基于GBDT模型的篮球比赛“四因素”对球队制胜模式影响研究——以巴黎奥运会与东京奥运会男子篮球赛为例  

Impact of Basketball's“Four Factors”on Team Winning Patterns Based on the GBDT Model——A Case Study of Men's Basketball Games at the Paris and Tokyo Olympics

作　　者：张煜明 敬龙军 Zhang Yuming;Jing Longjun(School of Physical Education,Hunan University of Science and Technology,Xiangtan 411201,China)

机构地区：[1]湖南科技大学体育学院,湖南湘潭411201

出　　处：《体育科技文献通报》2025年第3期11-14,共4页Bulletin of Sport Science & Technology

摘　　要：为探究篮球比赛“四因素”对球队制胜模式的影响,本文运用GBDT模型,将“四因素”即有效投篮命中率(Effective Field Goal Percentage, eFG%)、失误率(Turnover Rate, TOV%)、进攻篮板率(Offensive Rebound Rate, ORB%)、罚球率(Free Throw Rate, FTR%)作为自变量,比赛净胜分作为因变量,对近2届奥运会男子篮球赛的“四因素”数据进行回归分析。结果显示:该模型能够阐释自变量(“四因素”)与净胜分之间的关联,不仅具备较高的拟合度,同时拥有较高的精度;自变量(“四因素”)对模型输出值的影响程度排序为eFG%>TOV%>ORB%>FTR%。To investigate the influence of basketball's “Four Factors” on team winning patterns, this study employs a Gradient Boosted Decision Trees(GBDT) model, utilizing the Four Factors—effective field goal percentage(eFG%),turnover rate(TOV%),offensive rebound rate(ORB%),and free throw rate(FTR%)—as independent variables and net scoring margin as the dependent variable.Regression analysis was conducted on data from men's basketball games at the recent Paris and Tokyo Olympics.The results demonstrate that the GBDT model effectively captures the relationship between the Four Factors and net scoring margin, exhibiting both strong goodness-of-fit and high predictive accuracy.The relative importance of the Four Factors in determining model outputs follows the order: eFG%>TOV%>ORB%>FTR%.

关 键 词：GBDT模型 四因素 制胜模式 男子篮球赛 

分 类 号：G841[文化科学—体育训练]