Michigan State’s loss to Northwestern looked bad, but in the heat of the moment, it was the kind of thing one doesn’t blink at in college basketball: A middling team beating a good one in a conference game at home. It happens nightly, just about, at least throughout January and February.
But in hindsight, it looks much worse. We updated our model’s bracketology today, and the Spartans are down to a projected 9-seed. Our model might be overreacting. We also thought it was overreacting to Kentucky’s early struggles. Tom Izzo’s team’s résumé, to date, is mostly that loss to Northwestern. The road win over Duke is good, but it’s not as good as it appeared at the time. With a deep Big Ten looming, Sparty’s got work to do.
Here’s what else we noticed:
Moving Up: Creighton, Saint Louis
Creighton bounced back, demolishing St. John’s and escaping a treacherous trip to Storrs. As for Saint Louis, as disappointing as the Minnesota loss felt, the Billikens gained a lot of ground with what they did to Indiana State and NC State. They’re the class of the A-10, and with some quirky A-10 movement lower down the seed list, that’s a good thing to be.
Moving Down: Michigan State, Oklahoma State
There isn’t much room for error in the Big 12, and Oklahoma State’s loss to TCU did a number on them. They’re still projected to make the field, but their margin for error isn’t large, and their schedule’s one where losses can mount fast.
Moving In: Colorado, SMU, Richmond, VCU, St. Mary’s, UConn, Belmont (auto-bid), Bryant (auto-bid)
Nebraska-Omaha and Washington aren’t good, but beating the two by a combined 75 will get the numbers to notice. They’ve noticed Colorado. Meanwhile, SMU slid across the line with a win over ECU in their conference opener, and Richmond comes back around after winning at Vanderbilt and beating Loyola on a neutral floor. VCU rocked Western Carolina, helping them cross the threshold with the assist from some of that weird movement further down in the A-10 (haven’t dug into this, so take it with a grain of salt, but it looks at a high level like the A-10 is stratifying, creating easier bad wins and more rewarding good wins for teams like VCU, Richmond, and SLU). St. Mary’s is suddenly 8-1, and while their best wins aren’t great (Northern Iowa and South Dakota State aren’t even in NIT at-large bid territory), they’re helped by BYU’s rise as our model looks forward, and the defensive display against Colorado State was impressive even if Colorado State’s nothing special. Quiet things catch the model’s eye, because sometimes the quiet things change expectations. Finally, UConn played Creighton tight, in a rare instance of a game that helped both teams’ standing (Creighton avoided a very possible loss, UConn improved its outlook going forward).
In automatic bid territory, Belmont moves in as the OVC favorite after Murray State’s loss to Austin Peay, and Bryant—who’s risen from 312th in KenPom to the top 200 and plays like lightning offensively—took the NEC favoriteship from Mount St. Mary’s.
Moving Out: Marquette, Oklahoma, NC State, Mississippi, Maryland, Penn State, Murray State (auto-bid), Mount St. Mary’s (auto-bid)
Marquette’s an interesting case. They’ve been rather inconsistent, and while plenty of teams will be inconsistent, Marquette’s done it earlier, making it noticeable. Their losses to Seton Hall and Xavier were understandable, but those still pushed them below the waterline. Oklahoma played well enough, but their schedule is daunting: The top five in the Big 12 are great. Possibly too great for the Oklahoma schools and TCU to have much of a chance. NC State was already close to the bubble, so the SLU loss pushed them over, which is the same thing that happened to Mississippi after their loss to Dayton. Maryland and Penn State didn’t play, but movement around them edged them out. The field is really, really close by nature, with single games at the end of the season moving teams around by one or two seed lines. Little things right now have a multiplied effect in our model’s projections. Which is part of why things are so chaotic week-to-week right now in these brackets.
Last Notes: Wisconsin and Louisville
Wisconsin and Louisville moved three seed lines in opposite directions following the Badgers’ win over the weekend, with Madison’s finest even taking over the Big Ten favoriteship, by our model’s estimation. This is a good time to note that our model, and the systems that feed into our model (the same ratings systems the committee uses), don’t account for absences due to illness, injury, what have you. Personally, I don’t think Wisconsin’s the Big Ten favorite right now, and I think our model’s forward-looking portion is overreacting to the result because it and its feeders don’t know how shorthanded Louisville was. But in terms of the impact of the result of that one game, my best guess is that it’s going to be inconsequential enough for both teams that its role in the ratings goes relatively unchecked by the committee when all’s said and done. Take all of that for what it’s worth.
