clock menu more-arrow no yes mobile

Filed under:

College Football Mathematical Preview: Independents and the Group of Five

We have taken a deep dive into the math of MSU’s schedule, the Big Ten and the rest of the Power Five. Today, it’s time to examine the rest of the FBS.

NCAA Football: Tulsa at Michigan State
MSU handled Tulsa in Spartan Stadium in 2019. Will the Golden Hurricane challenge for the AAC in 2021?
Mike Carter-USA TODAY Sports

This month I have been rolling out the results of my annual simulation of the college football season, based on a 100,000-cycle Monte Carlo simulation that uses the consensus preseason rankings of all 130 FBS teams as an input. So far, I have performed an in-depth analysis of Michigan State’s schedule, summarized the Big Ten race, and took a look at the rest of the Power Five.

Today, it is time to complete a quick survey of the independents and the Group of Five in order to complete this part of the puzzle. Let’s jump right in.

Independents (Notre Dame) Overview

Table 1 below gives the results of my simulation for the group of FBS independent teams. This table and the ones that follow use the same format and show the consensus rank of each team, the projected record assuming all the favored teams win, and the record if I “disrupt” the simulation by forcing a historically accurate number of upsets.

The table also contains the strengths of schedule for each team (with the national or conference rank in parenthesis) as well as the odds for each team to win a division title, conference title, make the College Football Playoff, and win the national title. Note that a probability of zero (0.0E+00) simply means that outcome was not observed in any of the 100,000 simulations. So, while those outcomes are still possible, they are extremely unlikely.

Table 1: Full Season simulation results for the FBS independents

When it comes to the independents, the real question is “will Notre Dame challenge for a playoff spot?” Based on Table 1, the answer to that question appears to be “yes.” My simulation gives the Irish a 21 percent chance to make the College Football Playoff again this year and predicts an overall record of 11-1, but with an expected win total of only 8.77. What this means is that Notre Dame’s schedule sets up for the Irish to be a narrow favorite in a lot of games. Based on the preseason rankings, Notre Dame (ranked No. 11) will face four top-20 opponents: North Carolina (No. 9), Wisconsin (10), Cincinnati (13), and USC (17). But, three of those games are in South Bend and the Irish will likely be a narrow favorite in all three.

Only the neutral site game with Wisconsin in Chicago projects as a loss, but that game essentially looks like a push. Notre Dame’s toughest true road game is at Virginia Tech (35) where the Irish project to be a touchdown favorite.

So, it looks to be an interesting season in South Bend. If Notre Dame can get to 11-1 with the 23rd toughest schedule in the nation, I think a CFP berth is quite likely. The road is tough, but manageable. If Notre Dame is a little better (or a little worse) than expected and depending on luck, finishes between 7-5 and 12-0 are completely reasonable.

As for the other independent teams, Liberty also projects to go 11-1, but with a strength of schedule outside of the top-100. That said, if Liberty can score an upset at Ole Miss (No. 26) and versus Louisiana (32) to run the table, then a nice bowl game might be in the Flames’ future.

Group of Five Overview

Table 2 below gives on overview of the simulation results for the American Athletic Conference (AAC).

Table 2: Full Season simulation results for the American Athletic Conference

The AAC used to be divided into East and West divisions, but the departure of UCONN left the league with just 11 teams. So, the AAC adopted the Big 12 strategy of pitting the top-two teams against each other in the end-of-year AAC Championship Game. Thus finishing in the top-two is akin to “winning the division” in Table 2.

As the table shows, Cincinnati (No. 13) is a heavy favorite overall, with UCF (36) as the most likely challenger. The Bearcats host the Golden Knights this year, but the two teams project to win all of their other conference games and meet again for the conference championship.

If I were to pick a dark horse, Houston (73) looks like an interesting choice, as the Cougars host Memphis and avoid both Cincinnati and UCF entirely in the regular season. Houston’s strength of schedule is the easiest in the league and in the top-10 easiest in the country.

As for Cincinnati, the Bearcats travel to both Indiana (23) and Notre Dame (11), which gives them an opportunity to pad their resume and perhaps even make a case for a playoff spot. That said, Cincinnati would need a 12-0 record and a little luck form the Power Five in order for this to happen.

Next, let’s take a look at the Sun Belt Conference.

Table 3: Full Season simulation results for the Sun Belt Conference

The “Fun Belt” appears to be a three-team race between Coastal Carolina (No. 29) and Appalachian State (47) in the East Division, and Louisiana (32) in the West. Overall, Coastal Carolina has the best odds, but the Chanticleers have to travel to Appalachian State, which the disruptive simulation flags as a potential upset. In this scenario, App State wins the East, but loses to the Ragin’ Cajuns in the championship game.

