Advanced Stats: Which Conference Is Having The Best March?

Between Selection Sunday and the morning of Wednesday, March 27, there have been 108 games of men’s Division I college basketball played across for different tournaments. There are all kinds of ways to assess a conference’s success in those games. The Big Ten and ACC are tied with the most wins, at 13. The ACC has the best winning percentage, at .813 with—believe it or not—the Atlantic Sun right behind them at .750.

The Big Ten and Big East both sent ten teams to the various tournaments. In the Big East’s case, that’s impressive, as there are only ten teams in that league. But nobody would say that the Big East, with only DePaul left in the CBI, is having a good postseason run.

But looking at teams and wins is a simplistic way of assessing conference performance. This is 2019, and we have advanced metrics we can use. So what do the stats say?

Typically, I use KenPom for articles like this, but for this one, we’re going to use T-Rank instead. The reason is very simple. T-Rank lets you run your numbers with different cutoff points in time. So if I wanted to just look at numbers since Selection Sunday, I could do that.

Hey, look, there’s the Big Ten’s own Purdue Boilermakers at the top. Blowing out Villanova will sure make your efficiency margin look good. I tried to do a breakdown of these numbers, but zero to three games is a very small sample size, and I got completely weird and unreasonable results.

So instead I ran the numbers for the season prior to Selection Sunday, and compared those to the overall numbers for all games, including those in the NCAA, NIT, CBI, and CIT. Those leagues who are having a good postseason run should show some improvement from their end-of-regular-season-plus-conference-tournament numbers.

If I do that, the end result is that the league who has improved the most since Selection Sunday is... the Big South? Believe it or not, the Big South sent six teams to postseason play: Gardner-Webb, Longwood, Charleston Southern, Hampton (who is still alive), Presbyterian, and Campbell. (Campbell is the Fighting Camels, and they have the most badass logo in the country.) Not only did Gardner-Webb hang tough against Virginia (relative to expectations), but Presbyterian also picked up two nice road wins.

The AAC also put up good numbers. A significant portion of that is attributable to UCF, who played Duke far closer than the computers expected. Houston has also rolled through tournament play so far, and USF is still alive in the CBI.

The Pac-12 has been improving, led by Oregon’s emergence as a respectable power conference team. The Big 12 and SEC are also in positive territory.

Which of course leads to the question that most readers of this blog want to know about—what about the Big Ten? According to my T-Rank analysis, the average Big Ten team has reduced their efficiency margin by about 0.11 points per 100 possessions. That may not sound like a lot, but remember, the one to three games played since Selection Sunday make up a small portion of the overall total.

We’ve already talked about Purdue, but the Big Ten team that has actually shown the most improvement has been Penn State. Isn’t that nuts? Pat Chambers’ team hasn’t even played a game, but their strength of schedule has been given a boost by teams they played throughout the year doing well. The Nittany Lions definitely rounded into form late in the season, but who ever heard of a team peaking after their season was already over?

Unsurprisingly, the Big Ten’s poor overall showing is being driven by the three team’s you’d expect: Wisconsin, who got embarrassed against Oregon, and Nebraska and Indiana, who both looked ho-hum before getting bounced out of the NIT. The Big Ten is still the best conference in the country, but since Selection Sunday the computers say we’ve been eighth-best.

T-Rank also gives odds that teams will win it all. Unsurprisingly, the ACC looks best there, with three number one seeds still alive. There’s a 41.7% chance that the national champion will be one of the four ACC teams left alive. The Big Ten is second, though, with a 20.7% chance of hoisting their first trophy in nineteen years.

But if you want to knock the ACC (and I will take every chance that I can get), by the same metric that made the Big Ten look bad, they look even worse. And it’s not their NIT teams that are dragging them down, but their No. 1 seeds. Duke and Virginia both haven’t put up the types of margins of victory that the computers would expect to see, with both of them seeing their efficiency margin drop by more than a full point.

At the end of the day, you can find numbers that support an argument that just about any conference is having a good or a bad March. Readers are encouraged to dig into the numbers for themselves. T-Rank’s site isn’t nearly as user-friendly as KenPom, and some of the best material can be difficult to find, but the advantage is that it’s not behind a paywall. Poke around and build a case for your favorite team (if they’re still alive), and when this weekend is over we’ll hope to see three Big Ten teams in the Final Four.