(1.清华大学 体育部，北京 100084；2.北京体育大学 中国篮球运动学院，北京 100084；3.上海体育学院 体育教育训练学院，上海 200438；4.北京体育大学 体育工程学院，北京 100084) 摘 要：为了探究高水平篮球比赛中技战术指标与比赛结果的因果关系，选取2019年男篮世界杯92场比赛作为研究对象。运用加权最小二乘法(WLS回归)和分位数回归(QR回归)，将常规变量、高阶变量和情境变量作为自变量，各参赛球队的最终比赛胜率作为因变量进行统计建模。结果显示：(1)相比较对阵弱队，各参赛队在对阵强队时，胜率会明显降低。对于所有球队，防守篮板(P<0.01)是关键的制胜指标，增加防守篮板会增加比赛获胜的概率。此外，对于低胜率球队，助攻(P<0.05)和犯规(P<0.01)与比赛结果呈现显著的正相关；对于高胜率球队，失误(P<0.01)和犯规(P<0.01)对比赛结果有显著的负向影响。(2)分位数回归相比较传统线性回归模型，可以更详细地刻画因变量与自变量之间的关系，获得不同百分位点比赛结果的影响因素，为提升球队的运动表现提供科学依据。 |
ZHANG Shao-liang1，DONG Rui2，WANG Xing2，ZHANG Ming-xin3，CUI Yi-xiong4
(1.Department of Physical Education，Tsinghua University，Beijing 100084，China；2.China Basketball College，Beijing Sport University，Beijing 100084，China；3.School of Physical Education and Sport Training，Shanghai University of Sport，Shanghai 200438，China；4.School of Sports Engineering，Beijing Sport University，Beijing 100084，China) Abstract: The aim of this study was to identify the casual relationship between the match performance indicators and match outcome, and comprised of 92 games during the 2019 Men’s Basketball World Cup of FIBA. Weighted least squares (WLS) and quantile regression (QR) were used to construct model to explore the relationship between independent variables (traditional variables, advanced variables and situational variables) and dependent variables (the final winning percentage for every team). The study showed that: (1) the winning percentage will decreases when confronting with strong teams than competing with weak teams at this tournament. For all teams, defensive rebounds (P < 0.01) were the key performance indicator which was positively related to match outcome. In addition, assists (P < 0.05) and personal fouls (P < 0.01) were significantly positively related to match outcome for low winning percentage teams, whereas turnovers (P < 0.01) and personal fouls (P < 0.01) were significantly negatively related to match outcome for high winning percentage teams. (2) Quantile regression can provide more detailed data information and identify the influencing factors at different percentiles compared with traditional linear regression, which can provide scientific evidence to improve teams’ sports performance. |
下载本期全文：点击下载