Greetings football fans. Tonight will start a week’s worth of analysis with a look into SOS so far as well as the weekly GCR prediction analysis. Plus the weekly, and yet, not really surprising, prediction of the committee’s top 4. On Tuesday, I’ll rank the conferences heading into Championship week, then on Thursday, I’ll predict every game from Division 1 and talk about how this weekend’s matchups can and should impact the playoff. If you have any questions or requests for analysis, send it in – this is the week to get everything out from under the rug, so bring it on!
Before we get into the GCR predictions from last week, there are now 79 teams eligible for bowls (this does not count Missouri who won 6, but had a denied appeal to qualify). That means one eligible team will sit out. If I had to guess at who, I would say one of the MAC 6-6 quartet: Eastern Michigan (104/93); Kent State (86/82); Ohio (50), or Toledo (97/88). The only other Group of 5 6-6 teams are Florida International (107/94) but they have that win over Miami (32) to boost them, and Tulane (111/96) – they are possibles.
The predictions this week were hit or miss. Hit if we are talking FBS where the GCR was 45-19 and miss in the FCS playoffs (plus 2 SWAC regular season games) at 4-6. For the year now, the GCR is 532-276 for a solid .667 correctness percentage. Here are the tables. Clearly the FCS prediction formula will need an overhaul over the summer.
| Decile | FBS | FCS | Total | Expected Wins |
| 50.0-59.9% | 10-10 (,500) | 0-5 (.000) | 10-15 (.400) | 13.75 |
| 60.0-69.9% | 10-6 (.625) | 3-1 (.750) | 13-7 (.650) | 13.00 |
| 70.0-79.9% | 12-3 (.800) | 1-0 (1.000) | 13-3 (.813) | 12.00 |
| 80.0-89.9% | 7-0 (1.000) | 0-0 (—) | 7-0 (1.000) | 5.95 |
| 90.0%+ | 6-0 (1.000) | 0-0 (—) | 6-0 (1.000) | 5.70 |
| Total | 45-19 (.703) | 4-6 (.400) | 49-25 (.662) | 50.40 |
| Decile | FBS | FCS | Total | Expected Wins |
| 50.0-59.9% | 75-68 (.524) | 83-73 (.532) | 158-141 (.528) | 164.5 |
| 60.0-69.9% | 81-37 (.686) | 73-45 (.619) | 154-82 (.653) | 153.4 |
| 70.0-79.9% | 76-19 (.800) | 51-24 (.680) | 127-43 (.747) | 127.5 |
| 80.0-89.9% | 42-3 (.933) | 27-4 (.871) | 69-7 (.908) | 64.6 |
| 90.0%+ | 26-1 (.963) | 19-2 (.905) | 45-3 (.938) | 45.6 |
| Total | 300-128 (.701) | 253-148 (.631) | 553-276 (.667) | 555.6 |
Last summer, in one of my first blogs, I posted a pre-season strength of schedule for each team based on how their 2019 opponents performed in 2018, adjusting for home/away/neutral. South Carolina (73/71) was the only team facing opponents who had combined for over 100 wins the year before. So, I thought it’d be fun to see how the predictions came out – did I hit the mark on the top 10 at least? Keeping in mind, I’d never done anything preseason before, and the method made an assumption that teams would repeat success or not so much year over year. I didn’t expect the results to match – more did it for fun. I cut the SOS data a couple of different ways to help explain (since this is G’s Explanation after all) the SOS logic in the formula. The only caveat is the season is not over and all teams, even those who have no more games, will be impacted by opponent’s (and opponent’s opponent’s) postseason exploits. But since the regular season is over (except for the Army/Navy game in two weeks), it’s a good time to look at where we stand. First, let’s look at those preseason predictions (numbers are SOS ranking, not Total Ranking).
| Preseason SOS | Team (Current SOS) | Current SOS | Team (Preseason SOS |
| 1 | South Carolina (1) | 1 | South Carolina (1) |
| 2 | Southern Cal (26) | 2 | Rutgers (21T) |
| 3 | Tennessee (28) | 3 | Texas A&M (10) |
| 4 | Florida State (5) | 4 | Auburn (16) |
| 5 | LSU (48) | 5 | Florida State (4) |
| 6 | Louisville (20) | 6 | Michigan (18T) |
| 7 | UCLA (13) | 7 | Texas (42) |
| 8 | North Carolina (35) | 8 | Penn State (24) |
| 9 | Georgia Tech (11) | 9 | Maryland (29) |
| 10 | Texas A&M (3) | 10 | Michigan St (31) |
We see the prediction wasn’t half bad. 5 of the the top 10 currently have a top 13 schedule. LSU was the biggest miss dropping from preseason 5th to 48th – we’ll analyze that a little closer in the next segment. The biggest jump into the current top 10 is Texas from 42nd to 7th. On Tuesday when I compare conferences, it should be no surprise that the Big 10 and the SEC are fighting it out for the top spot. 8 of the current top 10 are from those conferences.
I decided to go a little deeper into the SOS. I looked at each of the top 10 SOS and tried to show the why behind the score. Let’s look at the table.
We see South Carolina’s opponent’s already have 96 wins and 9 of them still have games (not Charleston Southern, Vanderbilt, or Missouri). Three of their opponents play next Saturday (Clemson, Georgia, and App State) and then the bowls come into play so it looks like 100 opponent wins are doable. Not that the fact would help a 4-8 team feel better, but it looks like the toughest schedule is a near-lock. The SOS is where I started what would one day be the GCR so it’s a good place to look every now and then.
Now is probably a good time to point out the top 4. I have the same teams the committee will post in the top 4, but in a different order. I have Clemson, Ohio State, LSU, and Georgia. Thursday I’ll make a prediction on the final 4, but this Tuesday the committee will publish Ohio State, LSU, Clemson, and Georgia. Utah will be 5th and Oklahoma probably 6th (although I think Baylor will beat them in the rematch). I wondered how the playoff group would look in the SOS analysis I did for the toughest schedules. Here’s that table:
Going into the game against Virginia, Clemson opponent’s have an aggregate losing record (more on why this is Tuesday). Compare the away opponent’s winning percentage from these 4 to the 10 above. Only LSU had a non-losing opponent’s record. UGA’s neutral game was the Cocktail Party in Jacksonville against Florida so their non-home opponent’s record is above .500 (32-28 .533). Clemson so far has played 0 teams with 10 wins. They’ve won their last 7 games by more than 30 so their performance score is tremendous, but they haven’t really been challenged. If they beat Virginia (9-3), they won’t face a double digit winner until the semis. If they lose, they will be 0-1 since UVA will pick up their 10th.
I’ll keep an eye on these as we go. That’s it for this edition. Plug back in on Tuesday and see how your favorite conference stacks up. Thanks, as always, and please keep the comments, questions, challenges, and sharing with others happening. G
