NCAA Tournament Probabilities: UConn’s Path Clears

We wrote earlier this week about the difficulty of UConn’s NCAA Tournament path. It wasn’t just perception. It was a measurable thing. The presence of so many good teams in their region nearly guaranteed UConn would have to go through multiple teams who were around the best on their seed line.

Nearly.

After checking in Thursday at only 15.5% likely to win the NCAA Tournament, per our lukewarm kenpom model, the Huskies are up to 20.5% today. Auburn went down. That is a big, big deal.

The thing with Auburn all season is that they were so good when they were good that it was hard to know how much to trust numerical evaluations of them. They were a dominant team at times, but they no-showed at times, and Bruce Pearl had coached successfully in single-elimination settings before but also lacked the good–guards blueprint said to be so essential to NCAA Tournament success.

I’m not sure we have an answer yet on Auburn. It was just one game. Yale is not a terrible team. But we no longer have to ask the question.

Our probabilities for the rest of the NCAA Tournament can be found here as well as below, for archival purposes. (The latest NIT ones are down there too, but they haven’t changed much from the last time we posted about them.) We’ll have thoughts on each of yesterday’s games in here as well. But first, we want to talk again about path, and this time, we’re going to do it with a different approach: Kenpom ranking vs. probability ranking. Here’s the list, for the top fourteen remaining teams:

There’s not a lot there.

Especially at first glance.

Upon second glance, you can see UNC has a bad path, drawing Michigan State in the second round while Alabama most likely awaits in the Sweet Sixteen. You would expect a 1-seed to have a better ranking in the second column than the first, but the opposite is happening, which speaks to how much better Michigan State is than the sum of their performance this year (Arizona is the best 2-seed left, so that’s part of it, but Alabama is only the third-best 4/5-seed remaining, so it’s mostly the Michigan State piece and then a little Arizona).

But overall?

The path thing has already worked itself out. Auburn’s elimination has cleared the way for UConn if it wasn’t clear enough already. The tournament fundamentally changed in that ugly, ugly, ugly final minute yesterday afternoon.

Thoughts on yesterday’s other 15 games:

• We were worried specifically about the Vladislav Goldin matchup for Northwestern, and Goldin was efficient, but FAU didn’t seem to go to him as much as they could have? Credit to Northwestern defensively, and credit to Northwestern especially for limiting the guy to nine boards. That was the difference, even before FAU’s guards failed to do what FAU’s guards failed to do.
• For those curious about NET manipulation and kenpom reactions to blowouts and things like that, UConn’s adjusted efficiency margin on kenpom only improved by 0.4 points per 100 possessions after beating up Stetson. That’s not very much, and I think we should all agree that not everyone could have done that to Stetson, so some improvement makes intuitive sense.
• Colgate, as suspected, didn’t put up much of a fight against Baylor. The Raiders drop to 0–5 in NCAA Tournament games under Matt Langel, who is a great coach but hasn’t found that next necessary gear. Can it be found? Or is Colgate at its ceiling?
• Was Clemson crushing New Mexico the biggest win compared to the betting line? Was that the spread’s biggest miss of the first round? It is fair to view the ACC as potentially having been undervalued. That doesn’t mean it’s a great league, but these results aren’t entirely meaningless.
• San Diego State looked bad, and as it often goes with those guys, it’s hard to know whether that was because of how they play or if they played badly. They’re not particularly inspiring in the potential UConn matchup, and it’s hard to trust Jaedon LeDee to tear up Yale after Johni Broome’s big game wasn’t big enough.
• This was a little obscured by how bad the first half went for Marquette, but Tyler Kolek looked perfectly healthy, which is something that was really surprising to me at least. It’s possible I went too far into baseball brain because it was an oblique injury, but I did not expect him to be himself on the offensive end. That is a big, big deal.
• Colorado/Florida was a great game and also wasn’t? Some rough stuff from the Buffs down the stretch, right up to settling for a midrange jumper from behind the backboard with four seconds left. But KJ Simpson got the roll, and Colorado remains dangerous.
• Texas A&M shot 57% on threes, with Wade Taylor seven for ten all by himself. When the Aggies are good, they’re really good, and after a lot of folks were excited about this team entering the season, it wouldn’t be the craziest thing to see them put together a run.
• Longwood was the best 16-seed on paper. Puts some perspective on how shocking it remains that Fairleigh Dickinson once won a first round game.
• People were so eager for Purdue to struggle that they treated a nine-point halftime lead over a 16-seed as Purdue being in danger. The Boilermakers’ win probability at that moment was, per ESPN, 98.2%. No trouble at all this year. The monkey isn’t fully off the back, but that’s at least one demon slayed.
• Utah State looked great, and while TCU had limped into the finish line, the deal with Utah State not playing any high-majors in the regular season was that they were more of an unknown than some other Mountain West teams. That could have gone either way, and I think a lot of us—myself included—overweighted the possibility it would go the way where Utah State wasn’t that good. They looked great. Much scarier situation for Purdue than if TCU had won, even if TCU and Utah State are now about equal in kenpom.
• Vermont, like Colgate, still hasn’t won an NCAA Tournament first round game in this otherwise highly successful stint under their head coach. As with Colgate, this isn’t necessarily a knock, and the sample is very small, but the Catamounts haven’t gotten over the hump. They’ve been better than Colgate, to the degree where you’d have expected a win by now (not really the case with Colgate), but it hasn’t come, and they don’t necessarily look any closer right now than they’ve sometimes been.
• A performance just below one point per possession was not a scenario in which I expected James Madison to beat Wisconsin. The Badgers were just too bad in their shotmaking early, and they could never catch up. Credit to JMU for a lot of that on the defensive end, and for maintaining the gap once they’d built it, but that’s one that’s going to eat at Wisconsin, because they didn’t play anywhere near their best.
• Absolutely wacky game between Alabama and Charleston, which is what you’d hope for between Nate Oats and Pat Kelsey. So many easy looks. Grand Canyon is tough, and UNC or Michigan State would obviously be a tough opponent as well, but the path isn’t bad for the Crimson Tide. Slightly overperforms on that list above.
• Grand Canyon was the better team against Saint Mary’s. I saw a lot of small-time college basketball journalists (much bigger than me, to be clear) discrediting Bryce Drew’s work because Grand Canyon is investing heavily in basketball, but he’s still brought a program to new heights, and generally small-time college basketball journalists break their backs bowing down to celebrate that. The school being for-profit and drawing some fraud-adjacent investigations doesn’t change what Bryce Drew has done there, or what these players have accomplished.

We’ll have a post up tomorrow reacting to today’s games. Good luck out there to those with rooting interests, in whatever form they come.

For the Archives