Now, my season to date ratings do give a little more weight to recent games, but certainly don't ignore the earlier games, which in some cases, like Denver, should be. Or in the case of New Orleans where they've gotten very hot late, continuing to include the earlier games can temper how high they get in the ratings. Or Green Bay built up a huge early lead and tailed off a bit late in the year, but perhaps they aren't really playing the best right now. Thus, the full season ratings, while pretty accurate, may not be as accurate as we'd like.
Ideally, one looks at the performance chart for a team and identifies trends and selects which games to consider including any mitigating factors. When this was done for Pittsburgh at Denver, we saw how one could have easily picked the upset.
As a quick way to look at the latter half of the season though, I've run my ratings for the last 8 weeks of the regular season plus the wildcard round of the playoffs. Doing this, New Orleans moves to #1 and New England to #2 ahead of #3 Green Bay. The Giants also move up to #6.
So, doing the Superbowl chances using these ratings, we get a slightly different picture.
Green Bay is no longer the favorite even with their home field advantage, not even in their own conference. New England is now the far and away favorite as they have the easier road. But despite their improvement, given that some future/potential opponents also improved, the Giants chances have gone down. Denver also had improved a bit but now has to go to an improved New England so their chances are down.
The NFC appears to have the stronger teams, 3 of the top 4 chances belonging to that conference and the conference itself having a 56% chance of winning the Superbowl.
For comparison, here is the full season ratings Superbowl chances.