NCAA Tournament Probabilities: UConn, Purdue, and…Duke?

After all of that, Houston is done, and Duke is the third-likeliest team in the country to win this national championship.

Like nearly everyone else, we were a little gutted by Houston’s loss last night, especially given how it happened, with yet another injury striking the Cougars at the worst time. That the injury came to Jamal Shead, maybe the most admirable player in college basketball, made it even worse.

Our NCAA Tournament probabilities are updated, and they indicate about a 5-in-8 chance that one of UConn, Purdue, and Duke wins this championship. That number’s probably a little higher in the real world, because of UConn’s home-crowd advantage tonight (we dove into that on Thursday), but it’s more likely than not that the champion is one of those three, and it’s about a coin toss that the champion is either UConn or Purdue.

Some quick thoughts on yesterday’s action:

Yet another good team crapped out behind a terrible shooting performance. This time, it was Marquette.

The argument against college basketball, from NBA fans, is that it’s too ugly. LeBron James complained earlier this season about how little spacing there is in the college game. Is that fair? Ehh. It’s subjective. If you want to watch beautiful basketball, good luck finding it in America before the NBA Playoffs. If you want to watch wholehearted competition, college basketball routinely delivers.

Either way, the reason ugly basketball is so prevalent in the college game is related to two correlated things:

First, there’s just not enough talent in the world to fully stock college basketball with deadeye shooters. This will probably never change. The three-point line’s been around long enough that whatever progress comes from seven-year-olds shooting threes has probably happened.

Second, a sport so inclined toward single elimination play disproportionately rewards things that don’t slump: Offensive rebounding. Defense. Booty ball. Even the midrange game, with its lower efficiency, can be a big asset in March. We’re not still stuck in the days when many thought Villanova couldn’t win a title because they shot too many threes, but you better be a damn good shooting team right now, or you better be able to adjust, or you better be great at something else. The easiest thing for coaches to prioritize is to be great at something else. That’s how you get approaches like Houston’s. But the thing that Marquette and Arizona both failed to do this weekend was adjust when the situation proved that necessary. They ran out of tricks in the bag. NC State moves on.

Kyle Filipowski has become a great villain, and Jared McCain is leaning into the Duke ethos as well. Jon Scheyer remains kind of hard to hate as a coach. He’s a little too genial and not quite smarmy enough. But Duke? The Duke hate is back.

In the Midwest, how about Braden Smith? 15 assists. Two turnovers. And he’s only a sophomore! Purdue has a lot of season left, but when the time comes to look ahead to the post-Edey era, the presence and development of Braden Smith portends not only good Purdue basketball, but a big pivot towards aesthetic Purdue basketball. A lot hinges on tomorrow, but if tomorrow does go well for Purdue, there won’t be many programs in a better spot than Matt Painter’s. It’s not just Zach Edey. And when he’s gone, people might suddenly really enjoy watching this team play.

Speaking of tomorrow: Tennessee did it. It was tense and back-and-forth, but Tennessee made it happen. I wonder if Creighton’s defense did them in here. Tennessee’s solid at protecting the ball, but they had no issue whatsoever holding onto it. Creighton seemed so focused on limiting Dalton Knecht’s open looks that they didn’t do anything else defensively, and paradoxically, they’re the kind of team you’d expect to maybe just try to outgun Knecht. I wasn’t watching all that closely—I was more focused on the Duke/Houston game—but I was a little surprised in how Creighton approached it. From what I did see, it was the defensive version of the failure–to–adjust thing that took down Marquette and Arizona.

As for today:

UConn and Illinois are the two most efficient offenses in the country, adjusted for opponent (that’s what kenpom’s AdjEM is). The difference, of course, is that UConn plays much better defense. But! Illinois’s defense has played better than its average self within the tournament, or at least produced better results (Iowa State may have helped them out with the finishing on Thursday). Ultimately, you would think that UConn would have so many more ways to beat Illinois that they’d get it done, but if Illinois throws the booty ball curveball and goes full 2023 Jalen Pickett with Marcus Domask…maybe. Illinois’s paths seem to be either to hope they can outrun UConn (doubtful, but a track meet’s a track meet) or to muck it up. I think the mucking it up option might work. This isn’t any brilliant basketball thought by me, but: Look for a lot of screens to try to get Domask isolated against Cam Spencer.

Clemson and Alabama are two of the three worst teams left in this. Alabama, though, is significantly better. I think there’s a temptation with Alabama to say they’re “due” for an off night. That isn’t normally how it works, though, and…wasn’t that night what happened against Grand Canyon? Alabama, more than any other team in the tournament, gains more ground the more the sample size shrinks. They do best in the broadest sample, but not when the sample is single-elimination. That isn’t a broad sample. It’s a series of tiny samples. If we were ranking teams when they’re playing at their best, it probably goes 1) UConn, 2) Purdue, 3) Alabama. Despite Alabama being the third-worst team left in this.

For the Archives