Math Major Explains Method to March Madness
ROBERT SIEGEL, host:
The NCAA Men's Basketball Tournament is underway, and we're going to hear now about some March Mathness. Thousands of Americans have filled in their office pool brackets, picking winners in dozens of games from this week's for presumably most predictable match-ups between favorite regional number one seeds and underdog regional number 16 seeds, all the way to the presumably unpredictable final four.
Well, Neil Goodson, a senior math major at the College of Charleston in South Carolina, is part of a two-person team that's been working on a system to predict the tournament winners.
And he joins us now from Charleston. Hi yah. Welcome.
Mr. NEIL GOODSON (Math Major, College of Charleston): Hi.
SIEGEL: And you'll note that like some business announcements that are made after the markets close, we are letting your wisdom out to the public here, after people have had to file all of their brackets in their office pools, so you're not going to influence betting all over the country right now?
Mr. GOODSON: Yes.
SIEGEL: First of all, I want you to explain the assignment at the College of Charleston that has led you and your colleague to come up with a system for predicting the NCAA Men's Basketball Tournament.
Mr. GOODSON: Okay. Well, we are in an operations research class, and in that class we look at applying math to real world problems, and we chose to look at the problem of sports-ranking or ranking teams.
SIEGEL: And, in this case, the NCAA ranks 64 teams - it used to be 65 teams for some reason, and they assign them different positions depending on which are the best and which are the weakest teams entering the tournament.
So, where do you pick up? Where do you apply your math there to predict it?
Mr. GOODSON: Okay. Well, we have several different algorithms or models that we look at to come up with our own ranking of the teams. Building matrices by using linear algebra that have, you know, 342 rows by 342 columns. Each column and each row represent the team and we're entering data into those, you know, just win-loss data, different point differentials between two teams, and we look at all these throughout the year. And then, we also weigh each game that we've looked at more heavily as it occurred towards the end of the year.
SIEGEL: Uh-huh, because a game that was played in early March would be more relevant than something that might have done in November or December.
Mr. GOODSON: That's what we think, and we hope that that will prove our models correct.
SIEGEL: And you assume there will be upsets?
Mr. GOODSON: Right. There will be upsets; they always happen. And some of our models do predict some upsets. For example, USC and Kansas state played tonight. We - our model - most of our models have predicted Kansas state, the weaker team, to beat USC, or supposed weaker team. Now, we think that they are a better team.
SIEGEL: Well, you - it's possible that you'll be heard by some USC fans in Los Angeles, well, that game is well underway, so you could be a genius late around this evening if you heard it that time, and you'll be the bearer of some very unhappy news, however, for the University of Southern California.
Well, here's your moment of truth right now. Who's going to win NCAA Men's Basketball Tournament this year?
Mr. GOODSON: Most of our models are pointing towards Kansas, now they are one seed, but they are not the overall one seed. So, we'll see how well we do.
SIEGEL: Well, Neil Goodson, has your professor graded you yet, by the way? Or is he waiting to see the results of the tournament and figuring out what your grade will be then?
Mr. GOODSON: Oh, no. The jury's still out on the grade.
(Soundbite of laughter)
SIEGEL: I see. We have to see results from the games.
Mr. GOODSON: That's right.
SIEGEL: Well, good luck to you.
Neil Goodson, a senior math major and along with Colin Stevenson, the developer of a plan this year to predict the outcome of the NCAA Men's College Basketball Tournament. Thanks a lot.
Mr. GOODSON: Thank you.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.