic beans' here, and anyone who thinks the Broncos are 'clearly' a better team hasn't paid enough attention to what the Patriots team looks like 'now'.

It's ok though... it was fun watching the 15-1 Pittsburgh Steelers put stock in the whooping they put on the Patriots in Oct. of 2004, when the Patriots were killed by injuries.

NOTE: If we were talking about the Patriots vs. Colts, I wouldn't be saying the same thing. The Colts ARE a better team than the Patriots. The Colts should easily be favored against the Pats.

141
by Scott de B. (not verified) :: Thu, 01/12/2006 - 11:43am

It allows a one-loss 2000 Nebraska to play for the title despite giving up 63 points (6 rushing touchdowns? to Chris Brown) in the last game.

Yet that was the correct decision (of all the one-loss teams, Nebraska lost to the strongest team), and any DVOA system would have shown that.

142
by masocc (not verified) :: Thu, 01/12/2006 - 11:50am

I've been tryingt to figure out how I keep being a day early for everything this week, then it hit me:

1st, MMQB came out on Sunday night. Now we get the weekly rundown on Tuesday evening!

PLEASE! Stop the insanity! Today's... Thursday, right?

143
by Starshatterer (not verified) :: Thu, 01/12/2006 - 12:04pm

Al 45 (#141 )--

"Magic beans" are true-but-useless stats like "Brady and Belichick are 10-0 in the playoffs" or "Brady and Belichick are 1-3 against Denver." I don't believe that Denver's regular-season win in 2001 or New England's playoff wins from the same season, are going to have any impact on Saturday.

Here are some useful stats:

1. Denver's DVOA, standard and weighted, was better for the year.

2. New England's DVOA trendline is way up, which has, in the past, shown a team that can cause playoff upsets.

3. Denver's playing at home, and there is an enormous HFA at Mile High/Invesco.

The rest of it's whether you think that factors 1 and 3 outweigh factor 2, or vice versa.

144
by hwc (not verified) :: Thu, 01/12/2006 - 12:19pm

I'm continually amazed when people think they can confidently predict the outcome of NFL playoff games.

If you asked the two most expert people involved with Saturday's match-up (Shanahan and Belichick), I think they would both laugh at the notion of predictions and would both point out that playoff games usually hinge on making one or two key plays at critical points in the game. I guarantee that they would both roll their eyes at predictions based on a regular season matchup, strenth of schedule, Sagarin ratings, DVOA tables, or power rankings.

I also be that both of them would tell you this is the kind of game they live for -- a matchup of two quality, well-coached teams in the tournament.

IMO, fans or pundits who think they know which team will win are just fooling themselves.

145
by dryheat (not verified) :: Thu, 01/12/2006 - 12:23pm

I thought Magic Beans, as defined by Aaron, was the attitude of Patriots fans that weeks 1-9 never happened.

I'm thinking Denver wins 24-20, but a Patriots win by the same score would hardly be shocking, I don't think.

146
by B (not verified) :: Thu, 01/12/2006 - 1:10pm

I think magic beans are fallacious reasoning statements like "Brady and Bellicheck never lose in the playoffs." or "Shannahan has never won a playoff game without Elway."

147
by Starshatterer (not verified) :: Thu, 01/12/2006 - 1:29pm

Right, I phrased it wrong. But I stand by the examples I gave.

148
by Will Allen (not verified) :: Thu, 01/12/2006 - 2:16pm

hwc, anybody who says the "know" what the future holds is by definition an idiot. However, people who are well informed and diligent can gain insight as to probabilities, and this skill is like most others regarding human beings; there is a bell shaped curve with a few people who are good at it, a few who are horrible, and the rest somewhere in the middle.

I suspect predicting NFL games is much like investing; there are a VERY few mutual fund managers or institutional investors who can indeed beat the market averages over a long period of time, say two or three decades (think Buffett or Lynch), but it almost impossible to identify them ahead of time, which is why people are better off in index funds.

