Enigma has a database of 31 million U.S. business locations. The database includes card transaction revenue, industry classifications, business registrations, and a lot of other things that have absolutely nothing to do with basketball. So naturally we tested all of it against the results of NCAA basketball tournaments, past and present, to see if any variety of business data somehow correlates with winners (and then worry about causation later).
So, we ran 590 business metrics against every March Madness game over the last ten tournaments: pizza shop revenue, dispensaries per church, the ratio of Pilates studios to BBQ joints, the number of people who list themselves as CEO on their business registration. For each game, we checked which team’s home territory (city and state) had more of a given metric, and whether that team won.
Some of the results were genuinely strange.
The ratio of divorce lawyers to wedding venues, for instance, happened to pick the winning team in 62% of games this year. It also picked winners at a 61% rate across the previous five tournaments. Of all the strange metrics we tested, this was the one that kept showing up, year after year, for reasons we cannot begin to explain. Delis per capita went 66%. Average pizza shop revenue, measured from actual card transactions, went 64%. We have no idea why.
Test 590 things and some will look good by chance. Different metrics spike and collapse from year to year, and nothing here constitutes a prediction system. But as a way to fill out a bracket when you don't know anything about basketball, or to file a side bracket next to your real one just for the absurdity of it, we think it's pretty hard to beat. So we also made a website for fans and anti-fans alike to explore spurious connections between business data and the world of sports: https://bracketology.enigmacorp.dev/
The best part of this project wasn’t the bracket-building itself, but searching through the textured landscape of data about each place where a March Madness university resides. Four cities made the Final Four, and each one has a very specific commercial character.
Ann Arbor, Michigan (Champion)
Storrs, Connecticut (Runner-up)
Tucson, Arizona (Final Four)
Champaign, Illinois (Final Four)
Compared to the teams they eliminated in the Elite Eight, this year's Final Four cities had:
93% fewer people calling themselves CEO on their business registrations. Champaign had 70. Durham, home of the Duke team they beat, had 11,624. Ann Arbor had 172. West Lafayette, Purdue's home, had 657.
The Final Four cities also had far more dispensaries per religious organization, and far more Thai restaurants per Mexican restaurant. Again: we are not claiming that humility, cannabis access, or Thai food wins basketball games. But we did run the numbers.
[GRAPHIC 2 — CHAMPIONSHIP MATCHUP goes here]
Going into the final, nine of the top ten metrics sided with UConn. Connecticut had more delis, higher pizza shop revenue, more divorce lawyers per wedding venue, and fewer haunted houses (4 versus Michigan's 30).
Michigan won anyway.
The only metric that sided with Michigan: average business revenue growth. Michigan's businesses were growing faster than Connecticut's. As for the rest of this analysis, make of it what you will.
91 of our 590 metrics happened to pick the exact Final Four. Among them: psychics per capita. We'll leave it there, but you can learn much more in last year’s newsletter spotlight on the mystical-industrial complex.
And as for retrospective bracketology, you can explore the full dataset at bracketology.enigmacorp.dev. Next March, when someone invites you to a bracket pool and you've never watched a game in your life, you'll know where to go.