Contributed Poster Presentations: Section on Statistics in Sports

Competitive balance is essential for sports leagues to maintain fan engagement and financial success. We investigate competitive balance across several professional leagues in soccer, basketball, football, and ice hockey using a metric based on the Bradley-Terry model. Men's soccer leagues in Europe and North America from 2004-present were analyzed, finding second divisions consistently more balanced than first. MLS proved more comparable to European second tiers in parity. Among major U.S. leagues since 2005, the NBA and NFL showed far lower balance than MLB, NHL and MLS. Incorporating playoff structures led to the NBA's lower balance being amplified while the NFL became more balanced. The metric also revealed higher parity in soccer versus basketball worldwide. Results suggest financial inequality, league format, playoff systems, and sports' inherent dynamics substantially impact balance. While limited by its narrow time frame and focus on standings over scheduling, the analysis provides valuable comparative insights and contributes towards the goal of the optimal viewing experience for fans. 

Athletes who compete in the Olympic Games participate in many other competitions leading up to the games event, and scores in these competitions are possible tools to use to predict Olympic performance. However, at each competition the name of a gymnast is not always perfectly recorded, as nicknames and other variants of names are often used. In this project, we use record linkage to identify gymnasts in data from 2022 and 2023 international competitions. We propose an adaptation to the Jaro-Winkler Similarity score based on the specific discrepancies in names in this dataset and we then use Bayesian Hierarchical Modeling on the linked data to predict winning gymnasts in Women's Artistic Gymnastics.