Ever since Kansas’ loss in the Elite Eight to Villanova last week, a major talking point has been Bill Self’s performance in the NCAA Tournament.
Though the 13th-year KU coach has had nearly unprecedented success winning conference regular-season titles, he has only two Final Fours and one national championship with the Jayhawks while earning high seeds nearly every season. Self’s KU teams have reached only one Final Four (2008) in six attempts as a No. 1 seed in the NCAA Tournament.
For me, it’s been hard to compare his record to other coaches simply because of the whims of the NCAA Tournament. Sometimes, the bracket opens up for you. Sometimes, like this year, you get stuck with a 2 seed playing as well as any 1 seed, and a loss might be more understandable with better context.
One potential way to better compare coaches’ NCAA Tournament success would be to use Ken Pomeroy’s win-probability numbers, which go back to the 2009-10 season.
Premium content for only $0.99
For the most comprehensive local coverage, subscribe today.
Here’s an example: In 2009-10, Pomeroy gave KU a 95.1-percent chance of beating Lehigh and a 74.8-percent chance of beating Northern Iowa based on both teams’ rankings. Add those together, and we would have expected KU for 1.7 victories in its first two games of the tournament. As we know, KU fell to UNI, meaning it only picked up one win instead — meaning we could say Self was -0.7 on expected wins that particular tournament.
Using that method, let’s take a look at KU’s tournament performance over the past seven seasons. (Mobile users can tap here to view the table.)
There are positives to looking at things this way. For one, we are comparing KU to itself to see if it over- or underachieved its own play that particular season. This also takes into account competition better than simply looking at seeds, as oftentimes, the committee mis-seeds teams, which can create a tougher draw (see Wichita State as a 7 seed in 2015).
Some potential negatives: This penalizes teams a bit for strong regular seasons, as it is comparing what the team should do based on its past performance. There also is no penalty for missing the NCAA Tournament, which Self has not done in any of his seasons.
The data also only goes back to 2010, and there’s a chance for huge gains from one long tournament run, though perhaps that should be considered a positive for coaches who can overperform when it matters most.
In any case, Self is -0.53 wins above expected over the last seven seasons with KU. Here’s how that compares to other top coaches at blue-blood programs. (Mobile users can tap here to view.)
A couple of interesting notes: Self has “overperformed” in two of seven NCAA Tournaments with KU … and one of those was this season. KU’s three victories were actually more than expected, a result of KU receiving a tough second-round draw in Connecticut and an extremely difficult Elite Eight opponent in Villanova (KenPom actually had the Wildcats as a slight favorite).
Kentucky coach John Calipari tops this list easily, helped by three strong postseason runs. We can argue whether 2013-14 UK deserves credit for having played itself to a No. 8 seed with 10 regular-season losses in the same year it advanced to the NCAA title game, but even with that criticism, we have to give Calipari credit for winning those late games.
Coach K ranks second, while Tom Izzo — hurt by this year’s early exit — comes in third despite a strong postseason reputation. Roy Williams, who could still improve or hurt his numbers this season, has been one of the most steady NCAA Tournament coaches on the list.
Self’s biggest issue lately seems to be in coin-flip games. In the last four seasons, KU is 7-1 when it has a 68-percent win probability or better. In three games with win probabilities between 49 percent and 52 percent, though, the Jayhawks are 0-3.
It’s not just that Self has had early losses. Unlike his peers, he hasn’t had many deep runs to balance out those times when his team has been bounced early.
Move the sample size back to 2008, of course, and these numbers would look different. Perhaps if KU wouldn’t have had such successful regular seasons, its postseason performance wouldn’t be as damning either.
In this exercise, though, it’s clear that over the last seven seasons, Self isn’t performing up to the level of his rivals — at least not in the most important weeks of the season.