I have no doubt that there have been some people who have consistently made money wagering NFL games over the past 20 years, but there are damned few, and they ain't talkin'. As to the Pats/Broncos, on a neutral field, I'd give the Pats a tiny edge, since it just isn't the same team that the Broncos beat earlier this year. In Denver, however, I give the Broncos a tiny edge. It isn't a game, however, in which the public has made such an obvious mistake in evaluation that it becomes a great game to bet on.

149
by admin :: Thu, 01/12/2006 - 3:00pm

Will apparently has psychic access to tomorrow's previews.

150
by dryheat (not verified) :: Thu, 01/12/2006 - 6:39pm

Quoting Aaron, from the 1/4 Playoff Fantasy Draft.....

but I just can’t buy the “Patriots magic beans� theory. That’s what I call the belief of Patriots fans that nothing that happened in the regular season matters and that no injuries to the Patriots matter, because the Patriots will always win the Super Bowl as long as Bill Belichick is coaching and Tom Brady is playing quarterback. But there’s no way to turn Ellis Hobbs and James Sanders into Ronnie Lott and Deion Sanders. People act as if Bill Belichick has some magic beans that instantly render the other teams helpless in the playoffs. I don’t think the Broncos are affected by the magic beans.

151
by thad (not verified) :: Thu, 01/12/2006 - 11:33pm

Many of you have made comments about the Denver New England game. Yet I don't think there has been one about the Denver defensive line, which seems godawful.
I do not think anyone has mentioned the fact that Denver got only 28 sacks this year.
I hear so many comments about how "you have to pressure the qb"
If that is true, is Denver doomed?
Has there been a team that got a sack on only 4% of their opponents pass plays and made it to the conference championship?
Ok there probably has, but I would guess not too many.
I am not really saying Denver won't try to get pressure, I am saying they will try to blitz and get torched, much like last year.

152
by Björn (not verified) :: Thu, 01/12/2006 - 11:46pm

Regarding the Denver-Baltimore game.

I try not to think about it.

One interesting thing was watching Boller play. As a Bronco fan and Plummer booster I couldn't help but think: "Is this how Jake looked before this year? Have I been that blind?"

153
by Björn (not verified) :: Thu, 01/12/2006 - 11:54pm

RE: 151

It seems to me that Denver often gets pressure, but just doesn't get the sacks. Case and point, Denver only sacked Donovan Mcnabb twice in Week 8, despite numerous blitzes and tons of really good pressure. It was in that game that Denver rushed 9 guys on one play. They put the corners on the wide receivers, and the safties and linebackers on the line, with the linemen, of course. The fast incomplete pass that Mcnabb threw was just as good as a 1 yard sack, in my opinion.

154
by Al 45 (not verified) :: Fri, 01/13/2006 - 12:01am

But there’s no way to turn Ellis Hobbs and James Sanders into Ronnie Lott and Deion Sanders.

Belichick doesn't have to turn them into Ronnie Lott and Deion Sanders. This analogy is ridiculous since it makes the comparison to two Hall of Famers (most likely).

With the front 7 of the Patriots, Ellis Hobbs and Artrell Hawkins (no more James Sanders or Arturo 'freakin' Freeman) simply have to be average. That's what I don't think people get.

Do people really think Randall Gay was Deion Sanders last year? No, he wasn't even close. Yet he was league average, and guess what... they absolutely dominated the best passing offense in the league. Why? Because they were able to play their ASSIGNMENTS at a league average level.

The Patriots problems came this year when the cornerbacks and safeties were repeatedly blowing their assignments. Now that they understand their assignments and have enough in game experience to read offensive sets, they're playing at a much higher level.

Again... Aaron's example of 'magic beans' is flawed, because it makes a ridiculous analogy suggesting that people think Belichick is turning players into Hall of Famers.

Nobody is suggesting that. What they are suggesting is that he gets more out of players than other coaches in the league... and that I don't think one can question.

