Below is my full Survival Model rankings for the NCAA tournament in 2012. The numbers are tough to interpret, but they represent the relative risk of failing (losing) at a given time period. Lower numbers represent better chances to advance:
| Team | Seed | Relative Risk |
| Kentucky |
1 |
1.1634 |
| Michigan St |
1 |
1.2112 |
| Ohio St |
2 |
1.2226 |
| Syracuse |
1 |
1.2685 |
| Kansas |
2 |
1.2788 |
| Missouri |
2 |
1.2793 |
| North Carolina |
1 |
1.3119 |
| Memphis |
8 |
1.3284 |
| Wichita St. |
5 |
1.3454 |
| Indiana |
4 |
1.4021 |
| Vanderbilt |
5 |
1.4192 |
| Baylor |
3 |
1.4405 |
| New Mexico |
5 |
1.4519 |
| Wisconsin |
4 |
1.4555 |
| Florida |
7 |
1.4901 |
| St. Louis |
9 |
1.4903 |
| Georgetown |
3 |
1.4920 |
| Duke |
2 |
1.4995 |
| Murray St. |
6 |
1.4997 |
| Gonzaga |
7 |
1.5098 |
| Marquette |
3 |
1.5195 |
| VCU |
12 |
1.5287 |
| Louisville |
4 |
1.5398 |
| Kansas St. |
8 |
1.5741 |
| Florida St. |
3 |
1.5778 |
| Belmont |
14 |
1.5798 |
| Michigan |
4 |
1.5904 |
| Temple |
5 |
1.6065 |
| Iowa St. |
8 |
1.6275 |
| Connecticut |
9 |
1.6529 |
| Cincinnati |
6 |
1.6858 |
| Harvard |
12 |
1.7268 |
| Nevada Las Vegas |
6 |
1.7418 |
| San Diego St. |
6 |
1.7695 |
| Purdue |
10 |
1.8008 |
| Alabama |
9 |
1.8058 |
| Creighton |
8 |
1.8175 |
| Long Beach St. |
12 |
1.8276 |
| Virginia |
10 |
1.8439 |
| Texas |
11 |
1.8573 |
| St. Mary’s |
7 |
1.8901 |
| Brigham Young |
14 |
1.9048 |
| Iona |
14 |
1.9903 |
| Notre Dame |
7 |
2.0195 |
| California |
12 |
2.0266 |
| New Mexico St. |
13 |
2.0462 |
| Lehigh |
15 |
2.0652 |
| West Virginia |
10 |
2.0705 |
| Xavier |
10 |
2.0722 |
| Southern Mississippi |
9 |
2.0873 |
| Ohio |
13 |
2.1188 |
| Davidson |
13 |
2.1709 |
| North Carolina St. |
11 |
2.1808 |
| Montana |
13 |
2.2171 |
| South Dakota St. |
14 |
2.2462 |
| St. Bonaventure |
14 |
2.4850 |
| NC Asheville |
16 |
2.6297 |
| Colorado St. |
11 |
2.6780 |
| Vermont |
16 |
2.8825 |
| South Florida |
12 |
2.9453 |
| Lamar |
16 |
3.0594 |
| Long Island |
16 |
3.2212 |
| Detroit |
15 |
3.2273 |
| Colorado |
11 |
3.3386 |
| Loyola MD |
15 |
3.4423 |
| Norfolk St. |
15 |
5.6642 |
| Western Kentucky |
16 |
7.4901 |
| Mississippi Valley St. |
16 |
23.1318 |
Don’t care what the analysis says . If you want to bet me the 6 teams ahead of North Carolina in the field other than Kentucky will get further than NC i will take that bet all day . 🙂
Why do you pick Florida over Missouri despite their weaker survival ranking?
Hey Alex,
The Survival Analysis model is good, but it is not perfect. There are often more early upsets than it anticipates because it is calibrated to predict long-term survival. Mizzou has the largest variance of prediction, and high seeds with that profile tend to not do as well as the model predicts. Examples would be Notre Dame last year, Tennessee that year they were a two seed, Gonzaga as a 3 in the Adam Morrison year, etc.
Luke and I wanted to spice up the bracket a bit, and picking against the Tigers when they play a pretty good Florida team seemed like it could accomplish that while staying true to the model. Feel free to disagree and advance the Tigers past Florida.
Well your picks look outstanding so far, if you would have combined them with your most confident turnover/rebounding upset picks your bracket would look like one of the best in the nation.
Someone’s going to have to help me the math, rankings and bracket selection regarding Cincinnati making it to the Sweet 16 over a higher rated FSU. While I agree that might happen your completed bracket doesn’t match your ratings. The same could be said NC State/SD State and for ‘Bama/Creighton. There’s probably others I missed. In other words why does your completed bracket differ from your risk ratings?
How is the survival index calculated?