Back to Prediction Models with Sports Data
University of Michigan

Prediction Models with Sports Data

In this course the learner will be shown how to generate forecasts of game results in professional sports using Python. The main emphasis of the course is on teaching the method of logistic regression as a way of modeling game results, using data on team expenditures. The learner is taken through the process of modeling past results, and then using the model to forecast the outcome games not yet played. The course will show the learner how to evaluate the reliability of a model using data on betting odds. The analysis is applied first to the English Premier League, then the NBA and NHL. The course also provides an overview of the relationship between data analytics and gambling, its history and the social issues that arise in relation to sports betting, including the personal risks.

Status: Data Analysis
Status: Regression Analysis
IntermediateCourse33 hours

Featured reviews

WV

5.0Reviewed Apr 11, 2024

Very interesting course, even though some of the data prep is kind of weird it's nice to see things done a bit differently

BB

4.0Reviewed Jul 10, 2023

I found the material from weeks 2 and 4 very interesting!

All reviews

Showing: 7 of 7

VALAY SHAH
5.0
Reviewed Jul 18, 2022
Надежда Вахрушева
5.0
Reviewed Feb 15, 2022
Brendan Bebb
4.0
Reviewed Jul 11, 2023
Péter
3.0
Reviewed Aug 6, 2025
JOSE EDUARDO LARIS ECHEVERRIA
1.0
Reviewed Jun 12, 2023
William VL
5.0
Reviewed Apr 12, 2024
Moulay Ali Elmghari Tabib
5.0
Reviewed Jul 31, 2023