Belichick and Mangini devise defensive schemes that put their players in positions to succeed (so long as they maintain their assignments). They don't put them in positions where they have to perform to 110% of their ability in order to succeed.

Right now, the only real loss in the secondary is Rodney Harrison.. and that is HUGE. However, Ellis Hobbs has absolutely replaced the level of play provided by Randall Gay last year, and Asante Samuel and Eugene Wilson are still there from last year.

Hawkins is nowhere near Rodney Harrison, but he gives you a decent cover safety, and while he's not a guy you'd base a defense around, he's playing his position at a league average level, and not making bonehead mistakes.

That's all you need with the dominance of the Patriots front 7.

155
by hwc (not verified) :: Fri, 01/13/2006 - 12:20am

Speaking of "beans", I wonder what kind of beans the Broncos ate before their playoff sinkers facing the Colts the last two seasons?

After those two meltdowns, I really think that Denver has to provide a little "show me" before we annoit them as poised and ready for postseason dominance.

156
by thad (not verified) :: Fri, 01/13/2006 - 12:26am

re 153,
the Denver secondary is great no doubt.
They forced 269 incomplete passes, most in the league.

157
by B (not verified) :: Fri, 01/13/2006 - 12:52am

155: Denver had to travel to a place with a significant home field advantage and face a team with a higher WDVOA and DVOA. I'm just sayin.

158
by Kibbles (not verified) :: Fri, 01/13/2006 - 6:15am

Re #137: The Patriots were actually a 3-pt underdog at home vs. the Colts in last years Divisional Playoff.
I could have sworn that they ended the week as a pickem.

Re #139: By the way, Kibs, I asked this earlier in the thread…and in a deeply sincere way, no Snark intended…I’d like a Broncos’ fan to offer a take on the Denver game against the Ravens. I thought that would make a good tape for BB to study (though yesterday I read that he had the Pats studying the Broncos v Eagles game…that crazy Belichick).

I don't think film is going to help much. It was just an uncharacteristically bad game for Denver. Watching the Baltimore film to figure out how to beat Denver would be about as helpful as watching last season's NE/Miami film to figure out how to beat the Patriots.

Besides, my impression from the game was that it was more one-sided than the final score indicated.

Re #140: Everyone wants to dismiss the Patriots performance over the past 6-7 weeks because it was against ‘bad’ or ‘weak’ NFL teams. Yet, if you all worship at the altar of DVOA, and DVOA adjusts for strength of schedule, it’s quite clear that what the Patriots did was far above what would be expected of an average team against those ‘weak’ opponents.
Yes, DVOA clearly shows that the Patriots performed above what would be expected from an "average" team over the last 8 weeks. That said, Denver's weighted DVOA is still 20% higher, and HFA is usually worth 17% on top of that. It seems pretty clear that, at least according to what has happened on the football field, Denver is still drastically better than New England (and pretty much everyone else, except Indy).

Re #151: Many of you have made comments about the Denver New England game. Yet I don’t think there has been one about the Denver defensive line, which seems godawful.
I do not think anyone has mentioned the fact that Denver got only 28 sacks this year.
I hear so many comments about how “you have to pressure the qb�
If that is true, is Denver doomed?
Has there been a team that got a sack on only 4% of their opponents pass plays and made it to the conference championship?
Ok there probably has, but I would guess not too many.
I am not really saying Denver won’t try to get pressure, I am saying they will try to blitz and get torched, much like last year.
So what you're saying that the team that pressures the other QB more regularly will have a huge advantage?

Well, New England finished the season +5 in sack differential (sacks - sacks allowed). Denver finished... +5.

Also, Denver's pass defense is #4 in the entire NFL. A lot of QBs have dumped the ball off in a hurry to avoid the pressure that Denver generates (which is pretty strong pressure), and they've made a lot of bad decisions in the face of a heavy passrush (while Denver has had lots of protection and has made good decisions, leading to a +20 turnover differential, second best in the NFL).

Re #153: It was in that game that Denver rushed 9 guys on one play.
They actually did it on 3 consecutive plays. That Larry Coyer has stones like I've never seen before.

