Ok, my updated analysis:
What I did was compare the rankings of the SB winners since 1991 in the following categories:
QB rating; Rush yards; Rush attempts; Off 3rd down efficiency; Def yards allowed, Def Pts allowed, Def 3rd down efficiency, and Turn/over differential (the combo stat--part offensive, part defensive).
To summarize the results:
Top 3 categories were Def pts allowed, QB rating, and Turnover Diff (with a little more than 1 ranking difference between the 3).
Rush attempts, Off 3rd down efficiency, and Def yards allowed were all right behind T/O differential. Rush yardage and Def 3rd down efficiency were bringing up the rear, well behind the others.
Here's how the SB winners ranked in those categories:
Rush 3rd DEF 3rd TO
QB Rating Yds Att Eff Yds Pts Eff Diff
2010 3 Rodgers 24 20 8 5 2 9 4
2009 1 Brees 7 6 6 25 20 14 3
2008 24 Roethlisberger 23 9 14 1 1 1 11
2007 25 E.Manning 4 8 12 7 17 5 26
2006 1 P. Manning 18 18 1 21 23 32 7
2005 3 Roethlisberger 5 1 23 4 3 20 10
2004 9 Brady 7 5 5 9 2 21 8
2003 10 Brady 27 12 15 7 1 7 2
2002 3 Johnson 27 23 26 1 1 3 1
2001 6 Brady 13 8 8 24 6 15 9
2000 20 Dilfer 5 5 11 2 1 5 1
1999 1 Warner 5 15 1 6 4 10 10
1998 5 Elway 2 2 6 11 8 21 6
1997 7 Elway 4 6 3 5 6 1 5
1996 2 Favre 11 14 2 1 1 2 2
1995 3 Aikman 2 4 6 9 3 26 13
1994 1 Young 6 5 1 8 6 16 3
1993 2 Aikman 2 6 7 10 2 15 10
1992 3 Aikman 5 4 6 1 5 1 9
1991 2 Rypien 7 1 1 3 2 6 1
131 204 172 162 160 114 230 141
6.55 10.2 8.6 8.1 8 5.7 11.5 7.05
OK, looking at the Super Bowl winners for the last 20 years, heres the conclusions that Im drawing:
1) SB winners needed to be multi-faceted, not one-dimensional. They needed to be good/toward the top of the league in multiple categories. Almost all of the SB winners were good in most of the categories, potentially ranking in the lower half of the league in 1-2 areas at most.
2) Being deficient in one area could be compensated for by being elite in another category, with an elite Defense (to include ball-hawking TOs) & elite QB being the most important categories. Teams that were in the lower half in more than 1-2 categories (2006 Colts, 2002 TB) were able to compensate by having an elite D or QB.
3) The running game was not irrelevant. Neither the 2007 Giants or the 2000 Ravens had elite QBs, but their running game compensated. The 2007 Giants ranked in the lower half of the league in QB, Defense, and TO ratio (ranking in the lower half of the league in all three categories), but their running game compensated.
4) An elite D wasnt absolutely essential (2009 Saints, 2006 Colts, 2007 NYGthough the Giants D did pick it up big time at the end of the year & through the playoffs), as long as you had an elite QB (Saints/Colts) or a good running game (Giants).
So, having a balanced team, including balanced offense, improves your odds of winning the SB.
Having a good running game is not irrelevantit improves your odds of winning the SB.