Baseball prediction using ensemble learning

As the salaries of baseball players continue to skyrocket and with the ever-increasingpopularity of fantasy baseball, the desire for more accurate predictions of players futureperformances is building both for baseball executives and baseball fans. While most existingwork in performance prediction uses purely statistical methods, this thesis showcases researchin combining multiple machine learning techniques to improve on current prediction systemsby increasing the accuracy of projections in several key offensive statistical categories. Byusing the statistics of players from the past thirty years, the goal of this research is to moreaccurately learn from this data how a players performance changes over time and applythis knowledge to predicting future performance. Results have shown that using machinelearning techniques to predict a players performance is comparable to the accuracy seen bysome of the best prediction systems currently available.