Re #156: re 153,
the Denver secondary is great no doubt.
They forced 269 incomplete passes, most in the league.
Misleading statistic. Denver also leads the league in passing attempts against, which is a big reason why (much like how Marino lead the league in yards, completions, and incompletions... and not at all coincidentally, passing attempts, too).

Anyway, Denver faced the most incompletions in the NFL... and the second most completions. And the most attempts by a HUGE margin. It's all part of the same thing. However, they allowed the 3rd lowest completion percentage, the 6th lowest yards per attempt, and had the sixth most interceptions. The unit was very good, but you really have to look at per-attempt stats to appreciate it.

Re #157: 155: Denver had to travel to a place with a significant home field advantage and face a team with a higher WDVOA and DVOA. I’m just sayin.
Exactly.
Difference between Denver and Indy, 2003: 7.6% + 17% HFA = 24.6% (in Indy's favor).
Difference between Denver and Indy, 2004: 15.1% + 17% HFA = 32.1% (in Indy's favor).
Difference between Denver and NE, 2005: 32.1% + 17% HFA = 49.1% (in Denver's favor).

Just sayin' that this season is a little bit different than the last two.

159
by Harry (not verified) :: Fri, 01/13/2006 - 12:17pm

Good analysis Kibbles. This is exactly why Denver-NE is such an interesting game from a DVOA perspective. Every traditional analysis looking at player-by-player matchups and strategy suggests NE has a very good shot, and this is why astute observers like Ron Jaworski think the Pats will win. DVOA clearly suggests the contrary, that Denver not only will win, but should win fairly easily. So if Denver wins big the game will be a huge validation for DVOA. If Denver loses, or even if they win by only a small margin, than it will be equally clear that DVOA is missing a key variable.

160
by b-man (not verified) :: Fri, 01/13/2006 - 12:26pm

So this is a big mismatch for DVOA. Denver should be expected to win by around 9pts. Has there ever been such a large disparity in the playoffs since DVOA has been tracked? Maybe last years Phi/Atl or the previous years Ten/Bal?

161
by spenceKarl (not verified) :: Fri, 01/13/2006 - 12:33pm

RE: 159 (Harry)

So if Denver wins big the game will be a huge validation for DVOA. If Denver loses, or even if they win by only a small margin, than it will be equally clear that DVOA is missing a key variable.

Again, you can't just subtract one team's DVOA from another to pick a winner. Individual matchups and a deeper look into how these teams have played overall and of late will give you a better sense of a possible outcome. Remember, of course, that any statistical formula is not a replacement for your own judgment, just a tool to use in analyzing performance.

162
by mactbone (not verified) :: Fri, 01/13/2006 - 12:44pm

Even if the Den/NE game ends up wildly different than what DVOA suggests that doesn't mean there's some missing variable. Well, it means that there aren't observable variables that can be included. Football is not just one team performing at the same level for 16 games against another performing the same way for 16 games. There are fluctuations and DVOA was built to predict those to an extent but the larger goal is to find out why teams win and what a team needs to do to win. DVOA has some predictive power but what it realy does is give you probabilities like the beginning season forecasts. DVOA (our interpretation of it) says that there's a high probability that Denver will win, not just Denver will win. You can take it to mean that Denver should win according to DVOA. However I don't think you can come away after that game and say that DVOA is fatally flawed if Denver doesn't win by 9.

163
by James, London (not verified) :: Fri, 01/13/2006 - 12:45pm

I'm rooting for either the Steelers OR Denver to win, I don't care which one (or even if both win.

I just don't want to go through Brady vs Manning 3. That would be horrible.

164
by spenceKarl (not verified) :: Fri, 01/13/2006 - 12:50pm

RE: 160 (b-man)

"So this is a big mismatch for DVOA."

