Regimes in baseball players’ career data

• Autores: Marcus Bendtsen
• Localización: Data mining and knowledge discovery, ISSN 1384-5810, Vol. 31, Nº 6, 2017, págs. 1580-1621
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • In this paper we investigate how we can use gated Bayesian networks, a type of probabilistic graphical model, to represent regimesin baseball players’ career data. We find that baseball players do indeed go through different regimes throughout their career, where each regime can be associated with a certain level of performance. We show that some of the transitions between regimes happen in conjunction with major events in the players’ career, such as being traded or injured, but that some transitions cannot be explained by such events. The resulting model is a tool for managers and coaches that can be used to identify where transitions have occurred, as well as an online monitoring tool to detect which regime the player currently is in.