This was too tempting and I had to try! Claude to the rescue...

So here's what I did - I pointed Claude to MaxPreps for the DuKane Conference and asked it to generate an RPI. I used the same formula described here, with your win/loss, your opponents win/loss and your opponents' opponents' win/loss. I also layered in the home/away dynamics (0.7/1.3). Then I let Claude work - it took about 75 minutes. Here's what it spit out:

Team / Conf / Overall / WP / OWP / OOWP / RPI

StCE 2-1, 13-4, .774, .632, .669
StCN 3-0, 9-0, 1.00, .522, .558, .650
LP
WN
GN
WWS
Batavia
Geneva

There are too many digits for me to copy, and it's not important with such a small sample size - but it shows that this can be done! (Although, maybe I burned a million tokens doing it?!)

Like I said, it's WAY too early and this doesn't matter yet, but it was also fascinating to see the commentary Claude created:

"St. Charles East ranks #1 despite having the 3rd-best conference record (2-1) and a conference loss to Lake Park. Their enormous non-conference schedule — 17 completed games featuring strong opponents like Marist (16-0), Lincoln-Way East (12-4), Stillman Valley (12-2), and Lemont (12-4) — produces a very high OWP (.6326) and OOWP (.6397), which the model rewards heavily.

St. Charles North ranks #2 with the only perfect record (9-0) and a perfect WP (1.000). They lead the conference outright at 3-0, but their schedule lacks the same quality depth as SCE's, keeping their OWP lower (.5221). They still score very well overall."

I know that this information is gong to be weaponized now, but that's not what I was trying to prove - only that this can be done, and it would be tons of fun if it were regularly available!!