Rules:

Rule 1.
Please use a "*" or "nm" in the subject line to denote topic only posts.
Any post or attempt to take things "off the board" will result in a ban. Do not use other poster's real names or personal info.
Don't be a Tom Izzo.

Admins - Rich, BBA, Nikos
Ceremonial Moderator - the wino
Moderator Emeritus - Beave
Honorary Moderators - BigLake, Blue Man

POTY - Craig

📌 🏈 Michigan at Minnesota*
📌 To register, email victardgoblue@gmail *

The Victors

MGoBlog || UMHoops || BB90 || Scores || Cards || Chat || HOF || Games

Post a New Message

Home | Search | Login | Register | Settings

Yeah take SP+, calculate expected wins for each team, compare vs. Vegas (ideally with wagered money corrections not just raw O/U like you were doing)

Posted on August 13, 2026 at 12:06:57 PM by LewisBullox

Take the deltas, update, rerun until converged. I figured getting vegas odds might be problematic at some point, but I figured if only looking at top 25 or 50 teams.

I (Claude) don't mind giving it a shot from scratch, but might want to crosscheck if I get around to it.

Back to the board

Thread:

+148 for over 8.5 wins, -184 for under 8.5 wins. Basically 38% chance of 9 wins or higher, 62% chance of 8 wins or lower. (m) - By suede August 13, 2026 at 10:45:20 AM
- I read this as injuries. If healthy, I'd put it at more like 58-42 in our favor. But, yeah, Bryce misses the "wrong" 2 games and here we are.* - By ftb August 13, 2026 at 10:54:54 AM
- How does that simulation work? What are its assumptions? You are putting a lot of value in worthless preseason estimates. Enjoy the season* - By Rich August 13, 2026 at 10:46:39 AM
  - Read the Connelly article that I linked below. - By GritsGoBlue August 13, 2026 at 11:12:22 AM
  - Agree. I trust my gut over suede’s stupid computer projections* - By Poster P August 13, 2026 at 10:50:48 AM
  - Take two teams' SP+, take the difference to get a forecast line (after accounting for HFA)l, convert that to win probability, run Monte Carlo. Simple. (nm) - By suede August 13, 2026 at 10:47:39 AM
    - OK, but I believe our SP+ to be understated and IU to be overstated for example. There is no current year data to simply accept preseason SP+. You can if you want but that is part of why you have angst on something that is has no real data yet... - By Rich August 13, 2026 at 10:54:53 AM
      - I tend to have the same skepticism on IU, but looks like Vegas has them at O/U 10.5, which seems consistent (without running a MC) with their SP+ and schedule.* - By LewisBullox August 13, 2026 at 11:05:18 AM
        
        Yep, looks like Vegas has the under 10.5 wins a little more likely, which aligns with SP+. Here's the run card for IU from my simulator. (m) - By suede August 13, 2026 at 11:25:30 AM
        
        Looks tidy to me. I'm interested in extending your approach to do a complete Vegas corrected SP+. - By LewisBullox August 13, 2026 at 11:46:35 AM
        
        That would be really cool - are you visualizing a bunch of full FBS simulation runs with SP+ drift until the distribution of wins matches the Vegas win total over/under weighted to odds? (m) - By suede August 13, 2026 at 11:53:32 AM
        
        Yeah take SP+, calculate expected wins for each team, compare vs. Vegas (ideally with wagered money corrections not just raw O/U like you were doing) - By LewisBullox August 13, 2026 at 12:06:57 PM
        
        Love that approach, super curious to see it. The SP+ drift-informed-by-win distribution had a few interesting takeaways. (m) - By suede August 13, 2026 at 12:19:48 PM
      - I do this exercise precisely because there's no angst. I'm setting my own expectations properly. (nm) - By suede August 13, 2026 at 10:55:38 AM
        
        You have angst my friend. Great teams win games.* - By Lodz August 13, 2026 at 11:45:45 AM
  - Believe he's adjusting it to match Vegas* - By LewisBullox August 13, 2026 at 10:47:32 AM
    - Exactly. This takes the SP+ universe of power ratings and then back solves for what we would need to be in order to match it. (nm) - By suede August 13, 2026 at 10:48:22 AM
      - At this point of the year, it's just playing with numbers* - By Rich August 13, 2026 at 10:49:05 AM
        
        Whitt is going to field an extremely improved team but ppl are going to call him worse than Sherrone because 9 > 8 * - By Nikos August 13, 2026 at 10:52:57 AM
        
        That would only make sense if we are (again) repeatedly uncompetitive.* - By ftb August 13, 2026 at 10:55:50 AM
        
        I'll copy and paste what I said last night. (m) - By suede August 13, 2026 at 10:50:04 AM
        
        It starts with beating OU. OU was 6-6 in 2024 and leveraged themselves into the playoffs by beating Michigan on their home field. - By Rich August 13, 2026 at 10:57:19 AM
- 12 wins. Lock it* - By Poster P August 13, 2026 at 10:46:03 AM

Log in to reply.

Probation ends …