Two of the marquee Sun Belt teams will get a chance to pad their resumes and stake a claim on the Group of Five spot in the New Year’s Six. Appalachian State travels to Miami (15), while Louisiana travels to both Texas (19) and Liberty (39). An upset win in one of those contests will boost the odds for a New Year’s Six bowl bid.

Next up is the Mountain West.

Table 4: Full Season simulation results for the Mountain West Conference

Once again, Boise State (No. 38) is the clear favorite, both overall (37 percent) and in the Mountain Division (61 percent). The West Division appears to be much more competitive, with four teams owning division odds between 15 and 36 percent. That said, my simulations favor the Nevada Wolfpack (50) to sweep the West and pick up their only conference loss at Boise State. That said, San Jose State’s (71) schedule is significantly easier, and the Spartans (West Coast version) would become the instant favorite with an upset win over the Wolfpack in Reno.

There are also a few notable non-conference games, which could help or hurt the eventually Mountain West Champion’s odds at claiming a spot in the New Year’s Six bowls. They are: Boise State at UCF, Oklahoma State at Boise State, Fresno State at Oregon, Fresno State at UCLA, Nevada at Cal, Nevada at Kansas State, San Diego State at Arizona and San Jose State at USC. A few upset wins in this group would help lift the profile of the entire conference.

Let’s now take a look at the MAC.

Table 5: Full Season simulation results for the MAC

The projected standings paint a clear picture and suggest that Ball State (64) and Buffalo (82) will meet in the MAC Championship game (where Ball State would be favored). Both teams have the benefit of playing the projected second best teams in their respective divisions (Ohio, No. 94 and Toledo, No. 69) at home this year.

However, the raw odds paint a slightly different story, as eight of the 12 MAC teams have at least a 14 percent chance to win a division title. The MAC always feels unpredictable, and it has that feel this year as well. The teams are pretty evenly matched and the conference race looks wide open.

Can a MAC team post a strong enough record to challenge for a New Year’s Six bowl slot? That would likely require a team running the table in conference play and picking up a resume-building upset in the non-conference. There are opportunities, but they are long shots: Ball State at Penn State (-22), Toledo at Notre Dame (-23.5), Buffalo at Nebraska (-14), and Ohio at Northwestern (-17) are the most likely candidates.

Finally, here is what the math says about Conference USA.

Table 6: Full Season simulation results for the C-USA

Much like the MAC, C-USA has the look of a league that should have competitive and exciting division races. Both of my projected standings as well as the raw odds give Marshall (No. 84) the nod in the East Division, thanks in large part to one of the easiest schedules in all of the FBS. But, the Thundering Herd are projected to pick up a loss at Florida Atlantic (96). FAU has tough games scheduled at both UAB and Western Kentucky, which both project as close losses, but a minor upset in either of those contests could hand the Owls the division crown.

In the West, UAB (No. 70) and UTSA (85) are the two top-rated team and the ones most likely to claim the division title. UAB grades out with the third toughest conference schedule, as the Blazers draw both Marshall and UTSA on the road. While the raw odds still suggest that UAB will come out on top, the disruptive simulation suggests that the Roadrunners will edge the Blazer and then fall to Marshall in the conference title game.

Will the eventual Conference USA champion have a shot to play in the New Year’s Six? At this point, this seems like a longer shot than the MAC. UTSA’s highest ranked non-conference opponents are Memphis (62) and Illinois (83), while Marshall best game is at Appalachian State (47). Wins there simply won’t move the needle. On the flip side, FAU travels to Florida (No. 14) and UAB visits Georgia (No. 5) and both teams project as almost 30-point underdogs. Stranger things have happened, but the odds are not good.

New Year’s Six Odds

As a part of my preseason simulation, I include a simple algorithm, which attempts to pick out the best team from the Group of Five Champions, based on the final projected record and power ranking in any given simulated season. Based on this analysis, I can project the odds of the eventual representative in the New Year’s Six for the 2021 season as summarized below in Table 7.

Table 7: New Year’s Six Bowl odds for the Group of Five members

Consistent with the analysis above, the eventual champions from the AAC, Sun Belt and the Mountain West appear to have the best odds overall, as teams from those leagues claim top-seven slots. Not surprisingly, the five Group of Five teams ranked in the top 40 in the preseason have the best chance. That said, those teams only soak up 66 percent of the overall odds.

Last year at this time, Coastal Carolina was ranked outside of the top-100. But, the Chanticleers came out of nowhere in 2020 to run the table in the regular season. While the Chanticleers lost a close bowl game to Liberty, CCU ended the season with a 11-1 record and a top-15 ranking. Will we see a similar surprise in 2021? Only time will tell.

So far, we have taken a close look at Michigan State, the Big Ten, the rest of the Power Five, and today, the Group of Five/independents. Next time, it is time to put those pieces together and to make some predictions about the College Football Playoff and the rest of the New Year’s Six Bowl games. Until next time, enjoy, and Go Green.