Volume 14, Issue 4

The availability of pitch-tracking data has led to increased scrutiny of Major League Baseball umpires. While many studies have attempted to rate umpires based on their conformity to the rule book strike zone, players and managers tend to accept deviations from this zone, provided that umpires establish consistent zones within a game. Using tools from computational geometry, we propose new metrics for assessing the consistency and accuracy of an umpire’s ball and strike calls over the course of a game. We apply these metrics to pitch-tracking data on all ball and strike calls made during the 2017 MLB regular season, giving some characterizations of the variation in performance of MLB umpires. This analysis demonstrates that measures of consistency can complement current accuracy-based evaluations of umpires.

Each year, millions of people fill out a bracket to predict the outcome of the popular NCAA men’s college basketball tournament, known as March Madness . In this paper we present a new methodology for team ranking and use it to predict the NCAA basketball tournament. We evaluate our model in Kaggle’s March Machine Learning Mania competition, in which contestants were required to predict the results of all possible games in the tournament. Our model combines two methods: the least squares pairwise comparison model and isotonic regression. We use existing team rankings (such as seeds, Sagarin and Pomeroy ratings) and look for a monotonic, non-linear relationship between the ranks’ differences and the probability to win a game. We use the isotonic property to get new rankings that are consistent with both the observed outcomes of past tournaments and previous knowledge about the order of the teams. In the 2016 and 2017 competitions, submissions based on our methodology consistently placed in the top 5% out of over 800 other submissions. Using simulations, we show that the suggested model is usually better than commonly used linear and logistic models that use the same variables.

We address the fiercely debated question of whether the strongest European football clubs get special, preferential treatment from match officials in their decisions on the teams’ players over the course of the teams’ trophy winning streaks. To give an empirical answer to this question, we apply a rigorous econometric analysis for causal effect estimation to a self-constructed data set. We consider the two clubs in the Italian Serie A that experienced a prolonged winning streak during the period 2006–2016, namely Internazionale Milan (Inter) and Juventus Turin, as well as one team from the German Bundesliga (Borussia Dortmund) and one from the English Premier League (Manchester United) that also experienced a winning streak during the same period. This allows us to perform an analysis with enough statistical power to be able to estimate properly the effect of interest. The general opinion among fans, sports journalists, and insiders that the strongest clubs are favored by match officials’ decisions is supported only by the results of the analysis we run for Juventus, whereas for the other clubs under investigation, we did not find any significant bias. During its winning streak, more yellow cards and total booking points (an aggregated measure of yellow and red cards) were given to Juventus opponents. These effects are not only statistically significant, but also have a sizeable impact.

The National Football League (NFL) Scouting Combine takes place annually for the purpose of allowing NFL teams to evaluate prospects. The battery of six physical tests receives a great deal of attention, and are a focus of team personnel as well as fans of NFL teams. Recently, some have suggested that the current battery of tests should be modified. This work aims to characterize the multivariate dependence structure between tests for Combine prospects, for both typical and elite-level performers, for the purpose of better understanding the current battery of tests before making modifications. Through analysis of two pairwise dependence matrices, one quantifying dependence in the center of the distribution and the other quantifying dependence in the tails of the distribution, this analysis finds that several events show differing levels of association, and that fewer Combine events may be sufficient going forward.

We create a data-driven Markov model for generating 10-pin bowling scores from the Professional Bowlers Association. The model incorporates insights from the hot hand literature and makes use of Bayesian shrinkage. For realistic sample sizes, the proposed approach is superior to modeling via the empirical distribution. Investigation of player-specific model components allows for a richer comparison of players than is possible using raw game scores alone. An additional feature of the model is that it can be used for in-game prediction.