I promised you I would show you my DIY POWER RATING SYSTEM which measures teams as per last week's results. Check it out.
- Method: Aggregate net yards differential to-date + aggregate net points differential to-date x or -:- by a weekly updated SOS coefficient.
- Measures performance to date on an aggregated basis. As such, it is a dynamic measure that changes from week to week based on how a team continues to perform. NO OTHER FACTORS as when you are DEFINING POWER by ACTUAL REAL-WORLD RESULTS, everything else is already BAKED IN.
- Measures how well a team has performed using the most BASIC METRICS but without regard to wins or losses. In other words, how has a team DONE OVERALL – in terms of moving the ball and defending – against teams it’s ACTUALLY PLAYED. What’s its OVERALL AGGREGATE POWER OUTPUT as STATISTICALLY DEMONSTRATED. No INTANGIBLES or QUALIFIERS.
- It’s INDICATIVE, not PREDICTIVE as anything can happen on ANY GIVE SATURDAY.
- That said, I just put together a POINT SPREAD ANALOGUE which I hope to test over the balance of the season. It took me awhile to figure out what the size of the point intervals between orders of magnitude should be and how many points to allocate per interval, but I now think I have it. I'm interested in how well "my lines" stack up against the various gambling odds.
MY POWER RANKINGS BASED ON GAMES PLAYED IN THE WEEK ENDING 10-12-24
The first number in parenthesis following my calculated power rating for teams 1 through 25 is their FPI ranking; the second, their AP ranking.
1. Texas – 562.97 (1)/(1)
2. Ohio State – 510.13 (2)/(4)
3. Ole Miss – 497.53 (5)/(18)
4. Miami – 397.88 (10)/(6)
5. Alabama – 397.45 (3)/(7)
6. Penn State – 376.69 (9)/(3)
7. Notre Dame – 374.00 (6)/(12)
8. Oregon – 372.76 (8)/(2)
9. Indiana – 337.03 (15)/(16)
10. Georgia – 335.82 (4)/(5)
11. Tennessee – 334.15 (7)/(11)
12. Louisville – 315.39 (16)/(NR)
13. Washington – 310.87 (40)/(NR)
14. Missouri – 280.64 (19)/(19)
15. Boise State – 262.64 (22)/(15)
16. USC – 257.98 (13)/(NR)
17. Arkansas – 249.99 (25)/(NR)
18. Texas A&M – 248.92 (12)/(14)
19. Clemson – 245.82 (11)/10
20. Iowa State – 244.63 (17)/(9)
21. SMU – 237.63 (18)/(21)
22. Tulane – 223.79 (24)/(NR)
23. Georgia Tech – 219.09 (43)/(NR)
24. LSU – 198.49 (14)/(8)
25. Pitt – 196.86 (36)/(20)
Teams Ranked by FPI and/or the AP That Didn’t Make My Top 25:
Oklahoma, ranked 20th as per FPI and unranked by the AP, came in at –17.21.
Kansas State, ranked 21st as per FPI and 17th by the AP came in at 188.50.
Iowa, ranked 23rd as per FPI and unranked by the AP came in at 168.18.
BYU, ranked 29th as per FPI and 13th by the AP, came in at 174.86.
Michigan, ranked 33rd as per FPI and 24th by the AP, came in at –13.45.
Army ranked 51st as per FPI and 23rd by the AP, came in at 141.38
Illinois, ranked 57th as per FPI and 22nd by the AP, came in at 92.11.
Navy, ranked 70th as per FPI and 25th by the AP, came in at 99.73.
The only two instances where the FPI ranking and my calculations came out EXACTLY the same were with Texas at 1 and Oregon at 8. With Texas it was a double match with both FPI and the AP, while with Oregon, it was only with FPI.
These numbers will obviously change in the weeks ahead as teams play tougher and/or weaker teams and acquit themselves better or worse as the case may be.
Using my method and the point spread system I’ve developed off the back of it,
ND is a 9 point favorite on a neutral field against Georgia Tech.
Frankly, I think this system has legs, and I will test the point spread aspect of it over the balance of the season.
Basically, it’s just three simple steps: points differentials, yardage differentials, SOS co-efficient. With all of the data there IN PLAIN SIGHT online.
This is not an attempt at GOSPEL nor is it meant to be. It's just a COMPARATIVE MEASUREMENT TOOL BASED ON ACTUAL DATA. To me, the correlation between the DATA and the RANKINGS is AT LEAST AS PLAUSIBLE as that provided by FPI and whatever BLACK BOX METHOD beyond the eye-test the AP uses in its rankings, IF ANY.