THE MARCH MADNESS MODEL UPDATE
Updated Projections After Round 1
As always, the first round did NOT disappoint. We had an upset nobody expected in McNeese St. over Clemson, Michigan barely survived a scare against UC San Diego, and thrilling finished in St. Mary’s win over Vandy and Arkansas’ defeat of the Kansas Jayhawks. Our model was pretty successful in round one (including the first “First Four”, here are some diagnostics:
Accuracy: 83.3%
Mean absolute error of predicted margin of victory: 10.5
Comparing to KenPom (stats below), we did just as well if not better -
Accuracy: 80.6%
Mean absolute error of predicted margin of victory: 9.6
Additionally, we correctly predicted Drake to beat Missouri and Creighton to beat Louisville, both of which KenPom and Bart Torvik missed, and the betting markets incorrectly picked.
After updating the stats to account for round 1 games, and updating to include matchups that we did not predict, here is our updated tourney forecast. To check back on our initial projections, or review the methodology, you can refer to my pre-tourney article here:
Second Round
Sweet Sixteen
Elite Eight
Final Four
BONUS: Sleepers to Watch
A few weeks back, I wrote about teams to potentially watch out for as "big upset” candidates based on similarity scores to previous successful, winning underdogs. This also included underdogs to win multiple games in the tournament, with at least one of those being a 5+ seed-line upset. Using the findings from that analysis, here are the teams most equipped to win multiple games in a tournament, with one of those wins being a big (5+ seed-line) upset. Included in the table below are how each team ranks among all 2025 teams across college basketball, not just tournament teams, in terms of “big upset” and multiple tourney win team resemblance.
If you like these predictions, I encourage you to also check out the forecasts by
at Neil’s Substack and at Silver BulletinBring on the madness!