NCAA tournament selection process to consider using different metrics, analytics

printable NCAA Tournament bracket
Mike Ehrmann/Getty Images
5 Comments

Big news today coming down from the NCAA selection committee. In a story published on NCAA.org, Dan Gavitt, the senior vice president of basketball for the NCAA announced that there would be a coming together of the minds behind some of college basketball’s best metrics.

Ken Pomeroy (KenPom), Jeff Sagarin (Sagarin ratings), Kevin Pauga (KPI) and Ben Alamar (BPI) will all be meeting in Indianapolis next Friday to discuss integrating more advanced metrics into the NCAA tournament’s selection and seeding process. Less RPI is always a good thing, and with those four gentlemen in a room together, you can bet your mortgage on that line of thinking being pushed forward.

One of the goals of this meeting is to develop some kind of a composite metric, another tool that the selection committee can use to differentiate teams and to find a fair way to get the 68 best teams in the country properly seeded into a bracket.

I love it.

I don’t want to delve too much into why the RPI is an ineffective metric, but rest assured, there’s a reason that the smart people covering – and coaching in – this sport can recite KenPom rankings.

There are, however, two things that I do hope will come out of this:

1. Can we end the over-reliance on arbitrary cut-offs points for ranking wins? There shouldn’t really be a difference between beating the No. 50 team and the No. 51 team, but there is. How often do we talk about “Team X has this many top 50 wins”? (Hint: The answer is A LOT.) And I would rant here about ensuring that the value in winning road games shows up in this metric, but I know for a fact that the guys getting invited to this meeting understand that better than I do. It will be discussed.

2. More importantly, I have a plea for the gentleman heading to Indianapolis next weekend: Make sure that, when you leave that room, there is the understanding that the actual results still have to matter.

One of the flaws in the RPI is that it doesn’t take margin of victory into account. For example, Xavier lost by two points at Colorado and by 25 points at Villanova. In the RPI, Colorado’s win has the same value as Villanova’s win. Even my 16-month old son knows that’s silly, and he throws a tantrum when I make him eat with a fork. The metrics that incorporate the difference in those two results are always going to be more accurate.

But the concern, then, is just how much of a difference there will be in, say, a one point win and a one point loss. The fact of the matter is that both of those teams essentially played each other to a standstill, right? Whether or not a shot goes in on a final possession isn’t going to change how good either team is, can be or will be. Remember this finish between Butler and Gonzaga? Remember the big to-do everyone made about Brad Stevens’ reaction on the sideline? It was one of the most exciting and memorable shots of his coaching career, and he never even unfolded his arms? He was already walking towards Mark Few when the shot went in.

“That was a really lucky play for us. Great steal by Rose, got it off in time, made a great play,” Stevens said after the game. “I don’t think it tells the whole story about how good these two teams are, and shouldn’t take anything away from Gonzaga.”

And, to a point, he’s right.

Losing a close game on a fluky buzzer-beater doesn’t make you any worse of a team.

But it should have a tangible benefit to the team that won the game. You should want it to have a tangible benefit to the team that won the game, because that’s what makes these moments so exciting. That’s why Hinkle Fieldhouse went absolutely bananas when that shot from Jones went in. That’s why we collectively lose our minds anytime someone hits a big shot. That’s why, say, the three that Malik Monk hit to give Kentucky a win over North Carolina back in December was so meaningful. It didn’t mean that the Wildcats were suddenly that much of a better team that the Tar Heels, but it did mean that they had that win on their résumé.

And that needs to matter.

Winning needs to matter.

There should be a big difference between a one point loss and a 30 point loss, but there should also be a difference between a one point loss and a one point win.

Because otherwise, if winning the game doesn’t actually matter, what the hell are we doing all of this for?