What would happen if ten coaches got together in a conference and all scheduled the same 10 noncon opponents?
Now, presuming no interference, they schedule 10 sims that they can all beat and there are no upsets. Each team wins all ten games.

What would happen to these teams' SOS?
What would happen to these teams' RPI/Projection report standings?
If you did this for years on end, what would happen to the conference's prestige and also their individual prestige?
Would you consider this be against fairplay guidelines?

It would help the conference's prestige, but only to the extent the user teams were getting better and the ten games that are being scheduled in non-con were progressively tougher as well. It would take a lot of time and coordination, but you could raise the conference's RPI and prestige without playing quite as tough of a schedule. RPI (in real life at least) is 25% Winning Percentage, 50% opponent's winning percentage, and 25% opponent's opponents' winning percentage. TLDR: I think you could game the system a little bit, but not enough to cause issues. You would tank your opponent's winning percentage enough that you would still need 10 users to build solid teams that were capable of going 10-0 against 75ish RPI Sims. At that point, you're a top 8 conference anyway.

Let's break it down.

Winning Percentage (25%)
The winning percentage calc is straightforward. Conference play would be split 0.500. Non conference, with a conference of 12 playing 10 non-con games, 10 user teams winning all 10 games and 2 sim teams likely winning 2-3 games it would look something like this:

	Total Conf	Users	Sim
Non-Con W	105	100	5
Non-Con L	15	0	15
Conf W	81	80	1
Conf L	111	80	31
Winning%	0.596154	0.692308	0.115385

Opponent's Winning Percentage (50%)
Non-conference piece of this is ugly. All the sims playing user teams go 0-100. Assuming those ten teams took over a lower end conference, they'd have to start with non-con opponents in the 200+ range (EARLY), then maybe 125 range (MIDDLE), and ultimately sims in the 75 RPI range (LATE). This isn't concrete, but if you managed to target the right sim teams in those ranges, you could see maybe around 6-7 conference wins (EARLY), 9-10 conference wins(MIDDLE), and 12-13 conference wins(LATE). The results would look something like this:

Early	opponents
Total Conf	Users	Sim
15	0	15
105	100	5
72	60	12
120	100	20
0.278846	0.230769	0.519231

Mid	opponents
Total Conf	Users	Sim
15	0	15
105	100	5
102	90	12
90	70	20
0.375	0.346154	0.519231

Late	opponents
Total Conf	Users	Sim
15	0	15
105	100	5
142	130	12
50	30	20
0.503205	0.5	0.519231

Opponent's Opponents' Winning Percentage (25%)
Same as your conference, by default 16/26 games are close to 0.500 within your Sim's conference. Going back to initial winning percentage calc, 10 of their games would have winning percentage (from the user non-con) of about 0.692308. So 16 at .500 and 10 at .700 gets you about 0.580.

RPI calculation:
Putting it all together:

		Early	Mid	Late
25%	Winning Percentage	0.596	0.596	0.596
50%	Opponent's Winning Percentage	0.279	0.375	0.503
25%	Opponent's Opponents' Winning Percentage	0.580	0.580	0.580
	Conference RPI	0.433	0.482	0.546
	Approx. Place	25+	15-25	5-15

It would be pretty cool to see the experiment though. Also, if WIS RPI is calculated differently, it could change the analysis.

Wow thank you, I really appreciate this analysis.

I would get the admin’s explicit approval before actually doing something like this. While I don’t think there is significant benefit (or harm) in the isolated action, an entire conference working in concert on a rather important team power aspect (schedule) shows a level of coordination I probably wouldn’t be comfortable with, in general. Programs should all be competing with each other in all aspects, not forming factions and cabals (not to make your intentions sound nefarious, I know you’re probably just curious about how it plays out, as I am).

this makes more sense if everyone finds their own 10 sims, or some overlap, overlapping all 10 doesn't make a lot of sense. the goal is to find beatable opponents with solid records.

this is a pretty common discussion point in conferences, going back more than 10 years. the projection report change seble made was one of his finest (i know its not perfect but its a major improvement). it punishes high end teams for too-easy wins, so it doesn't really make a ton of sense for top conferences to take this too far.

however, it does make sense for lower end teams (everyone outside the NT at least) to do things like this. in d1 those wins are just so valuable to low prestige teams trying to raise their prestige, and trying to qualify for the PIT/NT. i often make a similar recommendation to conference mates, at least the lower end ones, to really work to schedule 8 or 9 wins. i recommend it because it makes sense for their program, but it does also benefit the conference, and i definitely mention that too.

for the RPI analysis post - WIS RPI does equal real RPI. your analysis looks pretty good, except that in the opponent's winning %, you are not removing the sim game against the team in question. OWP removes all games between the team in question and their respective opponents (OOWP does not have an analogue).