What a great weekend of college football – shootouts, blowouts, last minute scores, upsets, critical reviews, bad calls – all we could want in a weekend. Before we get to far into analysis, the GCR went 6-3 or a combined 7-3 season record. Oklahoma (36) bested Texas (38) in 4 OT, Clemson (6) was not upset by Miami FL (12) at home, and Duke (67) joined the “we won a game in 2020!” club by beating Syracuse (53). We still have 13 undefeated teams that have played at least 3 games. And the top 25 got all reshuffled – partly because of the games this last week and partly because the sample size is still low. Expect more large jumps or slides, really, for the rest of the season. We did have 3 first-2020-game teams this week: Houston (25) beat Tulane (55), Temple (73) lost to Navy (54) and Mercer (89 and last) came up short against Jacksonville St (66).
…..
The GCR has been verified and ready to rank the Division I teams from 1 to 89. But first, I’ve been reminded that we have new readers that may not feel the need to read all previous 108 posts (I know, surprise to me too, right?). So a quick sum-up of how the GCR works. First there are currently 130 teams in the FBS evenly split between the Power 5 (ACC, Big 10, Big 12, Pac 12, SEC and Notre Dame), and the Group of 5 (American, Conference-USA, MAC, Mountain West, Sun Belt and all other FBS Independents). Add to that 127 teams in the FCS (Big Sky, Big South, Colonial, Ivy, MEAC, Missouri Valley, Northeast, Ohio Valley, Patriot, Pioneer, Southern, Southland, SWAC, and Independents). Those 257 teams make up the totality of the GCR. While I don’t rank Division II, III, NAIA, or others), if a member team plays one of them I track the game. The GCR is split into two components: Strength of Schedule (SOS) and Performance scores. Those combine to determine the Total Score which is used to rank the teams in order. The SOS is calculated for each school using the number of wins each opponent has, adjusted for cross-league (e.g., a Power 5 team playing an FCS team will get a severe discount while the same team playing a Group of 5 team gets a lesser discount), and for place of the game (playing an 8 win Power 5 team at home is “easier” than playing them on the road – neutral is in the middle). There are a few more factors, but those are the main things. The Performance score is based on the binary outcome of the game (did you win or lose), point differential (in a diminishing return kind of way), and other lesser important factors (including place of the game). So a Power 5 team that plays an FCS team early in the season will experience a big deduction for SOS and a smaller one for Performance. Even more so if that FCS team doesn’t win other games early. You might have caught I said “early”. That’s excellent. The only reason timing matters is that GCR doesn’t ever stop calculating. Let’s use LSU (48) as an example. Last year they were the national champions and were ranked (by the AP and others which are completely irrelevant in the GCR) in the top 15. Unranked Mississippi St (49) beat them much to the surprise of the college football world. In the sports world, the Bulldogs get credit for beating a ranked team. But the Bayou Tigers are 1-2 and clearly not an elite team. The GCR solves this problem by recalculating every game every week. If LSU continues to lose, Mississippi St will get less “credit” for the win. I hope that makes sense. To tie that back to the FCS example: by the end of the season, that early game will be folded into the much more difficult rest of the schedule and be “equal” to another Power 5 team playing their patsy in week 12. The GCR has no other assumptions or biases. All 257 teams begin the season tied for first (or last if you are more of a glass half empty kind of person). Only the play on the field matters. Opponents, location, and score are the only variables that really matter in the formula. Hope that helps. Check out some of the early posts in August/September of 2019 if you’d like more specifics. Oh, yeah, I’ve been working on the GCR for about 8-9 years now – it started during the SEC championship streak. Before 2019, I simply emailed the rankings to some friends who encouraged me to open it up – for which I am grateful as that encouragement was the impetus of the blog. If you have any questions about how the GCR works (short of the formulas themselves – that’s a secret), comment and I’ll do my best.
…..
Ok. Here is this week’s top 5: holding at number 1 is UNC (3-0 1)after beating Virginia Tech (2-1 8), followed by Alabama (3-0 2) who battled a pesky Mississippi (1-2 41) squad, Georgia (3-0 3) who have proven they are a second half team by leaving Tennessee (2-1 9) behind, Oklahoma St (3-0 4) who were idle, and Notre Dame (3-0 5) who took care of business with Florida St (1-3 59).
…..
Without further ado, here is the full listing. I’ll be back in a few days with some analysis on what it all means. Thanks for reading, please comment and share with others. Later, G
