Volume 20, Issue 4

The existence and justification to the home advantage – the benefit a sports team receives when playing at home – has been studied across sport. The majority of research on this topic is limited to individual leagues in short time frames, which hinders extrapolation and a deeper understanding of possible causes. Using nearly two decades of data from the National Football League (NFL), the National Collegiate Athletic Association (NCAA), and high schools from across the United States, we provide a uniform approach to understanding the home advantage in American football. Our findings suggest home advantage is declining in the NFL and the highest levels of collegiate football, but not in amateur football. This increases the possibility that characteristics of the NCAA and NFL, such as travel improvements and instant replay, have helped level the playing field.

The baseball statistic “Wins Above Replacement” (WAR) has emerged as one of the most popular evaluation metrics. But it is not readily observed and tabulated; WAR is an estimate of a parameter in a vaguely defined model with all its attendant assumptions. Industry-standard models of WAR for starting pitchers from FanGraphs and Baseball Reference all assume that season-long averages are sufficient statistics for a pitcher’s performance. This provides an invalid mathematical foundation for many reasons, especially because WAR should not be linear with respect to any counting statistic. To repair this defect, as well as many others, we devise a new measure, Grid WAR, which accurately estimates a starting pitcher’s WAR on a per-game basis. The convexity of Grid WAR diminishes the impact of “blow-up” games and up-weights exceptional games, raising the valuation of pitchers like Sandy Koufax, Whitey Ford, and Catfish Hunter who exhibit fundamental game-by-game variance. Although Grid WAR is designed to accurately measure historical performance, it has predictive value insofar as a pitcher’s Grid WAR is better than Fangraphs’ FIP WAR at predicting future performance. Finally, at https://gridwar.xyz we host a Shiny app which displays the Grid WAR results of each MLB game since 1952, including career, season, and game level results, which updates automatically every morning.

We leverage Large Language Models (LLMs) to extract information from scouting report texts and improve predictions of National Hockey League (NHL) draft outcomes. In parallel, we derive statistical features based on a player’s on-ice performance leading up to the draft. These two datasets are then combined using ensemble machine learning models. We find that both on-ice statistics and scouting reports have predictive value, however combining them leads to the strongest results.

This paper attempts to identify football players who have a similar style to a player of interest. Playing style is not adequately quantified with traditional statistics, and therefore style statistics are created using tracking data. Tracking data allow us to monitor players throughout a match, and therefore include both “on-the-ball” and “off-the-ball” observations. Having developed style features, tractable discrepancy measures are introduced that are based on Kullback–Leibler divergence in the context of multivariate normal distributions. Examples are provided where a pool of players from the Chinese Super League are identified as having a playing style that is similar to players of interest.

Multi-competitor races often feature complicated within-race strategies that are difficult to capture when training data on race outcome level data. Models which do not account for race-level strategy may suffer from confounded inferences and predictions. We develop a generative model for multi-competitor races which explicitly models race-level effects like drafting and separates strategy from competitor ability. The model allows one to simulate full races from any real or created starting position opening new avenues for attributing value to within-race actions and performing counter-factual analyses. This methodology is sufficiently general to apply to any track based multi-competitor races where both tracking data is available and competitor movement is well described by simultaneous forward and lateral movements. We apply this methodology to one-mile horse races using frame-level tracking data provided by the New York Racing Association (NYRA) and the New York Thoroughbred Horsemen’s Association (NYTHA) for the Big Data Derby 2022 Kaggle Competition. We demonstrate how this model can yield new inferences, such as the estimation of horse-specific speed profiles and examples of posterior predictive counterfactual simulations to answer questions of interest such as starting lane impacts on race outcomes.

We perform an exploratory data analysis on a data-set for the top 16 professional darts players from the 2019 season. We use this data-set to fit player skill models which can then be used in dynamic zero-sum games (ZSGs) that model real-world matches between players. We propose an empirical Bayesian approach based on the Dirichlet-Multinomial (DM) model that overcomes limitations in the data. Specifically we introduce two DM-based skill models where the first model borrows strength from other darts players and the second model borrows strength from other regions of the dartboard. We find these DM-based models outperform simpler benchmark models with respect to Brier and Spherical scores, both of which are proper scoring rules. We also show in ZSGs settings that the difference between DM-based skill models and the simpler benchmark models is practically significant. Finally, we use our DM-based model to analyze specific situations that arose in real-world darts matches during the 2019 season.