How well all six models matched the 2014-15 final coaches' poll.

Table explanation: The Spearman Correlation Coefficients (SCC) below relate to how well the top 15, 25 and 35 teams, in the final coaches' poll, match with the predictions for the models listed. The Score column is a metric (whose maximum value is 2000) that gives more weight to the predictions near the top of the poll than to those near the bottom. The '0,1,2' column contains how many top 35 team rankings were only off by 0, 1, or 2 places, and this is followed by the average deviation between the actual ranking and the prediction. ("X" in the large table at the bottom represents the number of wins in the NIT tournament that a team earned.)

SCC15 SCC25 SCC35 0,1,2 Average
ZP2 0.97857 0.97462 0.95187 11,9,6 1.9143
PR2 0.96250 0.96962 0.94809 12,3,11 2.1143
50T 0.96250 0.96346 0.95658 12,2,12 1.9810
LN2 0.98036 0.97885 0.96008 16,5,8 1.6476
 ZPF 0.96071 0.97692 0.96377 9,13,6 1.7048
MCB 0.93214 0.92269 0.90700 3,17,2 2.7048
OCC 0.97143 0.95084 0.85761 10,9,4 3.2190

Commentary: All 7 models listed above achieved better SCC values, and average differences, than their average values for every one of these quantities since 1993! LN2 had the highest SCC-15 value above, which also tied its previous best SCC-15 value (in 2008-09), as well as the lowest average difference of the models (and it was the fifth lowest such total for LN2; the SCC-25 value above is the second lowest ever - for LN2). LN2 also predicted 29 of the top 35 teams within 2 places of their actual, final ranking. The 2014-15 tournament yielded the fifth best predictive match by the MCB model (using the avg. diff.), and the 17 teams, whose position was only off by one, is four more than the previous best recorded such value (in 1993-94) by this model. Not to be overshadowed, this year produced the third lowest average difference for the ZPF model, and its fourth lowest SCC-15 and SCC-25 values - not to mention 28 of the top 35 teams being only 2 or fewer positions away from their final place in the final coaches' poll (as predicted by the ZPF model). The ZP2 SCC-15 value listed above is higher than the one for ZPF, but this only tied for the fourth best such value (for ZP2). (OCC posted its second highest SCC-15 value this year as well.)

TeamName Poll Wins ZP2 PR2 50T LN2 ZPF MCB OCC
Duke 1 6 1 1 1 1 1 1 1
Wisconsin 2 5 2 2 2 2 2 3 2
Kentucky 3 4 3 3 3 3 3 2 3
Arizona 4 3 4 4 4 4 4 4 4
NotreDame 5 3 6 7 7 6 6 6 6
Gonzaga 6 3 5 6 6 5 5 5 5
MichiganSt 7 4 7 5 5 7 8 7 7
Virginia 8 1 10 10 10 10 10 10 11
Villanova 9 1 9 9 9 9 7 8 8
Louisville 10 3 8 8 8 8 9 9 9
WichitaSt 11 2 11 11 11 11 11 12 10
NorthCarolina 12 2 12 12 12 12 13 13 12
Oklahoma 13 2 13 13 13 13 14 14 13
NorthernIowa 14 1 15 16 16 14 12 17 15
Utah 15 2 16 14 14 16 17 11 14
Maryland 16 1 17 18 18 17 16 20 18
Kansas 17 1 14 15 15 15 15 16 17
WestVirginia 18 2 18 17 17 18 19 15 16
IowaSt 19 0 19 19 19 19 18 24 21
Arkansas 20 1 22 23 23 23 20 21 20
Baylor 21 0 24 24 24 24 24 29 30
Xavier 22 2 20 20 20 20 21 19 19
Butler 23 1 26 26 26 25 26 22 22
NCState 24 2 21 21 21 21 22 23 27
Oregon 25 1 28 28 30 27 28 31 24
SMU 26 0 29 31 28 28 25 33 32
Georgetown 27 1 25 25 25 26 27 25 23
UCLA 28 2 23 22 22 22 23 27 28
VACommonwealth 29 0 38 38 37 37 37 28 35
SanDiegoSt 30 1 31 30 32 30 30 18 26
OhioSt 31 1 30 29 31 31 32 30 29
Dayton 32 1 33 34 34 32 33 26 31
Iowa 33 1 32 33 33 33 34 32 42
Providence 34 1 27 27 29 29 29 38 25
Purdue 35T 0 49 48 47 46 46 49 48
LSU 35T 0 48 47 46 50 48 47 77
Cincinnati 35T 1 34 35 35 34 35 34 41
UAB 38T 0 87 84 68 67 65 45 78
MurraySt 38T "2" 54 57 64 73 73 100 33