In college basketball, it pays to pamper your players. Connecticut, Duke and North Carolina have each won the national championship recent years, and it’s no coincidence that the three are also among college basketball’s biggest spenders. With thousands of NCAA tournament brackets being filled out this March, with reasons for team selections ranging from point differentials to favorite mascots, why not bet on the teams that invest the most in their on-court success?
We looked at last year’s NCAA tournament bracket and financial filings that each school makes with the Department of Education, and we found that operating expenses are a surprisingly strong predictor for tournament success. Operating expenses are defined as “lodging, meals, transportation, uniforms, and equipment for coaches, team members, and support staff (including, but not limited to team managers and trainers).” Using operating expenses to predict on-court play makes sense in theory: the teams that eat better, travel better, sleep better and train better should, generally speaking, play better. It’s the same rationale that Mark Cuban, owner of the Dallas Mavericks, operated under when he purchased a private jet for his team’s travel.
See This Year’s Operating Expenses Bracket
And it works, too. Including the play-in round, the team with the higher operating expenses won in 70% of the games in last year’s tournament. If you used our system to fill out your bracket, you would have correctly picked 68.3% of the games from the round of 64 up to the Final Four. For comparison’s sake, taking the favored teams in those games would have yielded a 65% rate of correct picks. That’s not a large difference in total correct picks, but operating expenses offer a significant advantage if you want to capture those late-round points: team spending correctly predicted half of the Elite Eight and Final Four, whereas picking by the favorites predicted just three of the final eight teams and none in the Final Four.
At the very least, operating expenses are a great system for the first round. They wrongly predicted only six games in the round of 64, good for an 81% success rate. Of those six incorrect picks, two were 8/9 games (generally the most difficult to predict) and two were surprising upsets by Morehead State and Virginia Commonwealth (VCU).And it works, too. Including the play-in round, the team with the higher operating expenses won in 70% of the games in last year’s tournament. If you used our system to fill out your bracket, you would have correctly picked 68.3% of the games from the round of 64 up to the Final Four. For comparison’s sake, taking the favored teams in those games would have yielded a 65% rate of correct picks. That’s not a large difference in total correct picks, but operating expenses offer a significant advantage if you want to capture those late-round points: team spending correctly predicted half of the Elite Eight and Final Four, whereas picking by the favorites predicted just three of the final eight teams and none in the Final Four.
Higher operating expenses don’t always mean a higher seed, either. Our system picked nine upsets in the first round, and only three of them were 8/9 match-ups. Of the remaining six upsets, operating expenses correctly predicted four: Richmond (12) over Vanderbilt (5), Marquette (11) over Xavier (6), Gonzaga (11) over St. John’s (6) and Florida State (10) over Texas A&M (7). The accuracy of these predictions may illustrate how a well-funded team can out-perform its regular season record once the tournament starts.
Our approach of using operating expenses did not perform as well in the round of 32, but it still correctly picked nine out of the 16 games, including two surprising upsets: Marquette (11) over third-seeded Syracuse and Florida State (10) over second-seeded Notre Dame. More good news if your pool includes bonus points for picking upsets. Through the next two rounds, our system went an even 50% and correctly picked two of the teams in the Final Four: third-seeded Connecticut and fourth-seeded Kentucky. They were hardly Cinderella stories, but neither team was considered to be a favorite, either.
