Writers of Pro Football Prospectus 2008

02 Dec 2007

ESPN: Best-Ever Teams Have No Title Guarantees

This week's ESPN MNF column looks at two things: How the Patriots compare to the teams covered in Eddie Epstein's book Dominance, and how their near-loss to Philadelphia compares to other near-losses by the NFL's greatest teams. Suffice it to say, one or two close wins doesn't say much about the Pats' overall greatness.

Posted by: Aaron Schatz on 02 Dec 2007

38 comments, Last at 07 Dec 2007, 1:31am by DP

Comments

1
by register_allocation (not verified) :: Mon, 12/03/2007 - 12:41am

Epstein didn't pick the 1996 Packers as the best team ever, even though they had the highest API of any Super Bowl champion
Bah. If you're going to use an objective means to rank teams, it's a cheat to bail on it when it tells you something you don't like.

2
by Shot 'n Freud (not verified) :: Mon, 12/03/2007 - 12:45am

Quite a revealing article title. And all this time, I thought I could stop watching at the end of the regular season.

#1 - excellent point.

3
by Lance (not verified) :: Mon, 12/03/2007 - 1:01am

I'm also not convinced by a recitation of stats. The 1995 Packers were more or less the same team that had struggled against Dallas and the 49ers in the previous few years. My guess is that if you imported the 1993 Cowboys or 1994 49ers and played them against the 1996 Packers, the former two win.

4
by Lance (not verified) :: Mon, 12/03/2007 - 1:02am

That should be *1996 Packers

5
by Jake (not verified) :: Mon, 12/03/2007 - 1:06am

#3 I'm not convinced by your guess.

6
by Fizzman (not verified) :: Mon, 12/03/2007 - 1:12am

#1:

But API can only cover the regular season (since most teams aren't in the playoffs), and, as stated in the article, the Bears were selected as the top team based on their post-season dominance. That's a lot different from simply "I like the Bears better, so I'm just going to say they're the best."

7
by register_allocation (not verified) :: Mon, 12/03/2007 - 1:29am

#6 Surely his technique had to deal with bye weeks. Why not deal with teams that don't make it to the postseason as if they were on a bye?

8
by register_allocation (not verified) :: Mon, 12/03/2007 - 1:50am

#3 -- The Packers never struggled against the 49ers. The last two times the 49ers beat the packers were 1998 (divisional playoffs), and 1990.

9
by navin (not verified) :: Mon, 12/03/2007 - 2:04am

I still hate Don Beebe.

10
by Temo (not verified) :: Mon, 12/03/2007 - 2:13am

His statistic measures dominance against that specific current-year competition. I really can't determine if there's any objective way to measure dominance of one team in one era to another team in another era.

11
by Fizzman (not verified) :: Mon, 12/03/2007 - 2:15am

#7
There are a number of good reasons why the regular season and playoffs are virtually never treated as the same data set statistically, some of which are relevant to this particular question (and some are not).
Any statistic that is counting/cumulative is obviously skewed by the extra games; you could look at per-game stats, but that raises the next objection. Playoffs are designed so that the level of competition (and presumably difficulty) is meant to increase with each round, a factor that clearly does NOT operate in the regular season.
Well, I started this off as if I were going to have a long list, but I don't at this late hour. Rebut away.

12
by CA (not verified) :: Mon, 12/03/2007 - 2:55am

Re: 1 Bah. If you’re going to use an objective means to rank teams, it’s a cheat to bail on it when it tells you something you don’t like.

Based on that logic, he should have picked the 2001 Rams. Why bail on the "best" team simply because you don't like that your objective means told you that a team that didn't win the Super Bowl was the best? Also, see #6.

13
by register_allocation (not verified) :: Mon, 12/03/2007 - 4:40am

Based on that logic, he should have picked the 2001 Rams.
OK. I have no problem with picking a team other than the 1996 Packers. I have a problem picking another team because it is "more dominant in the playoffs" with no objective metric offered to justify the conclusion.

Personally, I would argue that the 1996 49ers, Panthers, and Patriots (the teams the Packers beat to win the Superbowl) were clearly better than the 1985 Giants, Rams, and Patriots (the teams the Bears beat).

But without some kind of objective metric to use for comparison, there's no way to refute my claim other than "did not/did too".

14
by register_allocation (not verified) :: Mon, 12/03/2007 - 4:52am

Any statistic that is counting/cumulative is obviously skewed by the extra games
By this argument, shouldn't the 1972 Dolphins and 1962 Packers be excluded because they didn't play a 16 game regular season?

And yeah, the post season is not the same as the regular season, but does that mean the only way to compare post season performances is to use the "ooh shiney" method?