If you want to compare DVOAs for the matchup, why not use the Weighted DVOAs listed at the top of the page. They at least include the Patriots performance vs Jacksonville and don't include the irrelevant portions of week 17 (Broncos/Chargers after Jake leaves, Pats/Dolphins after Brady leaves).

So 38.1% for the Broncos and 15.3% for the Pats. Still a decent gap, but not as large as using the regular season total DVOA.

165
by Al 45 (not verified) :: Fri, 01/13/2006 - 12:52pm

than it will be equally clear that DVOA is missing a key variable

Well I think that's already known to be true. If DVOA is supposed to adjust for the strength of opponent, then Aaron stating -- during his Jacksonville vs. NE preview writeup -- that "but all the excitement ignores that the opposition in these games wasn’t exactly top notch" means that even he doesn't trust that it 'accounts' for strength of the opponent.

That would be like a baseball guy stating that such and such a hitter has a VORP of 99.9, and then stating... but that's tempered by the fact that he plays at Coors Field. Since VORP is park adjustment it's ridiculous to mention where the guy plays.

Likewise, if DVOA is supposedly adjusted for the strength of the team you play on any given sunday, then stating that they were playing lower level competition is equally as ridiculous.

Hey, I could be wrong... but from my cursory glance at DVOA, and from what I've been told on these boards, DVOA does account for strength of schedule. Therefore I find it odd that the quality of the opponent has all of a sudden become a tempering tool.

166
by spenceKarl (not verified) :: Fri, 01/13/2006 - 1:06pm

RE: 165 (Al 45)

"but all the excitement ignores that the opposition in these games wasn't exactly top notch" means that even he doesn’t trust that it 'accounts' for strength of the opponent.

I think you're taking him out of context. In rereading the paragraph that quote came from, along with the preceding paragraph, I think Aaron was refering to all the national hype and attention the Pats began to recieve during their four-game win steak. And what he was pointing out in that quote, is "Hey, look who they played!". I think that quote is actually in support of what DVOA does for opponent adjustments.

167
by b-man (not verified) :: Fri, 01/13/2006 - 1:08pm

164: Not sure if it makes sense to use WDVOA since Denver didn't even play last week. Also, it says that those W17 plays are elimnated from DVOA in the above tables as well.

168
by jeff (not verified) :: Fri, 01/13/2006 - 1:08pm

I hesitate to get into this again but the DVOA could stand a new category which represents a delta DVOA which would be a positive or negative number based upon whether a team has shown a positive or negative change in its DVOA over it's recent history. It may not happen very often but occasionally there may be a team whose season long DVOA just does not represent it's improvement during the course of the season and what it's chances are in the playoffs. I suggest the number go back only 5 or 6 games but could be as many as 8. Does anyone else think this other purely statistical number would be an improvement to the DVOA system?

169
by spenceKarl (not verified) :: Fri, 01/13/2006 - 1:14pm

RE: 168 (jeff)

I'm confused by the difference in your suggested improvement and what's being done with Weighted DVOA.

170
by Al 45 (not verified) :: Fri, 01/13/2006 - 1:19pm

Gotta disagree.

His exact statement is:

"There is no doubt that New England’s defense was dominant during their four-game winning streak, but all the excitement ignores that the opposition in these games wasn’t exactly top notch."

It's the start of a new paragraph, and therefore a new thought. He isn't quoting anyone else, or making an observation on someone elses thought. He's stating that the patriots defense has been dominant, but it ignores that the opposition isn't exactly 'top notch'.

He even then goes on to a poor analysis of the defense by stating that (in regards to Hawkins and Hobbs):

"but all of these changes were in place when the Patriots were slaughtered by Kansas City six weeks ago. Defensive end Richard Seymour and linebacker Tedy Bruschi were back from injury and Ellis Hobbs and Artell Hawkins had moved into the starting secondary."

Two things here... one, it's obvious he's attributing the defenses 'appearance' to the quality of teams they faced (why else would you mention that 'all the changes' were in place during the KC game)?

Two, that's absolutely horrific analysis since, while the changes 'were' in place, many of them had JUST taken place.

