We are excited to direct some well-deserved attention to a recently published book that includes significant contributions from Dr. Robert MacDonald, one of our associate economists here at Cirque Analytics. The book, authored by UC Irvine professor and economist Michael McBride, is titled Game Theory, Machine Learning, and Production in Sports: The Fair-Credit Baseball Statistics, and was released by World Scientific Publishing.
Dr. McBride’s book brings together ideas from economics, game theory, and machine learning to explore a central challenge in sports analytics: how to accurately measure the contributions of individual players within a team setting.
“Fair-Credit Baseball” Statistics
The book introduces the concept of Fair-Credit Baseball (FCB) statistics, a framework developed by McBride that relies on the Nobel-Prize-winning concept of a Shapley Value to better allocate individual credit for outcomes. This is especially important for baseball, as traditional statistics often attribute results to a single player, even when those results often depend on the actions of several players working together.
The FCB framework approaches baseball more like an economic production process, where outcomes are generated by multiple inputs. By applying tools and theory from mathematics, economics, and data science, the framework aims to more accurately measure how players collectively produce results on the field.
Robert MacDonald’s Contribution
Dr. MacDonald was a co-author on Chapters 6 and 7, which are the most technically demanding portions of the book. These chapters address the “training” of a machine learning model to efficiently and accurately record the thousands of in-game events that collectively form the basis for all baseball statistics. They also focus on evaluating the predictive performance of statistical models used to measure player contributions, examining how these models can be used to assess performance and test how well those models perform when applied to new data.
Dr. McBride had high praise for Dr. MacDonald’s contributions to the book, saying that “Robert is really the lead author for Chapters 6 and 7, and they could not have been written without him. He deserves the bulk of the credit for conducting the actual machine-learning analysis–from the designing of the models to the final testing of their performance. I learned a lot working with him.”
Using Economic Theory & Data Analytics to Answer Important Questions
Dr. MacDonald’s work on this project is a strong example of the approach we champion at Cirque Analytics: combining creativity, rigorous economic thinking, and modern analytical tools to solve complex, real-world problems. The questions he tackled in these chapters, such as how individuals contribute to team outcomes, how productivity can be measured across different environments, and how data-driven models can improve decision-making, are the same challenges that economists and data scientists face across a wide range of settings, from individual firms and consumers to complex markets and public policy.
While our clients rarely ask us to settle debates over MVP candidates in the most recent World Series, the skills and methodologies Dr. MacDonald applied in this work transfer directly to the work we do every day. We are proud of Robert’s contributions and look forward to the many ways his expertise will continue to create value, both for our clients and for the broader field of economic analysis.