A Two-Stage Bayesian Model for Predicting Winners in Major League Baseball
| Autor: | Tim B. Swartz, Tae Young Yang |
|---|---|
| Rok vydání: | 2021 |
| Předmět: | |
| Zdroj: | Journal of Data Science. 2:61-73 |
| ISSN: | 1683-8602 1680-743X |
| Popis: | The probability of winning a game in major league baseball depends on various factors relating to team strength including the past per- formance of the two teams, the batting ability of the two teams and the starting pitchers. These three factors change over time. We combine these factors by adopting contribution parameters, and include a home field ad- vantage variable in forming a two-stage Bayesian model. A Markov chain Monte Carlo algorithm is used to carry out Bayesian inference and to sim- ulate outcomes of future games. We apply the approach to data obtained from the 2001 regular season in major league baseball. The probability of winning a game in major league baseball (MLB) depends on various factors relating to team strength including the past performance of the two teams, the batting ability of the two teams and the starting pitchers. We define the relative strength of a team over a competing team at a given point in time by combining these three factors into a single measurement. Although other factors relating to team strength may influence the probability of winning, we as- sume that their effect is minor, and note that the three main factors mentioned above are those that are traditionally considered in the setting of betting odds by bookmakers (McCune 1989). Bookmakers are also aware that the home field advantage significantly influences the probability of winning, and likewise, we in- clude it in our calculations. In this paper, we propose a two-stage Bayesian model based on the relative strength variable and the home field advantage variable to predict the outcomes of games in MLB. MLB in the United States is divided into two leagues and six divisions. The American League (AL) has three divisions and the National League (NL) has three divisions. Each team plays 162 games in the regular season (April through |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|