Seymour had only been back for two games (it was his third game, and was still considered a good deal below 100%).

Bruschi was playing in his 5th game back which, when considering the situation, was basically the equivalent of playing in the opening game of the season (4 pre season games and your first regular game). Considering he had no preason practice, etc... I don't think it's a stretch to say that, during the KC game, he was no where near the ability he was at by the end of the season.

Artrell Hawkins was playing in his second game for the Patriots, at a position he had never played before in his life. Does Aaron really believe that a player can change from CB to safety and know your assignments perfectly in 2 weeks... all at the same time while attempting to learn the Patriots playbook? Come on now.

Now, either Aaron is over compensating with his analysis because he doesn't want people to think he's a Patriots schill... or he doesn't trust the formula enough to truly account for the strength of the team they're playing.

I'm not sure what I'm more troubled by... Aaron tempering the Patriots defensive performance with the fact the teams weren't 'top notch' while DVOA already accounts for that... or tossing aside the 'personel changes' because they were already in place in KC.

171
by jeff (not verified) :: Fri, 01/13/2006 - 1:26pm

Weighted DVOA may be fine because it clearly causes teams that are playing better to show an improvement in their season long calculated DVOA. I'm not convinced that weighted DVOA is enough in the case where a team that plays poorly through the first 8 games becomes a monster for the last 8 games. Would it help us or be of interest to us if we knew what teams have made the most improvement and/or how they compare to all teams for a shorter period of the season. Why is the Denver/ NE game generating so much interest? Because we know that NE has become the beast from the northeast again. When we look at the DVOA as presently configured do we see this?

172
by spenceKarl (not verified) :: Fri, 01/13/2006 - 1:38pm

RE: 170 (Al 45)

I think you're still misinterpreting the quote:

"There is no doubt that New England’s defense was dominant during their four-game winning streak, but all the excitement ignores that the opposition in these games wasn’t exactly top notch."

He didn't say DVOA ignores the opposition, he said "all the excitement" does. It's simple really: VOA says the wins were dominant, while DVOA considers and adjusts for the opposition. National excitement says the wins were dominant, but ignores the opposition.

173
by spenceKarl (not verified) :: Fri, 01/13/2006 - 1:56pm

RE: 171 (jeff)

Do you have any suggestions of how to improve the accuracy of Weighted DVOA for all teams and not just the Patriots?

174
by Al 45 (not verified) :: Fri, 01/13/2006 - 2:01pm

I think you’re still misinterpreting the quote:

I'm not misinterpreting at all. He goes on to explain that against KC (a team that's considered near 'top notch' ability, at least offensively) all these supposed changes were 'in place'.

If he was simply talking about everyone elses perceptions ignoring the conversation, then why did he go INTO DETAIL about the personel changes being in place the last time they faced a really good offensive team?

Sorry, but I really think it's you that's trying to read the quote the way you want. I'm firmly within the context of what he wrote.

175
by B (not verified) :: Fri, 01/13/2006 - 2:03pm

If the Patriots fall in Denver, I blame the FO power rankings message board curse, and specifically the postings from jeffy and AL 45. There's blood on your hands, and I expect an apology if I'm forced to watch my favorite team go down in flames.

176
by hwc (not verified) :: Fri, 01/13/2006 - 2:05pm

"Watching the Baltimore film to figure out how to beat Denver would be about as helpful as watching last season’s NE/Miami film to figure out how to beat the Patriots."

As a knowledgeable and frequent observer of Pats games, if I were advising a coach about what game film to study to disrupt Brady, my first suggestion would be to study all Miami/Pats games going back to Brady's rookie year. I would say that Miami's defense has consistently been the defense that gives Brady's offense the most trouble. The win/loss record or DVOA or Sagarin rankings between the two teams are largely irrelevant to that determination. It's been easy to see from watching the games that Miami's defense causes matchup problems for Brady.

