排球勝負預測模式之研究－以世界男排聯賽為例
No Thumbnail Available
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
排球運動發展至今，規則與技術已臻成熟，但球場上的變化瞬息萬變，大多數仍以教練主觀意識來決定戰術。因此藉由世界男排聯賽排球技術的數據，來探討排球技術的重要性與預測比賽結果的能力，進而在比賽時能客觀的進行球員調度與各種攻防戰術的調整。目的：本研究旨在利用區別分析及邏輯斯迴歸分析探討排球技術 (得分技術：扣球、攔網、發球、對方失誤；非得分技術：救球、舉球、接發球) 預測比賽結果的能力，並分析排球技術對比賽結果的影響力。藉此，教練才能依據科學及公平的理論來調整球員訓練的比重。方法：經由相關文獻探討堆砌出本研究的架構與方向，蒐集並加以分析2007年至2016年世界男排聯賽排球技術的數據。結果：區別分析挑選出5個自變數，邏輯斯迴歸分析挑選出12個自變數，兩個模式共同挑選出「扣球得分數」、「攔網得分數」、「發球得分數」對勝場具有預測能力，「扣球失誤數」、「接發球失誤數」對負場具有預測能力。結論：從研究結果得知，區別分析能挑選出重要的自變數，而邏輯斯迴歸分析能進一步分析勝場或負場的發生機率，建議搭配不同的統計方法來建構排球勝負預測模式。
With the development of volleyball rules and skills have matured. However, the changes on volleyball courts are hard to predict, and coaches usually decide tactics with their subjective consciousness. Therefore, the study tries to discuss the importance of volleyball skills and the prediction of outcomes by analyzing the data of volleyball skills of FIVB World League in order for coaches to arrange players and adjust offensive and defensive tactics. Purpose: The research aims to explore the ability of volleyball skills (scoring skills: spiking, blocking, serving, opponent error; non-scoring skills: digging, setting, receiving) to predict the results of games by using the discriminant analysis and the logistic regression analysis, and analyze the influence of volleyball skills on the results. Hence, coaches can adjust the proportion of players in the training according to scientific and fair methods. Methodology: Through the relevant literature reviews, the structure and the direction of the study are established by collecting and analyzing the data of volleyball skills of FIVB World League from 2007 to 2016. Results: There are 5 independent variables picked out from discriminant analysis, and 12 independent variables selected from the logistic regression analysis. From both models, the number of spiking scores, the number of opponent errors scores, the number of block scores, the number of serving scores are chosen to predict winning games, and the number of spiking errors and the number of receiving errors are used to predict losing games. Conclusion: From the results of the study, the important independent variables can be chosen from the discriminant analysis, and the logistic regression analysis can be used to analyze the probability of wins or loses. It is recommended to match different statistical methods to construct the prediction model of volleyball.
With the development of volleyball rules and skills have matured. However, the changes on volleyball courts are hard to predict, and coaches usually decide tactics with their subjective consciousness. Therefore, the study tries to discuss the importance of volleyball skills and the prediction of outcomes by analyzing the data of volleyball skills of FIVB World League in order for coaches to arrange players and adjust offensive and defensive tactics. Purpose: The research aims to explore the ability of volleyball skills (scoring skills: spiking, blocking, serving, opponent error; non-scoring skills: digging, setting, receiving) to predict the results of games by using the discriminant analysis and the logistic regression analysis, and analyze the influence of volleyball skills on the results. Hence, coaches can adjust the proportion of players in the training according to scientific and fair methods. Methodology: Through the relevant literature reviews, the structure and the direction of the study are established by collecting and analyzing the data of volleyball skills of FIVB World League from 2007 to 2016. Results: There are 5 independent variables picked out from discriminant analysis, and 12 independent variables selected from the logistic regression analysis. From both models, the number of spiking scores, the number of opponent errors scores, the number of block scores, the number of serving scores are chosen to predict winning games, and the number of spiking errors and the number of receiving errors are used to predict losing games. Conclusion: From the results of the study, the important independent variables can be chosen from the discriminant analysis, and the logistic regression analysis can be used to analyze the probability of wins or loses. It is recommended to match different statistical methods to construct the prediction model of volleyball.
Description
Keywords
得分技術, 非得分技術, 區別分析, 邏輯斯迴歸分析, Scoring Skills, Non-Scoring Skills, Discriminant Analysis, Logistic Regression Analysis