The season for the Michigan State Spartans has come to an end, but there is still a lot of football to be played before a National Champion is crowned. For the remainder of the season, I will be continuing to issue a few short pieces of analysis.
This weekend, there are nine game on the schedule, each of which will crown a new conference champion. Based on the results of those nine games, the College Football Playoff Committee will release the final set of rankings and the resulting bracket for the inaugural 12-team College Football Playoff. Let's discuss both topics.
Championship Weekend Bad Betting Advice
With the limited number of games this weekend, I will simply present my summary table of the nine games which includes my projected scores, the opening point spread, and the computers' projected point differentials. See Table 1 below.
From a betting perspective, the two computer models together are projecting upsets in three of the nine contests. Most notably, both algorithms are picking Penn State to upset Oregon in the Big Ten Championship Game and Southern Methodist to upset Clemson in the ACC Championship Game. My computer is also picking Ohio to win the MAC Championship over Miami (Ohio).
That said, all nine games this weekend have opening spreads under six points, so any upset is likely to be a minor one. My simulation of the weekend suggests that 3.8 plus-or-minus 1.5 upsets out of the nine games is most likely
As for picks against the spread, none of the games meet my standard criteria to recommend a wager. However, Table 1 above does indicate which team each computer believes will cover. The FPI favors the teams shaded in red while my algorithm favors a team shaded in green. If there is no red shaded team in a given row, both computers agree on the pick ATS.
If one is looking for the best bet against the spread, the strongest signal from both computers is for Penn State to cover versus Oregon (-3.5).
As for point total bets, my computer does have one suggestion for the weekend and that is to take the "over" on Tulane versus Army (47.5 points) in the American Athletic Conference Championship Game.
Playoff Scenarios
Of the nine games this weekend, only five of them are likely to impact the college football playoff. Based on the most recent College Football Playoff rankings, the winner of the Mountain West Championship between Boise State and UNLV will earn a playoff berth. The results of the other four Group of Five title games will not influence the final bracket.
Therefore, the final playoff field will be decided by the results of the Big Ten, SEC, Big 12, ACC, and Mountain West Championship Games. That means that there are only 32 possible scenarios. Based on the opening Vegas lines listed above in Table 1, we can calculate the odds for all 32 scenarios.
It is also possible to make a few assumptions in order to simplify the analysis and to make some educated guesses as to the seed each team will receive.
If Oregon beats Penn State, the Ducks will be the No. 1 seed. If Oregon loses and Texas beats Georgia, the Longhorns will be the No. 1 seed. If both Oregon and Texas lose, Penn State will be the No. 1 seed. The No. 2 seed will be the Big Ten or SEC champion which is not the No. 1 seed.
As for the rest of the playoff teams, I think that selection is very straightforward. I think that both the ACC and the Big 12 are one-bid leagues. Notre Dame, Ohio State, Tennessee, Indiana, Alabama, and the loser of the Big Ten and SEC Championship Games are all "in."
The only bit of uncertainty is if Clemson beats SMU, there is a chance that SMU would still get a bid over Alabama. I do not expect this to be the case.
If I make these assumptions and use the current college football playoff rankings, the No. 3, No. 4, and No. 12 seed will likely be selected using this order:
1. SMU (with a win over Clemson)
2. Boise State (with a win over UNLV)
3. Arizona State (with a win over Iowa State)
4. Iowa State (with a win over Arizona State)
5. Clemson (with a win over SMU)
6. UNLV (with a win over Boise State)
It is possible that Clemson could jump over the Big 12 Champion, but I believe that this is unlikely.
With this logic in place, the only remaining question for the committee is how far the losers of the Big Ten and SEC Championship Games will fall in the rankings. More importantly, will the committee utilize this uncertainty to create the best bracket possible by minimizing possible rematches?
The official selection guidelines direct the committee to focus solely on rankings and not to attempt to shuffle teams to optimize the bracket. As I have stated many times, this rule is idiotic and I hope the committee is both smart enough and brave enough to ignore it.
With the above assumption for the results of the Big 12, ACC, and Mountain West title games in place, this leaves only four scenarios to consider to fill out the rest of the bracket.
The most likely scenario (33%) is where Oregon and Texas are winners this weekend. In this case, my projection is as follows:
No. 1 Oregon
---No. 8 Georgia
---No. 9 Indiana
No. 4 Boise State (34%)/ Iowa State (27%)/ Arizona State (24%)/ Clemson (18%)
---No. 5 Notre Dame
---No. 12 UNLV (35%)/ Clemson (34%)/ Iowa State (16%)/ Arizona State (15%)
No. 3 SMU (47%)/ Boise State (34%)/ Iowa State (10%) / Arizona State (9%)
---No. 6 Penn State
---No. 11 Alabama
No. 2 Texas
---No. 7 Ohio State
---No. 10 Tennessee
This would require Notre Dame to move up to No. 3 in the final college football playoff poll, followed by Penn State, Ohio State, Georgia, Indiana, Tennessee, and Alabama.
