Before we show your our model’s FCS Bracketology, ahead of today’s selection show (Noon Eastern Time, ESPNU), a few notes on where we think our model’s wrong:
- It has Montana State seeded too low. We expect the Bobcats to be the 2-seed.
- It has the Ivy League collectively treated too kindly. We don’t really expect Dartmouth to make the field, and we don’t expect Harvard to get a bye.
- It has Alabama State on the bubble despite Alabama State seemingly choosing the Turkey Day Classic over the FCS Playoffs. (We never heard back explicitly from them about that, so we left Alabama State in our model’s simulations. For what it’s worth, our model actually has Alabama State as the last team in and Dartmouth as the first team out.)
- It appears overconfident.
We’ve talked plenty before about what went wrong with those first two things, mistakes we made when building the model. The fourth thing—that overconfidence, teams like Austin Peay having less than a 1% playoff probability across our 10,000 simulations—is new. To be fair, if you adjust the Ivy League’s treatment and remove Alabama State, you find teams like Austin Peay (and New Hampshire and Northern Arizona) a little bit more in the mix. But we’ll have upgrades to make this offseason. More on those at the end.
Now, the bracket:
Moving In: Youngstown State
The Penguins survived Northern Iowa on the road, and that should be enough to get them across the finish line, especially with so many others on the bubble failing to survive.
Moving Out: Lamar
Again, if we were making our own bracketology—separate from our model’s—we’d probably overrule it and put Lamar in ahead of Dartmouth. The committee’s in-season rankings reveals indicated they were lower on the Ivy League than our model expected, presumably because Ivy League teams play fewer nonconference games, making their lack of losses less impressive. Lamar doesn’t have a great résumé—they’re 8–4 overall and that loss yesterday to McNeese was a rough closing argument—but it’s the best of those we think are eligible for that 24th spot.
Moving Up: Abilene Christian, Yale, South Dakota State
Abilene Christian finished the job, winning the UAC automatic bid. Yale finished the job, trouncing Harvard to win the Ivy League automatic bid. South Dakota State finally pulled out a win, taking down North Dakota on one heck of a catch by Grahm Goering. All three move up into home games in our model’s projection.
Moving Down: Illinois State, Dartmouth, Monmouth
Illinois State wasn’t going to host this week anyway—thank you to the reader who reminded us of the IHSA State Championships happening in Normal—but now the committee doesn’t need them to formally decline. Southern Illinois played spoiler, and played it emphatically.
Part of why we’re so concerned about our model’s assessment of Dartmouth is that they lost their own finale, falling to Brown on a pick-six late in the fourth quarter.
Monmouth entered the weekend 9–2 and, per our model, 98% playoff-likely, needing only to beat 1–10 Albany to comfortably make the playoffs and likely host at least one game. Albany beat the Hawks, at one point leading 31–3.
Bubble Probabilities
Between automatic bids and at-large locks, our model has 21 of these 24 teams at 100% playoff probabilities. That’s the kind of number that makes me nervous, but I do think the model’s going to be right about all 21 of those teams. Beyond them…
- Monmouth: 80%
- North Dakota: 75%
- Alabama State*: 58%
- Dartmouth: 56%
- Lamar: 25%
- Southern Utah: 3%
- New Hampshire: 2%
- Northern Arizona: 1%
Technically, our model has Alabama State as its last team in, not Dartmouth. We talked about that above, and about Lamar and Monmouth. North Dakota’s an interesting one—they’re only 7–5 but they played a murderous schedule and our model thinks they’re the sixth-best team in the FCS at this moment in time.
Among teams our model’s writing off, I’m more curious about New Hampshire, Northern Arizona, Austin Peay, and Southern Illinois than I am about Southern Utah. UNH is 8–4 as a CAA team, and while the CAA isn’t what it once was, we do see the FCS committee lag behind reality when it comes to acknowledging the strength of some conferences. (See: The UAC, where Tarleton State should be a top-four seed today and might not be.) Northern Arizona and Southern Illinois are both 7–5. SIU gets dinged for playing a non-Division I school, but NAU didn’t do that. (NAU did lose by 20 yesterday at Weber State, so it’s not like NAU’s an innocent victim here.) I don’t think Austin Peay will make it, because of the UAC thing, but they beat an FBS team (Middle Tennessee), played another FBS team (Georgia, who beat Texas by more than they beat Peay), and nearly upset Tarleton yesterday in the finale.
National Championship Probabilities, Model Talk
At the moment, North Dakota State is 52% likely to win it all in the eyes of our model. Montana State is 21% likely. But, they’re on the same side of our model’s bracket, and as we said above, they probably won’t be on the same side of the real bracket. We’ll have the new probabilities up tomorrow morning. I’m curious how high that pair will rise. 80% combined? 85%?
As for next year:
We’ll probably want to increase our model’s uncertainty next season, and we’ll want to adjust it so that it uses the committee’s in-season rankings reveals, something we foolishly scorned when building this because they’re not that helpful in the basketball world (what the FCS committee does with those reveals is more similar to what the NCAA Tournament committee does than to what the CFP committee does). We’ll want to increase the importance of wins and decrease the importance of losses, or at least see if that can help when adjusting to this new world where Ivy League teams play in the playoff. If it doesn’t, we can always just add some conference-level adjustments in order to bring things in line. We’ll know a lot more after today.
**
