‘Listen to the model’: UNL class applies machine learning to March Madness – KETV Omaha

07

No one ever picked a perfect NCAA March Madness bracket. That hasn’t changed this tournament.But a group of University of Nebraska-Lincoln students followed their professor’s game plan to let a computer algorithm pick the winners for men’s basketball tournament in 2021.”A little bit of fun never hurt anybody,” journalism professor Matt Waite said.So instead of making their picks blindly, Waite’s sports data analytics students were tasked with applying their lessons to take their best shot.The odds of someone picking a perfect bracket of 63 games is 1 in 9.2 quintillion.”We were working through machine learning algorithms to predict things, and it just happened to coincide with the start of the tournament,” Waite said. “You’re presented with a probability that a team is going to win or lose and you can make decisions based off that.”Taking that data, Waite said students could pick one of three algorithms, feed it raw information in the form of possessions, offensive rebounds or three-point percentage and train it to predict future outcomes.Some things, Waite said, as happens in life on the court and in class, he could not see coming.”I did not predict that we would have five students better than 90% of brackets on ESPN’s Tournament Challenge,” he said.”It lives up to the name ‘March Madness,'” junior Thomas Baker said. “Who would have picked some of the lower seeds and all the upsets you’ve seen this year?”Baker follows the sport closely, knowing teams and players’ narratives throughout a season. He said his model ultimately helped him sense a few upsets and put him in first place among the classes’ brackets.”It actually also had Oregon over Iowa, which I was very excited by. I listened to the model 100% when they said Iowa was going to lose,” Baker said.Others, like senior Kaitlynn Johnson, went into this experiment without much hoops know-how, but armed with the knowledge of picking reliable metrics, she’s now in fourth place.”I had to look at like rolling means for possessions, and for points per possession and I had to look at the previous six games,” she said.Waite said he hopes these skills will help students find and share information that impacts communities.”If you want to understand how the world works, if you want to understand how government is working or how society is functioning, you need to know how to analyze data,” he said.

No one ever picked a perfect NCAA March Madness bracket. That hasn’t changed this tournament.

But a group of University of Nebraska-Lincoln students followed their professor’s game plan to let a computer algorithm pick the winners for men’s basketball tournament in 2021.

Advertisement

“A little bit of fun never hurt anybody,” journalism professor Matt Waite said.

So instead of making their picks blindly, Waite’s sports data analytics students were tasked with applying their lessons to take their best shot.

The odds of someone picking a perfect bracket of 63 games is 1 in 9.2 quintillion.

“We were working through machine learning algorithms to predict things, and it just happened to coincide with the start of the tournament,” Waite said. “You’re presented with a probability that a team is going to win or lose and you can make decisions based off that.”

Taking that data, Waite said students could pick one of three algorithms, feed it raw information in the form of possessions, offensive rebounds or three-point percentage and train it to predict future outcomes.

Some things, Waite said, as happens in life on the court and in class, he could not see coming.

“I did not predict that we would have five students better than 90% of brackets on ESPN’s Tournament Challenge,” he said.

“It lives up to the name ‘March Madness,'” junior Thomas Baker said. “Who would have picked some of the lower seeds and all the upsets you’ve seen this year?”

Baker follows the sport closely, knowing teams and players’ narratives throughout a season. He said his model ultimately helped him sense a few upsets and put him in first place among the classes’ brackets.

“It actually also had Oregon over Iowa, which I was very excited by. I listened to the model 100% when they said Iowa was going to lose,” Baker said.

Others, like senior Kaitlynn Johnson, went into this experiment without much hoops know-how, but armed with the knowledge of picking reliable metrics, she’s now in fourth place.

“I had to look at like rolling means for possessions, and for points per possession and I had to look at the previous six games,” she said.

Waite said he hopes these skills will help students find and share information that impacts communities.

“If you want to understand how the world works, if you want to understand how government is working or how society is functioning, you need to know how to analyze data,” he said.

Leave A Reply

Your email address will not be published.