Sagarin & Pomeroy data for the second round
If you’ve read my annual March Madness bracket column or if you are a stathead in general, you’ll know what the title of this post means. If you don’t, check out my column and come back.
Here are the next 16 matchups representing the second round of the NCAA Tournament. The Sagarin advantage represents the spread in Jeff Sagarin’s predictor ratings. If the number is positive, it means that Team A is the Sagarin favorite. If the number is negative, then it means Team B (and the lower seed) is actually the favorite.
The Pomeroy % represents the chances that Team A will win the game according to Ken Pomeroy’s Pythagorean calculation.
Over the last four years, teams with a 2+ point advantage in Sagarin’s “predictor” rating have won 156 of 198 games (78.7%). Over the last two years, if a team had at least a 65% expected win rate according to Pomeroy’s Pythagorean calculation, they won 61 of 76 games (80.2%).
I’ve also included the spread for the game at the World Sports Exchange. Oftentimes the spread is very close to the Sagarin spread.
| Team A | Team B | Sag Adv | Pom % | Spread |
|---|---|---|---|---|
| Kentucky | West Virginia | 2.6 | 63.0% | -3.5 |
| Florida | UCLA | 5.3 | 71.9% | -6 |
| Richmond | Morehead State | 6.2 | 72.0% | -4.5 |
| San Diego State | Temple | 5.7 | 75.0% | -5.5 |
| Pittsburgh | Butler | 8.8 | 82.4% | -7.5 |
| BYU | Gonzaga | 3.5 | 60.6% | +1.5 |
| Wisconsin | Kansas State | 5.5 | 73.7% | -3.5 |
| Connecticut | Cincinnati | 1.1 | 56.6% | -4 |
| North Carolina | Washington | 0.2 | 56.3% | -4 |
| Duke | Michigan | 9.8 | 83.5% | -11.5 |
| Ohio State | George Mason | 10.7 | 86.1% | -11 |
| Texas | Arizona | 5.4 | 75.9% | -5.5 |
| Purdue | VCU | 11.3 | 87.7% | -9 |
| Syracuse | Marquette | 3.6 | 67.7% | -4.5 |
| Kansas | Illinois | 5.4 | 69.9% | -8.5 |
| Notre Dame | Florida State | 4.4 | 71.5% | -5 |
[...] you’re looking for updated Sagarin and Pomeroy data, click here. Keep in mind that for record-keeping purposes, I use the static data from before round one to [...]