Our model’s probabilities are live for the NIT. They aren’t anything too special, but we’re unaware of other NIT probabilities out there, so perhaps they may be of use. Like most of our models, they work by simulating the remaining of each tournament 10,000 times. Here is how the model works, in full detail.
The probabilities are almost entirely kenpom. They start with the kenpom ratings. They then adjust those ratings in each simulation as the simulated tournament goes on, accounting for completed results. For example: Last year, when Fairleigh Dickinson beat Purdue, they naturally had a much better kenpom rating entering the second round than they did before the upset. In any scenario in which FDU reached the second round, they would be known to be better than they were thought to be entering the tournament. Our model accounts for this. However.
Our model does not run the simulations fully “hot.” It doesn’t adjust as much to each result as kenpom itself does. Instead, it runs them “lukewarm,” because present kenpom ratings tend to have stronger predictive power of future kenpom ratings than we get when we let the results go full bore. Also: We don’t account for other teams’ results the way kenpom does. We adjust it as an elo system, using his ratings as a starting point. This doesn’t create too significant of differences, but we are not fully reverse engineering kenpom over here. No college basketball genome project for us.
This is best viewed on a computer, but if viewing on an iPhone, sometimes double-tapping with two fingers will expand the screen to fit the full table.
Last updated: Tuesday 4/2
| Team | Second Round | Quarterfinals | Semifinals | Championship | Champion |
|---|---|---|---|---|---|
| Indiana State | 100.0% | 100.0% | 100.0% | 100.0% | 56.0% |
| Seton Hall | 100.0% | 100.0% | 100.0% | 100.0% | 44.0% |
| Utah | 100.0% | 100.0% | 100.0% | 0.0% | 0.0% |
| Georgia | 100.0% | 100.0% | 100.0% | 0.0% | 0.0% |
| UNLV | 100.0% | 100.0% | 0.0% | 0.0% | 0.0% |
| VCU | 100.0% | 100.0% | 0.0% | 0.0% | 0.0% |
| Ohio State | 100.0% | 100.0% | 0.0% | 0.0% | 0.0% |
| Cincinnati | 100.0% | 100.0% | 0.0% | 0.0% | 0.0% |
| Wake Forest | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Iowa | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Boston College | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| South Florida | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Minnesota | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Virginia Tech | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Bradley | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| North Texas | 100.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Villanova | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Princeton | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| SMU | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| San Francisco | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Appalachian State | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Saint Joseph’s | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Loyola (IL) | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Providence | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Butler | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| UCF | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Xavier | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| LSU | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Kansas State | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Richmond | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| UC Irvine | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| Cornell | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |