We already talked about how the current College Football Playoff and its four-team bracket was a pseudo-expansion of the previous BCS system, and it didn’t solve a whole lot of problems since that expansion still left well over two-thirds of college football teams starting their season with a zero-percent chance of reaching said playoff bracket.
We know that expansion is likely inevitable, it’s just deciding what form it will take. We at Forgotten5 decided to create and simulate a 24-team playoff to take the expansion to the logical extreme and decide just how much expansion would be too much.
Your reminder of the guidelines:
- If you win your conference, you are in
- If you’re a Power Five team, winning your conference also means a first-round bye
- Highest-ranked G5 team gets a first-round home game
- Without a conference title, you don’t get into the top 25 without at least nine wins
We used these rules to alter the CFP ranking from each year going into bowl season, then set up the bracket. Each matchup was simulated 10 times using WhatIf Sports and their delightfully easy simulator, with an 11th sim if there was a tie.
This was the year in reality where the playoff question boiled over and commentators and fans alike took sides, as the growth of talent in the G5 conferences was undeniable through many upsets of Power Five schools.
Boise stumbled away from the NY6, and Memphis did better than expected in their place. The Lane Train brought FAU through the C-USA Championship again, the volatile MAC produced a new winner, and Appalachian State hung on the repeat as Sun Belt champions.
But what would happen to them in the playoffs?
Unfortunately in this year, Memphis’ home victory over Iowa was the lone bright spot in postseason play for the Group of Five. Miami again gave the MAC a representative to get bludgeoned in the tournament, but Cincinnati, Appalachian State, and Boise State all struggled just as much.
FAU gets credit as the G5 team to perform the best on the road at Penn State, but that only soothes the wounds so much.
Full data below:
|Matchup||Winner||Final||Margin of Victory (Tiebreaker)|
|FAU-Penn State||Penn State||9-1||13.22|
|Minnesota-Notre Dame||Notre Dame||9-1||12|
|Penn State-Baylor||Penn State||9-1||21|
|Notre Dame-Ohio State||Ohio State||10-0||23.1|
|Penn State-Ohio State||Ohio State||9-1||18.78|
|Clemson-Ohio State||Ohio State||5-5*||(13-15)|
I think that this basic simulation, if anything, just makes me want to do it again. You’d get the opportunity for Western Kentucky’s crazy 2015 appearance, UCF’s strong showing in 2017, and Appalachian State nearly getting another upset to add to their historical ledger.
I also think that some of the close losses could turn into close wins if we ran the simulations enough times, much like the App State/Kentucky game from 2018, but I wanted to not get drowned in too much data for an initial simulation.
There obviously would be kinks to work out, like avoiding having the best G5 teams face other G5 teams in the opening round. This seems not unlike the typical season though. Mountain West teams and American teams play well more often and are more likely to advance, while the MAC, CUSA and Sun Belt are far less likely to achieve any success but just might pull one out of their hat if given the opportunities.
We want to increase both revenue and access across college football. The big name programs won’t lose any revenue, and might even gain from having multiple postseason home games like the pros do.
The smaller schools might just be increasing their odds of adding an ugly money game to their schedule, but there would at least be more money to share in and the slimmest chance of a victory that could do far more good than harm for the future of their program.
I think that there is really no such thing as “watering down” the regular season or the postseason via a playoff expansion, but I’d want to do far more simulations before deciding if the point of diminishing returns sits at 16 teams or if 24 teams is truly the sweet spot going forward, especially since the FCS level already provides a blueprint for how to make the 24-team model work.