On an unrelated note: I thought it was very interesting yesterday when Belichick was asked where he had spent the most time coaching this season relative to past seasons. His answer was "defensive front seven". When asked why, he said, "I thought spending more time with those guys would be the most productive use of my time this season." I find that interesting because, IMO, the Pats defensive front seven has been suffocating over the last half of the season and is probably the best in the NFL at this moment in time. Shanahan highlighted that in his press coverence this week, noting that the Pats allowed 2.9 yards per carry and 164 yards of total offense per game from December on.

177
by Al 45 (not verified) :: Fri, 01/13/2006 - 3:15pm

I would say that Miami’s defense has consistently been the defense that gives Brady’s offense the most trouble.

Absolutely they have. Although, they've had one of the better defenses, period, over that time.

Not to mention they had a PHENOMENAL secondary that were really terrific at blanketing NE receivers.

Personally, I think Brady's trouble had more to do with the athleticism and overall talent of the Miami defense, and in particular the secondary, than it did with any particular scheme.

178
by jeff (not verified) :: Fri, 01/13/2006 - 3:46pm

Could you back out of your role as the no. 1 DVOA apologist and tell me what is wrong with what I've suggested instead. Really, what I'm suggesting would apply to Miami and Washington this year and maybe alot of others in prior years. Maybe we could predict upsets better. Maybe you aren't that interested in seeing a better DVOA.

179
by chris (not verified) :: Fri, 01/13/2006 - 5:22pm

I've been a fan of DVOA for a couple years now but with the Pats below where most observers think they belong, it looks like DVOA needs seomthing more. One enhancement would be to run a DVOA for multiple subsets of games, then come up with a number describing which teams had the most variable rankings when looking at all possible DVOAs. This might give a sense of how "solid" any particular DVOA ranking is.

In the case of the Patriots it looks like their poor play earlier is really hurting them. I don't distrust what the DVOA says about their proven record, but I think the Pats are more likely to be better than their DVOA rating, whereas the Colts are very likely to be as good as their ranking. So the DVOA is a better predictor in the case of the Colts versus the Pats. My suggestion is to come up with a confidence factor to accompany the DVOA.

180
by chris (not verified) :: Fri, 01/13/2006 - 5:30pm

PS. You might think adding a confidence factor complicates the interpretation of the DVOA. Well, sometimes life IS complicated.

181
by B (not verified) :: Fri, 01/13/2006 - 5:59pm

179: That sounds like the variance stat that's normally included on the second chart, seems to be missing this week. If I remember correctly, however, both NE and Denver have pretty high variance. However, if memory serves, Denver's bad games have been on the road, whereas for NE they've been bad both at home and on the road, although most bad games were early in the season (KC being the obvious exception). This is all kind of a long-winded way of saying I think both NE and Den will bring thier A game on Saturday. The question is which A game will be better? I think the edge has to go to Denver. Don't get me wrong, I think last years Patriots team was better than this year's Denver squad, but the Patriots arn't that team anymore. The biggest reason they're not? No Rodney Harrison. Atrell Hawkins is a perfectly servicable strong saftey, but you can't replace a top-5 saftey like Harrison with a guy who never played the position before, and not expect some drop-off. Did anybody read the article on how the 2005 USC team is over-rated because of the greatness of the 2004 squad? Same basic concept here.

182
by thad (not verified) :: Fri, 01/13/2006 - 7:33pm

No Kibbles that was not what i was saying. I did not mention sack differential, only that 28 sacks is a very low number for a playoff team. I am saying that compared to other teams that have advanced in the playoffs, Denver is very poor in this regard.

183
by Christian X (not verified) :: Wed, 01/18/2006 - 2:24am

Aaron Schatz
is the biggest idiot of all time... how do you destroy a team and not make a judgement call that the steelers aren't more highly ranked... Moreover, your ranking system has been WRONG ALL YEAR... Schatz take some humility and rank teams on wins, heart and stats together... not some rediculous ranking that has the colts, who lost, in first... YOU'RE a real AMERICAN IDIOT...