Today's optimistic Vandermause Packers column from GBPG

[RashanGary]

I have numbers a page or two back, Vince. You're right about the colts though. I was getting my stat from nfl.com/history and then the last two years I switched. In the switch I made a mistake.


I still ended up taking out the highest two. The Colts would be one of those highest two. When doing that, it adds in 14 from when the raiders won and bumps the averages up to about even - 3.5. If you take out the high two on yours it ends up at 3.5 as well


I didn't want unusual exceptions throwing off the data. Ultimately it's how a defense/STs play in the playoffs, not the regular season that counts, but I don't know where to get that stat and over 27 years, after removing unusual data, I think a pretty reliable theme is gained from the regular season stat. The theme is that great QB's or good QB's both require the same thing - (a great surrounding cast) I don't think it's a stretch to say people far overrate the importance of the QB because they see a couple QB's win many and they assume it was the QB. Reality is that they need the same surrounding cast as the 1 time winners and maybe a bunch of guys who never won are better than the guys who did, but never had the surroudning team or health bounce the right way in the post season. It's a team game with team wins and I think Brett Favre was on to something when he says the QB takes more blame than he deserves in the bad times and gets more credit than he deserves in the good times. It's the team.

[vince]

I have numbers a page or two back, Vince. You're right about the colts though. I was getting my stat from nfl.com/history and then the last two years I switched. In the switch I made a mistake.


I still ended up taking out the highest two. The Colts would be one of those highest two. When doing that, it adds in 14 from when the raiders won and bumps the averages up to about even - 3.5. If you take out the high two on yours it ends up at 3.5 as well


The average defense in points allowed over the last 27 years is still 3.5 more or less after removing unusually high and low data. Go back adn look at the data. You'll see I removed the high two and low to from the 1 time winners and multi winners just to disregard the unusual data and focus on the typical situation that is required to win the SB. I didn't want unusual exceptions throwing off the data. The odds are you need a top 3 or 4 defense to win a SB, but circumstances arise where a defense plays bad in the reg season and turns it on in the playoffs. Ultimately it's how a defense plays int he playoffs, not the regular season that counts, but I don't know where to get that stat and over 27 years, after removing unusual data, I think a pretty reliable theme is gained. The theme is that great QB's or good QB's both require the same thing - (a great surrounding cast)

OK. However, that's where statistics get misused. In this case in particular, you don't (shouldn't) ignore ANY instances when averaging the results. That's the point of averaging - to come to a reasonable generalization about a set of statistics. Ignoring any outlyers, particularly in a set of ranked numbers where there is a limitation to the values (1 to 32), creates a bias which, in most cases including this one, can lead to faulty conclusions. Those teams won the Super Bowl and should be included in any pool of Super Bowl winners - plain and simple.

Averaging a lot of low numbers with one or two high numbers effectively gives much greater (appropriately) weight to the low numbers - without ignoring the instances that tend to disprove the conclusion.

Using your numbers would lead to the conclusion that a Top 3 defense is practically required to win the Super Bowl, particularly for one-time winners. I'd say that's why Harv pointed out how "telling" those (skewed) results are. Only about half the winners accomplished that though. Half didn't. A conclusion where there's only a 50/50 (or 60/40) chance of being right is not a good basis upon which to draw any conclusion.

The general theme, although you exaggerated it, obviously holds though in both instances.

[vince]

I went back and looked at your original posts, and after stating the statistics, you drew a conclusion that the Packers aren't likely to win the Super Bowl by asking this question...

Do we have a #2 defense?

By concluding that a #2 defense is required to win the Super Bowl is just not an accurate conclusion. A #2 or better defense was only achieved by 37% of the last 15 Super Bowl champions. An overwhelming percentage of Super Bowl Champions had lower than a #2 defense...

Obviously, it's difficult to win the Super Bowl, and I'm not arguing that the Packers will win it this year. I'm just pointing out that skewing results leads to faulty conclusions, and that the Packers could very well win the Super Bowl, but they will likely need a top 10 defense. Last year, they were 11th I believe.

[RashanGary]

Vince, I really wasn't trying to extrapolate all kinds of crazy assumptions out of it. I was trying to find a common tendency of SB winners - that's really it.

I think the the theme is that good and great QB's alike don't do it alone and shouldn't get the credit or blame that they get by some people here. That's really the only conclusion I drew from any of this. Maybe the stat seemed to do one thing or another, but I explained what I was doing the whole way and didn't extraoplate it in any specific, crazy way. As it relates to this team, I didn't specifically say it, but I also think that losing Brett is not nearly as important as many here think it is. IF the team is good and the QB is OK, they have a chance. Further, my problem with the Packers isn't the QB just yet, it's the defesne.

If our defense is top 10 (and I think it is), 11% of teams have won the SB with that type of defense. Our chances aren't very good. That's my point.

[texaspackerbacker]

I honestly don't have much faith in all this statistics stuff--there are always exceptions--like the Colts a couple of years ago, and occasionally teams like the Giants last year who get lucky and/or peak at the right time.

Just for curiosity, though, where did the Packers rank last season in points allowed and yardage allowed?

[vince]

Vince, I really wasn't trying to extrapolate all kinds of crazy assumptions out of it. I was trying to find a common tendency of SB winners - that's really it.

I think the the theme is that good and great QB's alike don't do it alone and shouldn't get the credit or blame that they get by some people here. That's really the only conclusion I drew from any of this. Maybe the stat seemed to do one thing or another, but I explained what I was doing the whole way and didn't extraoplate it in any specific, crazy way. As it relates to this team, I didn't specifically say it, but I also think that losing Brett is not nearly as important as many here think it is. IF the team is good and the QB is OK, they have a chance. Further, my problem with the Packers isn't the QB just yet, it's the defesne.

If our defense is top 10 (and I think it is), 11% of teams have won the SB with that type of defense. Our chances aren't very good. That's my point.

Agreed with all that JH. I'm not trying to beat you up or anything - just make clear that most Super Bowl winners don't have Top 2 or 3 defenses, even though that improves one's chances...

Last year, the Packers finished with the 11th best D in total yards and 6th in total points.