Establishing an Evaluate Model for Attack Power of Chinese Professional Baseball League Players

• Main Authors: LIN,CHIA-FENG, 林家鳳
• Other Authors: LEE,MONG-HONG
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/28546207088165938323

碩士 === 國立臺北大學 === 統計學系 === 104 === The purpose of this study is to investigate the major factors of a regular player of CPBL this year who will still be a regular player next year. The method of decision tree analysis of data mining technique is applied as the research method of this study. The records in regular CPBL fielders from 2010 to 2015 are adopted in this study. Hitting records and their height, weight and age of a player are chosen as explanatory variables. A player will still be CPBL regular players next year or not is the response variable. The impact factors of CPBL players remains in the regular team next year is summarized. And critical standards of players in the regular team are investigated. Finally, a forecast model is also established to provide players as self-review in the end of season and to be the references of players choosing by baseball organization. The result of this study shows that plate appearances (PA) and age are the most two important factors. Players with more PAs have higher probability to be CPBL regular players next year. Because under the sever conditions of competition, coaches will go through various assessments to determine whether a player can play the game. It is a predictable result that a player can receive affirmation performances then who will have a positive impact for staying in the coming year. The conditions of player himself are also very important. Young players can take advantage of the speed of the first and followed by constant practice to get success. On the contrary, when the age of player is increasing, although his speed can’t catch young players, he still can be a regular player by his playing skill. Keywords: CPBL, Decision Tree, CART