Our NIT Bracketology, NCAA Tournament Bracketology, and NIT/NCAAT Probabilities are all updated heading into the day. Here’s what’s new:
Moving In: Penn State
Moving Out: USC
The College Basketball Crown is rumored or reported to be awarding its automatic bids by NET. With USC (66th), Penn State (67th), and Iowa (68th) all on top of one another in those rankings, the Big Ten’s second Crown bid is probably going to wobble all the way until Selection Sunday. It’s possible our model will be a day late on catching the proper order of these teams, because in order to get these brackets and probabilities up before the next day’s games start, we have to make our model guess at how NET will react to a single day’s games.
Today, USC’s ahead of Penn State in the NET and in our model’s median final NET across ten thousand simulations. That pushes USC out of the NIT and Penn State into it.
There are ripples from this. Iowa gets the Big Ten’s NIT exempt bid, which is awarded by KNIT, and therefore gets a home game. There was other movement as well. UC Irvine moved up into one of the NCAA Tournament’s first four spots out, which makes them an NIT 1-seed, pushing SMU down to a 2.
There’s also a big question mark here. Penn State has already been mathematically eliminated from the Big Ten Tournament, meaning the Nittany Lions will spend next week idle after closing out the regular season on Saturday. The transfer portal doesn’t open until March 24th, so there’s no urgency for any departing players to depart (or for coaches to start actively recruiting in a legal way), but it’s hard to keep a team together, especially after that team went 15–16, the Nittany Lions’ likeliest final mark.
Our model accounts for this and accounts for complications driven by the Crown, namely that at least the first two teams the Crown invites from the Big Ten, Big East, and Big 12 are rumored to be contractually barred from playing in the NIT. There’s a strong possibility of Crown opt-outs pushing those automatic invites down the list. We don’t know what happens if Penn State declines an invite USC already declined. Are the Nits also banned from the NIT? There’s a lot of randomness baked into our model when it comes to who accepts and declines Crown and NIT invitations.
For what it’s worth, Penn State head coach Mike Rhoades said yesterday that the program will address postseason play as soon as the regular season is over. We will of course be tracking the developments. At the moment, our model has Penn State right on the edge between the Crown and the NIT, but it has the Nittany Lions only about 20% likely to play in the Crown and 28% likely to ultimately play in the NIT. We’re in a new era of postseason college basketball, where the transfer portal has upended tons of norms. This isn’t good or bad, but it’s different. There’s uncertainty.
Model Talk: The AP Poll and Seeding
This is NCAA Tournament-specific, so if you’re here for the NIT, you are dismissed (and God bless you).
We found a little error in our model that’s impacting NCAA Tournament seeding. It’ll be fixed before tomorrow’s simulations, at which point we might see teams like Clemson, Louisville, and Memphis drop in our model’s seed list. Why? Well, we find that the AP Poll released the Monday of Champ Week has some predictive value when it comes to seedings. This makes some intuitive sense: The committee watches basketball games all year where little numbers next to teams’ names tell viewers how good those teams are. For a team like Memphis, who is not particularly good but has strong résumé ratings and a good ranking in the AP Poll, this could really pay off.
Where our model was going too far was that it wasn’t stopping its AP Poll estimations this Sunday. It was running them through Selection Sunday, meaning teams who won a lot of games in their conference tournaments were set up to climb in the AP Poll. Who’s set to win a lot of games in their conference tournaments? Teams in weaker conferences like the ACC and the American.
This all said…Clemson, Louisville, and Memphis are all ranked 16th or better, and it’s very possible this is going to matter. It might not matter as much as it did in 2021, when the AP Poll was extremely predictive of seedings, but it’s possible Bracket Matrix is underrating these teams (and VCU) because bracketologists don’t really look at anything approximating the public vibe.
That said, this isn’t a part of our model we’ve tested particularly rigorously. As we start publishing our subjective Seed List, we’ll be looking out for instances where the AP Poll variable pushes our model away from the Bracket Matrix consensus. We do think the AP Poll matters, but we’re not going to stick our necks out too far on that without conducting a better study.
**
