Don't know, they say they take it into account, they wouldn't blow smoke at us would they? FWIW here are the ESPN SOS rankings for the top 12 Alabama #4 Georgia #9 Texas #11 OU #!2 A&M #15 ND #23 Ole Miss #24 Ohio State #33 Oregon #36 Indiana #39 BYU #46 Texas Tech #58 2025 Resume College Football Power Index - ESPN There are a number of metrics at the link click on which ever one you are interested in and it reranks every team from best to worst.
I'm surprised there's talk. It's not uncommon for a team to lose a close game early and later in the season surpass the winning opponent. Of course, the most notorious example for ND fans happened in 1993.
The Chair Mack Rhoades specifically addressed it and said this year they added the Strength of Record (SOR) metric. SOR is the next level derivative to SOS and far superior in that it takes into account how well you performed against a given schedule. SOS actually doesn't take into account wins or losses or how you performed which seems to be kinda the point of the whole ranking system - SOR actually takes the schedule and corrects for that failing. Here's the SOR rankings for the top 12: 2025 FEI Ratings — BCF Toys Indiana #1 Texas A&M #2 Ohio State #3 BYU #4 Alabama #5 Georgia #6 Ole Miss #7 Oregon #8 Tx Tech #9 Texas #11 Oklahoma #15 Notre Dame #23
Sid, that's my understanding as well. Here's how my best pal Gemini explained it for me: The SOR Calculation Process (Conceptual) Define the Benchmark: The calculation first determines the power rating (FPI) of an "average Top 25 team." This acts as a reference team (sometimes referred to hypothetically as "NSU"). Simulate the Schedule: A Monte Carlo simulation is used to simulate the average Top 25 team playing the actual schedule of the team being evaluated (Team A) thousands of times. For each game, the FPI is used to determine the probability that the average Top 25 team would win against that specific opponent (taking into account home-field advantage). Count the Results: The simulation counts how often the average Top 25 team achieves a record equal to or better than the actual record of Team A. Calculate the Percentage (The SOR): The Strength of Record is the resulting percentage (or probability). Example: If a team is 8-0, and the simulation shows the average Top 25 team only goes 8-0 or better 1% of the time against that schedule, the SOR is 1%. Interpretation of the Result A Lower SOR is Better: A lower percentage indicates a more impressive accomplishment. It means the team's record was harder to achieve, as the average Top 25 team would rarely match it. Combining Record and SOS: SOR essentially combines Strength of Schedule (the difficulty of the opponents) with Actual Performance (the win-loss record) into a single metric. It directly answers the question: "How much better did this team perform against their schedule than an elite team would be expected to?"
Interesting so BYU has played one ranked team (#17 Utah State). Alabama has played and beaten 4 ranked teams (in consecutive weeks) and BYU has a higher SOR ranking? What is SOR measuring again? And what is it telling us? Did I read that right? BYU record was harder to achieve than Alabamas record?
That's precisely what it's telling you. It's saying that statistically speaking using the historical results of tens of thousands of past games using literally millions of simulations that it's been found to be more difficult to be undefeated with BYU's schedule than to have one loss to a team that doesn't have a winning record with Bama's schedule. It seems self-evident to me that having a loss to an FSU team who is 1-4 in the vaunted ACC on your record will result in a lower "strength of record" or SOR than that of a team that does not have any losses at all, much less a loss against a team with such a record....any loss of course will hurt your record strength, losses against lousy teams will hurt it more. That Bama's SOR is still very high at #5 is testimony to how very difficult is their schedule. In my mind that is exactly why SOR is a far superior measure....strong schedule is great, adding the results of those games only enhances the value of the measure, imo. Should be a good game this week with BYU and Tx Tech.