Course Syllabus
Yale School of Management
MGT 817 – Sports Analytics
Fall 2017
Professors: Ed Kaplan (Evans Hall, Room 3550) and Toby Moskowitz (Evans Hall, Room 4508)
TA: Scott Rodilitz; RA: Kaushik Vasudevan
Class Meetings: Tuesdays in Evans Hall Room 4200 from 4:10 -7:00
First meeting: Tuesday, September 5
Detailed Schedule of Classes: see below.
Software: Microsoft Excel and another software package that can handle large data (R, Matlab, Python, Stata, SAS, . . . ).
Texts: Tobias Moskowitz and L. Jon Wertheim. Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won. New York: Crown Publishing, 2012 (paperback edition): ISBN 978-0-307-59180-7/
Wayne L. Winston. Mathletics: How gamblers, managers, and sports enthusiasts use mathematics in baseball, basketball, and football. Princeton, NJ: Princeton University Press, 2012 (paperback edition): ISBN 978-0-691-15458-9.
Additional readings will be posted on the Sports Analytics CANVAS website.
Course description: We will not have all the answers to many of the questions we address in the course. Our goal is teach students how to think about the issues, rather than what to think. The principles we aim to impart on students, taught through the lens of sports, will be useful in addressing many business-related issues, even outside of sports.
This course is designed to help students develop and apply analytical skills that are useful in a general business environment, applied specifically to sports. Students will learn how to apply methods and principles in a wide range of applications: evaluating performance and decision making, hypotheses testing, interpreting market-based evidence, identifying directions of causation, and quantifying the magnitude of various effects. Although the course focuses on applications in the sports context and uses approaches that are rapidly becoming important in the business of sports analytics, this is not a survey course about issues in sports.
Probability modeling and statistics are fundamental tools of management. Building upon earlier core courses in probability, statistics, economics, and modeling managerial decisions, Sports Analytics is designed to provide additional experience using these modeling tools via applications to problems in sports. Specific analytical tools include conditional probability, conditional expectation, random variables and probability distributions, hypothesis testing, and regression (including identification strategies such as instrumental variables, as well as logistic regression). We will consider various applications to major sports such as baseball, basketball, football, hockey and soccer. Sample applications include in-game decision making, rating sports teams and individual players, modeling win probability and win probability added in various sports, understanding the determinants of home field advantage, understanding referee behavior, sports gambling, and player evaluation. Students will gain practice applying analytical tools to these topics via regular problem sets and a term project.
Homework, Project, and Grading: Students must attend all class sessions and complete assigned readings before class meetings.
The group project accounts for 60% of the grade and the homework assignments combine for the remaining 40% of the grade.
Homework/Problem sets. There will be regular group problem sets posted on the Sports Analytics Canvas website. These assignments must be submitted on time with no exceptions. All group members are expected to contribute to the completion of group assignments. Signing a submitted assignment indicates contribution to the work; if you do not contribute to a group homework assignment, do not sign the submission. At the end of the course, each group member will turn in a signed sheet indicating their own % contribution to the group assignments as well as their assessment of each group member’s contribution to the assignments throughout the course. While it is permissible to informally consult other class members about aspects of assignments, it is not acceptable to collaborate across groups. All groups must submit original work.
Project. The class project will be done in the same groups and will require a write up and analysis of an issue in sports analytics. A list of possible topics will be provided, but you are also welcome to come up with your own topic, but must seek approval from the instructors. Research projects typically involve analyzing a sports-related issue using data analysis and a probability or empirical model. At the end of the course groups will present their projects to the whole class and turn in their written projects. The write-up should be in the form of a paper as if submitting to an academic journal. Details on the specifics of the project (paper write-up, presentations, and possible topics) will be provided in a separate document on the first day of class. We will also help groups obtain and identify data sources for their projects.
At the end of the course, each group member will also turn in a signed sheet indicating their own % contribution to the project as well as their assessment of each group member’s contribution to the project. A student will not receive a grade in the course without turning in this contribution document.
The timeline for group projects is:
Week 2 (Sept. 12) – Groups formed and submitted (via email).
Week 4 (Sept. 26) – Topics chosen and a student prepared preliminary outline of what the project will entail is turned in.
Week 7 (Oct. 24) – Groups turn in preliminary analysis associated with the project (at this point data should be obtained and perhaps some early tables, figures, etc. have been produced).
Week 11 (Nov. 28) – Groups turn in first draft of their paper for comments.
Week 13 (Dec. 12) –Groups turn in final draft of written project along with each student’s evaluation of all group members including themselves. Presentations also conducted.
Schedule of Classes
Date |
Topic |
Prof. in charge: |
Sept 5 |
From scoring to winning |
Kaplan |
Sept 12 |
Baseball: state space, run value added, win probability |
Kaplan |
Sept 19 |
State space models: hockey, soccer, basketball, football |
Kaplan |
Sept 26 |
Strategic in-game decision-making |
Moskowitz |
Oct 3 |
Home field advantage |
Moskowitz |
Oct 10 |
Guest: TBD |
|
Oct 24 |
Referee analytics |
Moskowitz |
Oct 31 |
Rating sports teams |
Kaplan |
Nov 7 |
The market for and evaluation of talent |
Moskowitz |
Nov 14 |
Office pools |
Kaplan |
Nov 28 |
Sports gambling and the Kelley criterion |
Moskowitz/Kaplan |
Dec 5 |
Optical tracking (BIG) data |
Moskowitz |
Dec 12 |
Project presentations |
|
Course Summary:
Date | Details | Due |
---|---|---|