How I Correctly Picked Villanova to Win the 2018 NCAA Men’s Basketball Tournament

Plus, a full breakdown of how my data-optimized bracket performed

My bracket on ESPN
  • Round of 32: 99.1st percentile
  • Sweet 16: 95.8th percentile
  • Elite 8: 79.4th percentile
  • Final Four: 86.2nd percentile
  • Championship: 94.5th percentile
Visualizing optimal strategy, from Optimizing Your 2016 March Madness Bracket by Robby Greer

The Final Four

Villanova

Villanova wasn’t a very unorthodox pick to win the tournament. In fact, Villanova was given a 17.96% chance of winning the tournament by FiveThirtyEight, the highest probability given to any team. However, optimal strategy weighs both probability and crowd pick distribution, and even considering those factors, it still recommended Villanova. For comparison, last year, Villanova was also the most likely to win the tournament, but the math told me to pick Gonzaga. Gonzaga beat Villanova in the Final Four before losing to UNC in the championship game.

Ranking of picks by optimal strategy (expected points over crowd — column G)

Michigan

Michigan was a slightly overvalued team; 3.7% of the crowd picked them to win the national championship, but they only had a 2.37% chance, according to FiveThirtyEight. Being overvalued is common with strong college sports programs, although they were less overvalued than true blue bloods Duke and Kansas.

Kansas

When I saw how the math had filled out my bracket, I was skeptical that the chips would fall according to plan in the Midwest region. However, it ended up being one of my best-performing regions, making 12 of 15 correct picks. I correctly picked Kansas and Duke to make it to the Elite Eight, despite a tough road for Kansas: the (presumably) best 16 seed in history, a strong Seton Hall team, and a capable Clemson team.

Loyola-Chicago

This year’s Cinderella takes its place alongside the VCU’s and George Mason’s of the world who have had their school’s name put on the map by their basketball team’s strong performance in March.

Notables

UMBC over Virginia

Why didn’t my model pick the first #16 upset of a #1 seed, when UMBC knocked off Virginia? Let me count the ways:

  1. FiveThirtyEight gave UMBC a 2.15% chance of winning the game
  2. Virginia was the #1 overall seed, not just any #1 seed.
  3. My model projected that Virginia would earn my bracket an average 125.92 points in my pools.

Before You Go

Enjoyed this post? Give it a few 👏, share it with friends, and follow me.

District of Columbia fanboy living in CDMX with @egbarnett. Product Manager at @TheoremCo.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store