15
by Jsaon (not verified) :: Mon, 12/03/2007 - 6:09am

2 Crazy things about the 1996 Packers:

-They were similar to this years Colts in that in the middle of the year their wr's were decimated by injury (they ended up as the most dominat team desite the fact their offense was crippled for multiple games by injuries)

-They never played a single game at full strength. Rison and Brooks never played together. In a theoretical game they would have both on their team making them even better

16
by Becephalus (not verified) :: Mon, 12/03/2007 - 8:12am

Why would they have both in a theoretical game? We are talking about the actual Packers, NOT the theoretical Packers.

Should the Lions have Barry Sanders this season theoretically as he may not have retired yet?

17
by Waverly (not verified) :: Mon, 12/03/2007 - 12:27pm

I don't think Epstein could have chosen the 2001 Rams, no matter how objective he wanted to be. He wrote those articles before the 2001 season.

18
by Otis Taylor 89 (not verified) :: Mon, 12/03/2007 - 12:31pm

#15

Every team, including the great ones, have had injury problems. The Colts had theirs during a weak spot in their schedule and have been greatly overrated. Even the 1972 Dolphins lost their starting QB for most of the season. You can justifiably cry a river when Manning goes down.

19
by B (not verified) :: Mon, 12/03/2007 - 12:37pm

16: The Theoretical Packers have nothing on Theoretical Rob Johnson.

20
by John (not verified) :: Mon, 12/03/2007 - 12:44pm

#18:

The Colts had theirs during a weak spot in their schedule and have been greatly overrated.

Very true. Those Patriots the Colts played without their #1 wide receiver and left tackle clearly were soft. What a weak spot in the schedule!

I'm not going to make any comments one way or another about whether the Colts have any room to use injuries as an excuse, but your assessment is a little awry.

21
by Boots Day (not verified) :: Mon, 12/03/2007 - 2:14pm

Maybe I need to get Epstein's book, but I'd sure like to see more about how those Packers got named the best team of all time. They finished "just" 13-3, and it's not like they only lost narrowly to very good teams. They got beat by the 9-7 Vikings 30-21, by the 9-7 Chiefs 27-20, and by the 10-6 Cowboys 21-6. They had a massively impressive point differential, outscoring their opponents 456-210 (leading the NFL in both categories), but part of that was because they poured it on at the end of the season, outscoring the opposition by 110-19 in their final three games, after the regular season had been decided.

Sure, they were a great team, but the best ever? I doubt even Mike Holmgren thinks that.

22
by Otis Taylor 89 (not verified) :: Mon, 12/03/2007 - 2:46pm

#18

The Colts still had two Pro Bowl receivers available for the Pats game, something most teams don't have on their best days. They then played a .500 SD, KC at home against a QB in his 1st start and ATL - if that isn't soft you probably don't think puppy crap isn't either.
Of course, the 1972 Dolphins played an entire year of only teams like that, which is one reason they wouldn't even be in my top 15 teams of all time (the 1973 team would, however).

23
by Richie (not verified) :: Mon, 12/03/2007 - 2:53pm

#22, why would you include the 1973 Dolphins, but not the 1972 Dolphins even though they are almost the same team?

24
by adwred (not verified) :: Mon, 12/03/2007 - 3:42pm

I would say, using statistics and my love of all things Packers, that the 1996 Packers were indeed the best NFL team ever.

25
by Otis Taylor 89 (not verified) :: Mon, 12/03/2007 - 3:44pm

#23
1972 didn't beat a playoff team during the regular season - of course they didn't play one either. The 1973 team played against 3 playoff teams during the regular season and beat teams in the playoffs that were superior to the 1972 playoff teams. Finally, I thought their defense was better in 1973 against better comp with everyone gunning for the Champ.

26
by Yaguar (not verified) :: Mon, 12/03/2007 - 4:07pm

22: "The Colts still had two Pro Bowl receivers available for the Pats game, something most teams don’t have on their best days."

No they didn't. This is the second time you've been flat-out wrong about something in as many posts, so I humbly suggest that you probably don't know what you're talking about.

27
by Otis Taylor 89 (not verified) :: Mon, 12/03/2007 - 4:21pm

#26
Opps, you are correct, Dallas Clark, although one of the top receiving TE's in the league, has not been to a Pro Bowl.

Please kick this ass. I'm not asking, I'm telling you. Kick my ass....