The trick here is to move Indiana up and past Tennessee in the rankings to avoid Georgia playing Tennessee in the No. 8/No. 9 game and Ohio State playing Indiana in the No. 7/No. 10 game.
There is also a more plausible seeding where Ohio State plays Tennessee in the No. 8/No. 9 game and No. 7 Georgia plays No. 10 Indiana in the first round, but this potentially sets up both a quarterfinal matchup between Oregon and Ohio State and round three of Texas and Georgia. That is less than ideal.
The second most likely scenario (27%) is where Oregon and Georgia win on Saturday. In this case, my projection is as follows:
No. 1 Oregon
---No. 8 Ohio State
---No. 9 Tennessee
No. 4 Boise State (34%)/ Iowa State (27%)/ Arizona State (24%)/ Clemson (18%)
---No. 5 Texas
---No. 12 UNLV (35%)/ Clemson (34%)/ Iowa State (16%)/ Arizona State (15%)
No. 3 SMU (47%)/ Boise State (34%)/ Iowa State (10%) / Arizona State (9%)
---No. 6 Penn State
---No. 11 Alabama
No. 2 Georgia
---No. 7 Notre Dame
---No. 10 Indiana
Here, I have Texas falling to just No. 3, followed by Penn State and Notre Dame. While the potential for a second round rematch between Oregon and Ohio State is possible, all other rematches are easily avoided.
This is an example of where some flexibility in where Notre Dame is placed is helpful. Putting the Fighting Irish as the No. 5 seed instead of the No. 7 seed would result in the less-than-desirable first round games pitting Penn State versus Indiana and Texas versus Alabama.
The third most likely scenario (22%) is where Penn State and Texas win on Saturday. In this case, my projection is as follows:
No. 1 Texas
---No. 8 Ohio State
---No. 9 Tennessee
No. 4 Boise State (34%)/ Iowa State (27%)/ Arizona State (24%)/ Clemson (18%)
---No. 5 Oregon
---No. 12 UNLV (35%)/ Clemson (34%)/ Iowa State (16%)/ Arizona State (15%)
No. 3 SMU (47%)/ Boise State (34%)/ Iowa State (10%) / Arizona State (9%)
---No. 6 Notre Dame
---No. 11 Alabama
No. 2 Penn State
---No. 7 Georgia
---No. 10 Indiana
This scenario is the easiest to justify, as it just requires Georgia to stay ahead of Ohio State, which seems likely. Texas could face Tennessee in the second round, but those two SEC teams did not play each other in the regular season. This is the scenario that both my algorithm and the FPI estimate to be the most likely.
Finally, the fourth most likely scenario (18%) is where Penn State and Georgia both win. In this case, my projection is as follows:
No. 1 Penn State
---No. 8 Ohio State
---No. 9 Tennessee
No. 4 Boise State (34%)/ Iowa State (27%)/ Arizona State (24%)/ Clemson (18%)
---No. 5 Texas
---No. 12 UNLV (35%)/ Clemson (34%)/ Iowa State (16%)/ Arizona State (15%)
No. 3 SMU (47%)/ Boise State (34%)/ Iowa State (10%) / Arizona State (9%)
---No. 6 Oregon
---No. 11 Alabama
No. 2 Georgia
---No. 7 Notre Dame
---No. 10 Indiana
The challenge here is that I need to drop Oregon below Texas in the final rankings in order to balance out the bracket. This is harder to justify but it creates an obviously better outcome. There is a possible Penn State/Ohio State rematch in the second round, but other than that rematches are avoided.
That is all the advice I have to give this week. Enjoy the games this weekend.
Saturday Morning Update
With Boise State's win over UNLV last night, the Broncos have effectively locked up either the No. 3 or the No. 4 seed, depending on the outcome of the ACC Championship Game. Here are the updated seed probabilities:
No. 3 Boise State (53%)/ SMU (47%)
No. 4 Boise State (47%)/ Iowa State (28%)/ Arizona State (25%)
No. 12 Clemson (53%)/ Iowa State (25%)/ Arizona State (22%)
- DT
- RB
- WR
- CB
- OLB
- PRO
- OLB
- CB
- OT
- TE
