2 Kansas upset 1 North Carolina. Image credit: Wikipedia |
First things first: Jeff Sagarin alerted me to the fact that I made an error in constructing the bracket based on his system. I had him picking UNC to win their region. He actually picked Kansas. The proof is in the post. An error such as this would, obviously, have hurt his system after yesterday's games. In the rankings below I have rectified this mistake. Apologies to my readers and Mr. Sagarin.
Second things second: 1 Kentucky did just what every system expected, while 2 Kansas did what few systems expected, that being advance to the Final Four.
In fact, only five systems put Kansas in the Final Four, while fourteen picked Carolina. Moreover, five had UNC them in the final game and three had them winning it all.
Rank | System | R64 | R32 | S16 | E8 | FF | NCG | Champion | PPR | Total | Pct | Correct |
1 | Survival Model | 250 | 200 | 280 | 240 | 0 | 0 | Kentucky | 640 | 970 | 99.8 | 75.0% |
2 | ESPN/USA Today Preseason | 230 | 200 | 280 | 240 | 0 | 0 | 160 | 950 | 99.6 | 71.7% | |
3 | Jeff Sagarin | 230 | 200 | 200 | 240 | 0 | 0 | Kentucky | 640 | 870 | 96.9 | 68.3% |
4 | AP Preseason | 210 | 220 | 280 | 160 | 0 | 0 | 160 | 870 | 96.9 | 68.3% | |
5 | Pomeroy | 220 | 180 | 200 | 240 | 0 | 0 | Kentucky | 640 | 840 | 94.0 | 65.0% |
6 | LRMC Bayesian | 200 | 180 | 200 | 240 | 0 | 0 | Kentucky | 640 | 820 | 91.4 | 61.7% |
7 | ESPN Simulation Model | 220 | 180 | 160 | 240 | 0 | 0 | Kentucky | 640 | 800 | 88.1 | 63.3% |
8 | ESPN Decision Tree | 210 | 220 | 200 | 160 | 0 | 0 | Kentucky | 640 | 790 | 86.0 | 65.0% |
9 | Sonny Moore | 230 | 200 | 200 | 160 | 0 | 0 | Ohio St | 480 | 790 | 86.0 | 66.7% |
10 | ESPN National Bracket | 220 | 200 | 200 | 160 | 0 | 0 | Kentucky | 480 | 780 | 83.8 | 65.0% |
11 | FiveThirtyEight | 220 | 180 | 200 | 160 | 0 | 0 | Kentucky | 640 | 760 | 78.8 | 63.3% |
12 | Pure Chalk | 220 | 220 | 200 | 80 | 0 | 0 | Kentucky | 480 | 720 | 67.3 | 65.0% |
12 | AP Postseason | 220 | 220 | 200 | 80 | 0 | 0 | Kentucky | 480 | 720 | 67.3 | 65.0% |
14 | Lunardi RPI | 200 | 200 | 240 | 80 | 0 | 0 | 160 | 720 | 67.3 | 61.7% | |
15 | Vegas | 210 | 220 | 200 | 80 | 0 | 0 | Kentucky | 480 | 710 | 63.2 | 63.3% |
16 | ESPN CBPI | 220 | 200 | 200 | 80 | 0 | 0 | Kentucky | 640 | 700 | 59.9 | 63.3% |
17 | ESPN/USA Today Postseason | 220 | 200 | 200 | 80 | 0 | 0 | Kentucky | 480 | 700 | 59.9 | 63.3% |
18 | ESPN InsideRPI | 200 | 180 | 200 | 80 | 0 | 0 | 160 | 660 | 46.9 | 58.3% | |
19 | Nolan Power Index | 190 | 160 | 200 | 80 | 0 | 0 | Kentucky | 480 | 630 | 37.5 | 55.0% |
*Rank sorted by total points (Total), possible points remaining (PPR) and percentage of correct picks (Correct). "Pct" denotes where the brackets rank among the rest in the ESPN Tournament Challenge.
Finally, yesterday's outcomes significantly boosted the standing of the computerized systems published by Ken Pomeroy and the aforementioned Jeff Sagarin (whose system now occupies third place).
Rank | System | South | East | West | Midwest | Correct | PPR |
1 | Survival Model | UK | OSU | KU | 75.0% | 640 | |
2 | ESPN/USA Today Preseason | UK | OSU | L'Ville | 71.7% | 160 | |
3 | Jeff Sagarin | UK | OSU | 68.3% | 640 | ||
4 | AP Preseason | UK | KU | 68.3% | 160 | ||
5 | Pomeroy | UK | OSU | KU | 65.0% | 640 | |
6 | LRMC Bayesian | UK | OSU | KU | 61.7% | 640 | |
7 | ESPN Simulation Model | UK | OSU | KU | 63.3% | 640 | |
8 | ESPN Decision Tree | UK | OSU | 65.0% | 640 | ||
9 | Sonny Moore | UK | OSU | 66.7% | 480 | ||
10 | ESPN National Bracket | UK | OSU | 65.0% | 480 | ||
11 | FiveThirtyEight | UK | OSU | 63.3% | 640 | ||
12 | Pure Chalk | UK | 65.0% | 480 | |||
12 | AP Postseason | UK | OSU | 65.0% | 480 | ||
14 | Lunardi RPI | UK | 61.7% | 160 | |||
15 | Vegas | UK | 63.3% | 480 | |||
16 | ESPN CBPI | UK | 63.3% | 640 | |||
17 | ESPN/USA Today Postseason | UK | 63.3% | 480 | |||
18 | ESPN InsideRPI | UK | 58.3% | 160 | |||
19 | Nolan Power Index | UK | 55.0% | 480 |
No comments:
Post a Comment