Hughes, you're thinking about it all wrong.  You're thinking about it in the context for one team.  While your math is probably right for one team, when it comes to the conference, you're missing a BIG piece of the puzzle.

Go check out Rails' thread.  

#Earthisnotflat

Actually, I do math for a living .. SO .. I will do some math on Rails thread right now.  My thinking is not at all wrong.  Which is why when the ACC when 119-1 against crappy SIM teams, they had the same RPI (or lower) than when they scheduled the other way.  Going to do math there now.

Pretty sure a LOT of people who play this "do math for a living"

Hughes - you are not building in the fact that your next 16 games are vs teams that are all 10 games over .500 from non-conf

Posted by mullycj on 3/16/2015 9:44:00 PM (view original):
Hughes - you are not building in the fact that your next 16 games are vs teams that are all 10 games over .500 from non-conf

This.  1000 times this.

Posted by hughesjr on 3/16/2015 8:50:00 PM (view original):
Lets do a couple of comparisons:

I go 5-5 against teams that are 10-0 and their opponents are 5-5 ... so my RPI is:  .500/4 + 1.000/2 +.500/4 = 0.750 (quite good)

I go 10-0 against teams that are 2-8 and whose opponents are 5-5 .. so my RPI is 1.00/4 + .125/2 + .500/4 = 0.438 (crappy)

I go 0-10 against teams that are 10-0 and whose opponents are 5-5 .. so my RPI is  0/4 +1.000/2 + .500/4 = 0.625 .. significantly better than going 10-0 against crappy SIMs

Remember, the biggest contributor to RPI (50% of it) is your opponent's winning percentage)

True. . . but that's your RPI, not your conference RPI. For the other 11 teams in the conference, your record goes in the "opponent's win%" category (which is worth 50%) and your opponents' win% is just 25%. So going 10-0 against a weak schedule is more beneficial to your conference RPI than going 0-10 against a tough schedule. Now going 10-0 against a weak schedule may not be as good as going 7-3 against a good schedule, but that's not really what's at issue here. For your conference RPI, your wins are worth more than your opponents. 

I think in real life the home-road factor is 1.4 to 0.6, so as an added bonus, if you go 10-0 against poor teams all on the road you're 14-0 in RPI. And if all at home you'd be only 6-0 in RPI. If it's 1.2 then 12-0.

Anyway, your conference mates will love putting your 14-0 record towards 50% of their RPI, while putting your opponents' combined 0-100 (random 5 home and 5 road games scheduled for Sim) record towards only 25% of their RPI.

Your RPI will be awful, but your conference mates will really benefit.

Side note is that the Projection Report seems to look at Top-100 or Top-50 wins. So i think it is sometimes worth the gamble to play 2 or 3 great teams, because if you steal a win then it's great for your resume, all else being equal.

Glad that pretty much everyone else recognized this.

I'd also say that 10-0 against all 0-10 teams is an extreme case and would hardly ever happen.  More likely, even if you didn't pay attention to which SIMs you're scheduling, you're likely to have a few with 1 or 2 wins, and several others with at least 3-5, maybe some over .500.