Here are ISU's final rankings:
Seed List: 24
NET: 21
KPI: 18
SOR: 25
BPI: 15
Sagarin: 15
KenPom: 16
The Committee slotting ISU at 24 was kind of a headscratcher, ending up on the 6 line perhaps was not. Two teams slotted above ISU, Marquette and Maryland, are the most annoying given their numbers. The others slotted are not so much so I don't think the 6 seed was a hose job.
Here are correlation coefficients (and of number of at large misses) between the Seed List and the Committee computer models:
NET: 0.9701 (6 teams missed)
KPI: 0.9689 (3 teams missed)
ESPN SOR: 0.9785 (5 teams missed)
ESPN BPI: 0.9634 (7 teams missed)
SAG: 0.9677 (7 teams missed)
KenPom: 0.9682 (6 teams missed)
NET had the 2nd highest correlation to the final Seed List but missed on 6 teams so while the Committee primarily uses the NET as a sorting tools for the Quads, they obviously did not heavily use NET rankings to pick the at large teams. This is the 2nd year in a row that KPI had the fewest misses. ISU's worst ranking was 25 in the ESPN SOR and the SOR has the overall best seeding correlation.
Joe Lunardi missed on one at large team picking TCU instead of Belmont. Jerry Palm and Bracket Matrix did likewise. I thought NC State would get in instead of St John's.
Seed List: 24
NET: 21
KPI: 18
SOR: 25
BPI: 15
Sagarin: 15
KenPom: 16
The Committee slotting ISU at 24 was kind of a headscratcher, ending up on the 6 line perhaps was not. Two teams slotted above ISU, Marquette and Maryland, are the most annoying given their numbers. The others slotted are not so much so I don't think the 6 seed was a hose job.
Here are correlation coefficients (and of number of at large misses) between the Seed List and the Committee computer models:
NET: 0.9701 (6 teams missed)
KPI: 0.9689 (3 teams missed)
ESPN SOR: 0.9785 (5 teams missed)
ESPN BPI: 0.9634 (7 teams missed)
SAG: 0.9677 (7 teams missed)
KenPom: 0.9682 (6 teams missed)
NET had the 2nd highest correlation to the final Seed List but missed on 6 teams so while the Committee primarily uses the NET as a sorting tools for the Quads, they obviously did not heavily use NET rankings to pick the at large teams. This is the 2nd year in a row that KPI had the fewest misses. ISU's worst ranking was 25 in the ESPN SOR and the SOR has the overall best seeding correlation.
Joe Lunardi missed on one at large team picking TCU instead of Belmont. Jerry Palm and Bracket Matrix did likewise. I thought NC State would get in instead of St John's.
