The error is probably in how the ranking system treats cupcakes.
Example: my team will play the #1, #2, and #3 teams in the country, as well as the #100, #101, and #102. Your team will play the #12, #13, and #14, and will also play the #71, #72, and #73.
Now, you'll beat your three cupcakes, and I'll beat my three cupcakes. Those games are all mathematical near-certain wins for each of us. So the real difference in our schedules is that I'm taking on the top 3, while you're taking on three easier teams from the Top 25.
But when the games are averaged out, my "weaker cupcakes" completely mask my "harder top end", and I get an average difficulty of 52. Meanwhile, your stronger cupcakes artificially inflate your schedule's average to 42.
Using blind math, without thinking it through, your schedule looks to be 10 spots harder than mine.
...
This isn't just theory, either. This is exactly what's happening with our schedule in comparison to West Virginia's. The top three teams we play are ranked #1, #3, and #9. Meanwhile, WVa's top three opponents are ranked #7, #16, and #23. But the bottom of our schedule has teams ranked in the 100-130 range, while the bottom of WVa's calendar is mostly in the 70s and 80s.
So most of the Strength of Schedule calculators show their schedule as hella harder than ours. When in reality, I'd rather play their slate, for sure. I can find a way to eke out wins against Youngstown State, Kansas, Iowa State, and Baylor, even if they are more of a challenge than Vandy, ETSU, Texas El Paso, and Charlotte.
