Probability models for the NCAA Regional Basketball tournaments


Authors: 
Schwertman, N. C., McCready, T. A., & Howard, L.
Category: 
Volume: 
45(1)
Pages: 
35-38
Year: 
1991
Publisher: 
The American Statistician
Abstract: 

The authors state that athletics in general can be used to demonstrate probabilistic concepts and that these may be reinforced due to the motivational aspects of the examples. In this article, the planification of the NCAA Regional Basketball tournaments is examined as a probability problem (e.g. what is the probability of team #3 winning the regionals?). In calculating the probability of any team winning, the students must: 1) evaluate the probability of each game played; 2) assume independence of each contest; and 3) incorporate the relative strength of each team in the model. Three probability models are considered, the third being the most adequate. It seems to provide a good fit to the data and is considered interesting to students without requiring a strong mathematical background. The key point in this article is that these models demonstrate the multiplication principle and the additive property of mutually exclusive events in addition to the motivation provided by the topic itself.

The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education