Every year, millions of people fill out a March Madness bracket, and many of them are completely convinced they’ve cracked the code and this is the year they will bring home the bragging rights. Some of us check rankings, some of us do a deep dive into stats, and some just go on instinct and hope for the best.

From the outside, it can look like most of us are just guessing, although arguably some guesses are more informed than others! But if you step back for a second, there’s more going on than that.

A Bracket Results from a Model

When someone fills out a bracket, they’re making a series of decisions based on a number of things they think matter. It could be the strength of the schedule. Maybe it’s the recent performance. Or maybe it’s a sense that one team just matches up better. It’s a filtered judgment.

What changes from person to person isn’t the process. It’s the data they are prioritizing. Sometimes model building is more explicit, and other times it’s more implicit. Sometimes a model is more sophisticated, and other times it’s maybe based more on what someone had for lunch.

Everyone is building a version of the same thing. It’s a simplified way of making sense of a complicated and unpredictable system. That’s one use of modeling. No one is writing equations or calling it that while they’re doing it, but the thinking is there. We’re deciding what matters, ignoring what doesn’t, making decisions, building connections, and then seeing what happens. 

The Part We Can’t Control

One reason brackets can fall apart so quickly has nothing to do with people doing it “wrong.” It’s because a single-elimination tournament leaves no room to recover from outliers – unexpected events. One game decides everything, and sometimes just a bad moment in a game shifts the outcome. A cold shooting night, a bad matchup, an injury, or a sudden switch of momentum late in the second half is all it takes to break a bracket.

There’s also a lot of noise in a March Madness bracket, and no matter how much data someone uses, the noise never fully goes away.

That’s actually a reality of many modeling situations, and similar features and similar thinking show up in much bigger decisions.

The Perspective is Familiar

What makes March Madness interesting from a modeling perspective is how familiar the thinking actually is. 

In finance, people build forecasts using past data, trends, and a lot of assumptions. The models can get pretty sophisticated, but at the end of the day, they still come down to someone’s judgment.

Public policy decisions aren’t made with perfect information either. A lot of the time, they’re made in pressured situations, with real consequences, and without knowing how things will play out.

And in engineering, the tradeoffs are pretty much constant. Improving efficiency might increase cost, but increasing safety might limit performance. Every decision shifts something else in a complex system.

None of these situations come with certainty, and they all require someone to decide what variables matter most and move forward anyway.

Bringing Brackets Into Class

For teachers, this is one of those times you don’t have to work to get their attention. They know the teams in the tournament. They’ve seen the highlights. A lot of them are filling out brackets at home or with friends anyway. 

That makes it a natural starting point for a modeling conversation. Ask students how they’re making their picks. What are they looking at? What are they ignoring? What are they prioritizing? Why? The answers are usually all over the place, which is exactly what you want.

From there, you can start to transform it into something more intentional. What would a more structured approach to the bracket look like? What variables might matter more? How would you test whether your approach is any good?

Now it’s not just a bracket. It’s a model they’ve built.

A Natural Connection to Modeling

This is the same kind of thinking students tackle in COMAP’s math modeling contests.

There isn’t a clean path laid out. They have to decide how to approach the problem, what to include, how different pieces relate, and then justify the choices they make.

Using March Madness brackets makes it easier to shift into a mathematical modeling mindset. Students are already invested, the context is familiar, and it gives them a way to experience modeling before they ever call it that.

It’s More than Just a Bracket

After the tournament ends, most students won’t remember what teams they picked, but they’ll remember what it felt like to try to predict something unpredictable. Because whether it’s a tournament, a business decision, or another real-world problem, the process of modeling is the same: Take in information, decide what information matters, make decisions, and learn something from what happens next.