Data for viewership isn't consistent. Would be nice to have some supporting data.
Based on this website, USF had 8 rated broadcasts in 2025 -
Week 1 vs Boise - ESPN - 2.04MM
Week 2 vs UF - SECN - not rated (rumored ~750k with a peak of ~1.5MM)
Week 3 vs Miami - CW - 720k
Week 4 vs SCSt - ESPN+ - not rated
Week 5 vs Charlotte - ESPN2 - 749k
Week 6 vs UNT - ESPN2 - 1.1MM
Week 7 vs FAU - ESPNU - 38k (lol)
Week 8 vs Memphis - ESPN2 - 392k
Week 9 vs UTSA - ESPN - 586k
Week 10 vs Navy - ESPN2 - 463k
Week 11 vs UAB - ESPN+ - not rated
Week 12 vs Rice - ESPN+ - not rated
8 games rated, 6,088,000 viewers, AVG of 761,000. One thing to note, absolutely nobody watches ESPNU. Only 7 games on ESPNU over the entire 14 week season drew >100,000k viewers, with the high being 234k. ESPN2 consistently drew 250-500k, ESPN broadcasts mainly being 1MM+, and ABC being 3MM+.
EDIT: after clicking on the link, the author states his article is 2024 data, but being written in December of 2025? Must be a typo. And considering it's the same source as I had, if you divide the 6,088,000 by 12 instead of 8, you get his 507k number. Therefore, he's counting weeks that games weren't broadcast as 0, which is inaccurate as he just doesn't have the data for non-rated channels/broadcasts. Dividing by # of broadcast games is more indicative of "average viewership" than a placeholder 0 when he doesn't have data. AKA his data skews teams who weren't on a rated channel much lower than they appear, which should give every team in the American a bump from the OP #'s.