Mark E. Glickman, JQAS Editor-in-Chief
The December 2015 issue (volume 11, issue 4) of the Journal of Quantitative Analysis in Sports (JQAS) consists of four articles with applications to hockey, American football, ski-jumping, and ultrarunning.
“A Finite Mixture Latent Trajectory Model for Modeling Ultrarunners’ Behavior in a 24-Hour Race” by Francesco Bartolucci and Thomas Brendan Murphy is the editor’s choice article for this issue and available for free download for the next 12 months. The article develops a model for ultrarunners’ performance that accounts for different strategies and behaviors for modulating speed and rest propensity throughout a race. The model assumes speed and rest trajectories can be described via latent clusters and that cluster membership may depend on runner covariates. The model is demonstrated on ultrarunner outcomes from the International Association of Ultrarunners 2013 World Championship.
“Riding a Probabilistic Support Vector Machine to the Stanley Cup” by Simon Demers compares the predictive performance of various team metrics in the context of NHL playoffs. Metrics were compared in the context of 105 best-of-seven NHL playoff series that took place between 2008 and 2014 using a relevance vector machine learning approach and the more common support vector machine (SVM). The probabilistic SVM results were used to derive playoff performance expectations for NHL teams and identify playoff under-achievers and over-achievers.
In “Consistency, Accuracy, and Fairness: A Study of Discretionary Penalties in the NFL,” Kevin Snyder and Michael Lopez evaluate the consistency of specific discretionary penalties in NFL football. The authors focus on examining the occurrence of holding and pass interference calls. After accounting for game and play specific variables, the authors find through a generalized linear mixed modeling approach that the probability of both penalty types is low at the beginning and ends of the game, but high in the middle.
Finally, “Fair Compensation for Gate and Wind Conditions in Ski Jumping—Estimated from Competition Data Using a Mixed Model” by Magne Aldrin investigates the fairness of the existing compensation system in competitive ski-jumping by an analysis of the results from 80 competitions. The compensation system is intended to account for differential wind conditions and the gate from which ski-jumpers start, but the question of whether the adjustment is reflected in jump performance is raised. The analysis is performed by examining the fit of various mixed effect regression models and results in a conclusion that the existing compensation system requires improvements to account for head winds and tail winds.
These articles are available to all members of the Section on Statistics in Sports and on a subscription basis from the JQAS website. Also, prospective authors can find the journal’s aims and scope and manuscript submission instructions there.
Leave a Reply