Editor’s Note: The March Madness of Artificial Intelligence

Published: March 19, 2024

This is the time of year where we all become college basketball experts.

You struggle in how far to take Auburn, which team below the two line should you put in the Final Four, the 12th seeded team you are because 12th seeds have beaten a fifth-seeded team in 19 of the last 23 years according to your friend’s Dad, or can Caitlin Clark finish off her great career in Iowa by cutting down the nets as a Hawkeye one last time.

It is a yearly battle of showing how smart you are among your friends, family and co-workers—until someone wins based on school colors, mascots, or choosing Gonzaga because they like the way it sounds when they say it.

But there is a new wrinkle in your group this year—the cousin’s boyfriend who is using an artificial intelligence (AI) program to help them fill out his bracket.

RFID Journal Live

AI in March Madness

A recent AP story noted AI application in bracketology circles is not new. But the advancement of AI into everyday life will have more people using it this year to win bragging rights and however big the cash prize is.

Before you use ChatGPT or another AI program to come up with the perfect Final Four, know this: machine learning alone cannot solve the issues of limited data and human elements of March Madness, according to those that have been using it for previous tournaments.

“All these things are art and science. And they’re just as much human psychology as they are statistics,” said Chris Ford, a data analyst. “You have to actually understand people. And that’s what’s so tricky about it.”

Incorporating AI to March Madness is not new. Besides the simulations that are run by the NCAA’s broadcasting partners that are on every pregame or betting show, the data science community Kaggle will be hosting “Machine Learning Madness” for the 10th straight year. Kaggle provides a large data set from past results—such as box scores with information on a team’s free-throw percentage, turnovers, and assists—for people to develop their algorithms.

And before reading this article, have you ever heard of Kaggle or used their information to fill out one of your seven brackets?

Sports Providing Life Lesson

Tim Chartier, described in the AP story as a “bracketology expert,” noted the NCAA Tournament’s historical results provide an unpredictable and small sample size — a challenge for machine learning models that rely on large sample sizes.

“We can’t even predict 63 games of a basketball tournament where we had 5,000 games that led up to it,” said Chartier. “The beauty of sports, and the beauty of life itself, is the randomness that we can’t predict.”

I am a big believer that sports provide lesson that translate across all walks of life. In this instance, it serves as a reminder that AI or any machine learning program is a tool to be used that humans ultimately have the control over. It can help provide answers, but it is dependent on the data it receives.

And oftentimes the randomness of life, especially in sports, cannot be predicted or be accounted for when imputing data.

Fill Out Your Bracket

Historically, the Men’s and Women’s Final Fours are be filled with 1 or 2 seeds. But there will be an outlier due to an injury or the nerves of a 19-year-old in front of 18,000 people in the arena—not to mention the millions watching—finally catching up to him at the free throw line.

Those are the human elements that AI cannot account for in basketball. AI will never be able to provide all the answers we want because life is random.

The true Madness of March is believing you will win your pool because you are using AI. Stick with choosing the school your Mom graduated from to win it all.