Agreed, but it got me thinking -- how do the numbers all add up comparing division games to non-division games? Here's out it shakes out:
2002:
Overall record of all division winners: 87-40-1
Record for games outside the division: 48-31-1
Record for intra-divisional games: 39-9
2003:
Overall record of all division winners: 94-34
Record for games outside the division: 57-23
Record for intra-divisional games: 37-11
2004:
Overall record of all division winners: 96-32
Record for games outside the division: 58-22
Record for intra-divisional games: 38-10
2005:
Overall record of all division winners: 94-34
Record for games outside the division: 53-27
Record for intra-divisional games: 41-7
2006:
Overall record of all division winners: 93-35
Record for games outside the division: 59-21
Record for intra-divisional games: 34-14
2007:
Overall record of all division winners: 95-33
Record for games outside the division: 56-24
Record for intra-divisional games: 39-9
2008:
Overall record of all division winners: 87-41
Record for games outside the division: 50-30
Record for intra-divisional games: 37-11
Leaving out the tie (because it makes my math a lot easier), that equates to the following totals:
Overall record of all division winners: 646-249, .722 winning %
Record for games outside the division: 381-178, .682 winning %
Record for intra-divisional games: 265-71, .789 winning %
So, for at least the last seven years, the division winner's records have been better (%-wise) within the division than outside of the division, by a significant margin.
It's still just a correlation, not causal, but here's my guess as to the reasons for this:
1) Better coaching -- good teams generally have good coaches. Good coaches generally prepare well, and will be better prepared than an average coach for a second meeting between two teams. Since you play all divisional foes two times in the regular season, and no other teams twice, my guess is that the division-winning teams have a good record in the second games against the divisional opponents (anyone want to take on tackling that research?). At the very least, the coaches' familiarity with divisional foes from previous years probably impacts this.;
2) Strength of schedule. Though worst to first has happened a little bit, it's not all that common. Thus, a division winner is likely to have been a good team the year before. Because good teams from the year before are slated to face the other good teams from the year before (in the two strength of schedule determined games on every team's schedule), it is likely that strength of schedule for these teams may impact the out of division records. Even though this only effects 2 games per team per year, that would be a total of 112 games in this exercise (2 SoS games per year X 8 division winners per year X 7 years of records). My guess is that the division winner's records in these games is lower than their overall record (again, anyone up for that research?).
Admittedly, there's nothing really useful here -- just thought the numbers ended up coming out interesting. My guesses above are pure speculation, but I can't think of any other reasons for the difference.