28
by TireSlasher (not verified) :: Mon, 12/03/2007 - 5:39pm

How the 1989 S.F. 49ers are not included in this conversation is baffling. Though they lost two regular season games, they lost those games by a total of 4 points, one of which they played a significant amount of the game without Montana. Their API is probably a bit lower because they often pulled starters late in games and allowed some garbage time TD'S. But when it mattered, in the playoffs, they were absolutely dominant. I think they were clearly the best team of the past 25 years. Better than the '85 Bears, '96 Packers and those great Dallas teams of the 1990's.

29
by Scott C. (not verified) :: Mon, 12/03/2007 - 5:49pm

#1, #8, #14, #15 ...

You clearly don't understand what the stat is measuring.

Lets start over -- it is measuring the regular season dominance of a team compared to all the teams THAT YEAR, and it strongly discounts wins against wimpy teams.

So, its a stat that must have balance among all the teams in its pool (everyone's opponent win-loss record has the same number of games). Bye weeks don't matter because the numbers are at the end of the season when all teams have the same # of games, and you can work with 12 and 14 game seasons just fine since the stat works only within a single season.
You can work with the pats now, because all bye weeks are done this year.

What it says, is that by its measurements the 2001 Rams had the most impressive regular season. And so far the pats are beating that pace.

Post season dominance is not measured, and to do so you would need to use a different tool.

The "Most dominating post season" list would surely be different. Lots of SB champs have gotten there with several narrow victories, while others have just smashed through the playoffs.

30
by Yaguar (not verified) :: Mon, 12/03/2007 - 7:10pm

27: Not necessary. I'm not an angry troll or anything, I just think you don't know what you're talking about.

31
by Otis Taylor 89 (not verified) :: Mon, 12/03/2007 - 7:32pm

#30

Sorry, it's a "Spinal Tap" quote, no offense taken.

32
by Jsaon (not verified) :: Mon, 12/03/2007 - 8:33pm

"Why would they have both in a theoretical game?"

Because both players were on their team that year.

"16: The Theoretical Packers have nothing on Theoretical Rob Johnson. "

That is just a nonsensical response. Anyone should be able to tell a difference between having all of your players healthy vs pretending a qb will perform better than he actually did.

"#15

Every team, including the great ones, have had injury problems."

Talk about oversimplification. To contend that every great team had exactly the same injury problems is not only impossible but ridiculous.

33
by register_allocation (not verified) :: Tue, 12/04/2007 - 3:58am

You clearly don’t understand what the stat is measuring.
Go back and re-read what I wrote. You have completely missed my point.

34
by Pat (not verified) :: Wed, 12/05/2007 - 12:52am

I have a problem picking another team because it is “more dominant in the playoffs� with no objective metric offered to justify the conclusion.

Um. He does. He ranked the teams based on a combination of record (15-1, ranked #2), Power Index (0.725, ranked #2) and performance against winning teams (PF 180, PA 71, ranked #1).

35
by register_allocation (not verified) :: Wed, 12/05/2007 - 1:18am

He ranked the teams based on a combination of record (15-1, ranked #2), Power Index (0.725, ranked #2) and performance against winning teams (PF 180, PA 71, ranked #1).
Odd how none of that information seems to be in the actual article.

In the article, there are several paragraphs about the "Adjusted Power Index", followed by the statement
Epstein didn't pick the 1996 Packers as the best team ever, even though they had the highest API of any Super Bowl champion. [...] Epstein picked the 1985 Bears as the best team ever, largely because they were even more dominant in the postseason than in the regular season.
I haven't read the book, so I can't say anything about Epstein's actual methods. But the above paragraph, at best, falls way short of justifying picking the '85 Bears over the '96 Packers. Just saying "more dominant in the postseason" is a cop-out.

If that isn't sufficient explanation of my point, I'm just going to give up; I don't see how I could make it any clearer.

36
by Nathan (not verified) :: Wed, 12/05/2007 - 8:34pm

Another vote for the 89 Niners.

37
by Scott (not verified) :: Thu, 12/06/2007 - 3:51am

It was a good article on December 1, but at this point I don't think the 07 Pats are any better than the 05 Colts, and that team didn't get out of the first round. 3 of the last 4 games for NE have been nail-biters that easily could have (and really should have) gone the other way.

Indy had basically zero close calls on their way to 13-0. The Pats look like a 12-0 team that's only fortunate they're not 9-3. The "greatest team of all time" should not sweat out 3 pt victories against teams with losing records and bad back up QB's.

I've been saying since late September they cannot sustain this type of play all season long. Usually the Pats peak late in a season, this year it looks like the opposite.

38
by DP (not verified) :: Fri, 12/07/2007 - 1:31am

The '89 49ers are particularly noted for their dominant postseason, in which they outscored their three opponents by a total of 100 points, 126-26.