- Joined
- Sep 17, 2009
- Messages
- 12,160
- Reaction score
- 12,871
I enjoy seeing these kind of posts. I'm kind of in the same place.
I've been going back and forth on how to get an accurate SOS, using the records or rating at the time they play or by adjusting the teams entire season as teams change. My thinking on adjusting the entire season is that in the long run it could be closer to all teams' true strength if I use the latest rankings for all weeks.
I think if you're going to do a quick and dirty analysis, that's the best way. In the end, you'd just have to accept the criticism that the final rankings don't account for momentum swings, and since that's a very subjective measure, it's a criticism I could live with I think.
I tried fiddling with assigning weights to games played versus games to come and over the course of the entire season, and I got close to a formula but nothing that works in all scenarios. It's definitely a tough one and one that would occupy quite a bit of time in trying to figure out. It's possible, though.
Some of the differences are large. Using one method the Browns are tied for 21st and the other they're tied for 1st.
Minny goes from 2nd to 10th.
Yeah, I'm almost not surprised because of how subjective and fickle those rankings can be. The achilles heel might just be those qualitative numbers which, as we know, don't always make much sense.
Good post.












