I would never recommend a coach on a rebuild come in and schedule the toughest opponents possible, and I've never heard any experienced coach suggest this. I think the common wisdom is pretty close to what you're suggesting here, aim for 5-8 wins (more if you can, but you can't control the weird upside down nights, and when you're bad, they happen more often), scheduling opponents that you are fairly confident you can handle. The best approach for the first few seasons is to focus on sims, and schedule them on the road. Look for teams that won't be graduating more than 4 players (fewer is generally better, but 0-1 may be too hard to handle as those upperclassmen sims get tough for rebuilds in the first years). Also look for teams that don't already have more than 3-4 humans on their schedule (and obviously pick sims that are not in human-dominated conferences). This last one is why I wait until pretty late in the season to schedule.

i agree this is roughly the conventional wisdom. i tell folks to schedule games you are likely to win, aiming for at least 7 wins across all scenarios and roughly about 9 for everyone outside the top 25 or so. i say target 9 wins, not 10, because you don't really want 10 sims you are 99% to beat, that is leaving too much on the table SOS wise. you really want 10 teams you are 90% to beat, more or less, when the goal is growing your program. once you hit the top 25 or so, scheduling tougher games you can learn more from has value where aiming for perhaps 8 overall wins (6 easy games and 4 50/50s or something) has some real appeal. but going for the 9ish wins is pretty much always the way to optimize seeding and prestige from regular seeding.

For some reason or another I’ve definitely been asked to schedule harder, when I know that my team is not ready. Either way hopefully some new coaches will see this and understand the math behind RPI and what to do during scheduling. I’ve absolutely been asked that question from new coaches more times than almost anything else in the game. So hopefully this helps, ultimately that was the goal really.

Well, your breakdown of RPI is also incorrect, which may sway some of your conclusions. RPI is 25% your record, 50% opponents' record, 25% opponents' opponents record.

Posted by dahsdebater on 9/18/2023 3:09:00 PM (view original):
Well, your breakdown of RPI is also incorrect, which may sway some of your conclusions. RPI is 25% your record, 50% opponents' record, 25% opponents' opponents record.

Where did I get mine from, you are correct but, for some reason I have seen it was 50% team record somewhere. My oversight, dummy me

This is why there is an argument for scheduling the winningest opponents you have some kind of shot against - the biggest chunk of RPI is based on your opponents winning percentage. With that being said, RPI is not as important for NT seeding as it used to be long ago. Even so, there's often still something to be said for scheduling the strongest sims from weak conferences because they have a good shot to win 20+ games and still be very beatable. As gill alluded to, this probably isn't the best strategy if you're targeting a deep tournament run, but if you're in a building/rebuilding phase or just having a down year, it's a great way to schedule your non-con to give yourself the best shot at a decent tournament seed.

I always wait until the end of RS1 to schedule. I want to have a good idea of what my team will look like. As I've only been playing at the higher end of D1 lately, I mostly schedule around promises, which means I won't schedule a press defense and weeks 6-9 will mostly be cupcake sims. That gives me enough time to see if my freshmen aren't getting enough minutes so that I can give them a few games on "Foul Trouble Only" against easy sims.

For the other 6-7 games, it really depends on how good I think my team is. If I think my team will have a good shot at the S16 or better, I'll try to schedule a top-10 SoS. Otherwise, just schedule some lower level human-coached teams that should end up in the top-100 RPI range.

Interesting to know about RPI. 25% your own record, 50% opponents’ record, and 25% opponents’ opponents record.

Now, does anyone know what the Projection Report formula is??

No because RPI is a known formula that uses the one from real life. The projection report is an unknown formula

Posted by craigaltonw on 9/23/2023 10:33:00 PM (view original):
Interesting to know about RPI. 25% your own record, 50% opponents’ record, and 25% opponents’ opponents record.

Now, does anyone know what the Projection Report formula is??

Also a key component to the RPI formula:

*Your* W-L record (that initial 25%) is adjusted based on if your results were at home or on the road.

Your opponents' W-L record, and your opponents' opponents' W-L record stay as-is.

So, for *your* record:
- home wins count as 0.6 wins.
- home losses count as 1.4 losses.
- road wins count as 1.4 wins.
- road losses count as only 0.6 losses.

So if a team plays all *home* games and goes 5-5, their W-L record for RPI purposes is actually 3-7.
.

Not sure if my ideas fit with the ultimate goal of a high RPI but this is a rough estimate of what I do for scheduling.

SSchedule 3-5 very likely wins or more in conference with 4 or less humans (including yourself)

5-8 humans in conference schedule 6-8 likely wins. I usually go with 3-4 guarenteed sim wins and 4-7 games that can go either way depending on season.

9-12 in conference go 8 very likely wins and 1-2 games against comparable coaches, and if feeling great about yyour eam...one against a titan to test your team and your skill.

If you are typically at the top of your conference you can include more either way games and count on enough wins in conference to still get NT bid.