We’re now a few weeks into the 2021 college football season and, despite the rough loss to Texas A&M this past weekend, I can’t help but feel a tinge of something that I haven’t felt in the past few years of Lobo football: hope. Make no mistake, this Lobo team isn’t anywhere near where Coach Gonzales plans to take the program, but based on what we’ve seen already and the schedule ahead, I started to wonder if the Lobos had a reasonable chance to be bowl eligible this year. After the past four years where that didn’t seem like a realistic thing to even be pondering, it felt weird, and good, to have that cross my mind again. So, I decided to go back to some of the previous posts I did to simulate the results of the season based on current win probabilities and see what the Lobo chances are looking like.
This turned out to be a little more involved that before, where I could just look at the Lobos’ schedule on ESPN and take the win probabilities they showed there for the entire season. They removed that page and (spoiler alert) I couldn’t find the probabilities for all of the games until after I did a bunch of work to manually calculate the probabilities myself. Whoops.
So, what I’ve done this time is taken the SP+ ratings from ESPN for every team in the FCS, then used those to calculate the win probabilities for each remaining game in the college football season. The formula I used for that is taken from this Reddit post, but long story short, you calculate the differences between the two teams’ SP+ ratings, adjust for the home team having an advantage, then calculate a z-score from that adjusted difference by assuming a standard deviation of 17, then plugging that into a standard normal CDF function.
At that point, I had win probabilities for every game remaining in the season, which I could then use to simulate the season. For this, I focused on Mountain West teams, simulating the remaining schedules 10,000 times and seeing what the results were. Ultimately, I ended up with the results in the plot below, where we can see what percentage of simulations ended up with a particular win total for each team.
There are some things that stand out pretty quickly to me. First, there are a lot of teams in good shape to become bowl eligible. In fact, seven of the 12 teams have a 90% or better probability of being bowl eligible. That would be a lot of teams bowling, and with one more, we’d be at the eight teams in bowls that we saw in 2015 (which featured the annoying Nevada vs. CSU bowl game in Arizona, so there were only seven actual bowls the conference was involved in). Second, Fresno State and SDSU look to be in really good shape this season, while Boise is in a weaker position than we might typically expect. That’s partially because they’ve lost a couple of heartbreaking games already and partially because they have a tough schedule remaining. SP+ still views them as the second best team in the conference, behind Fresno State.
The last thing that stands out to me is that UNLV is looking like they’re going to have another rough season ahead of them. Their game against the Lobos is probably their most winnable game remaining.
In fact, looking at a plot of the bowl probabilities, we see basically four tiers: The “Almost Certainly Bowlers” (BSU, FSU, Nevada, SDSU, SJSU, USU, Wyoming), “Probably Bowling” (AFA), “Hopeful Bowlers” (CSU, Hawaii, UNM), and “UNLV” (UNLV).
Looking at UNM specifically, we can see that the most likely outcomes for the season are for them to finish with either four or five wins. In fact, those outcomes make up for almost 60% of the simulations. There’s a 13% chance that they only win one more game and a 2% chance we end up in the nightmare scenario of losing out. Let’s not talk about that.
Instead, let’s talk about the good possibilities. These simulations give the Lobos a 25% chance of making a bowl game. That’s not exactly great, but it’s not bad. After all, that’s the same probability of flipping two coins and seeing them both land on heads.
Looking at the individual game probabilities for the rest of the season, their path seems clear. They need to win the one game they’re really favored in (UNLV), win both of the toss-up games (this weekend at UTEP and their final game vs. Utah State), and then hopefully take one from either Air Force or CSU. SP+ really doesn’t like the Lobos’ chances in the other four games, which makes sense, as SDSU, Boise, and Fresno State all look to be the class of the conference while Wyoming has gotten out to a hot start. We’ve seen the Lobos win games like those before, but if bowl eligibility is the goal, you don’t want to have to rely on getting those types of wins if you can avoid it.
Going back to the original question I asked myself before I sat down to do all this work, do the Lobos have a chance to make a bowl game, the answer seems to be yes. Not a great one, but definitely not an insignificant one. And if the Lobos win their next couple of games, their chances start to really look good.