Robertson Sam, Woods Carl, Gastin Paul
Centre for Exercise and Sports Science, School of Exercise and Nutrition Sciences, Deakin University, Australia.
School of Exercise and Health Sciences, Edith Cowan University, Australia.
J Sci Med Sport. 2015 Sep;18(5):601-6. doi: 10.1016/j.jsams.2014.07.019. Epub 2014 Aug 9.
To develop a physiological performance and anthropometric attribute model to predict Australian Football League draft selection.
Cross-sectional observational.
Data was obtained (n=4902) from three Under-18 Australian football competitions between 2010 and 2013. Players were allocated into one of the three groups, based on their highest level of selection in their final year of junior football (Australian Football League Drafted, n=292; National Championship, n=293; State-level club, n=4317). Physiological performance (vertical jumps, agility, speed and running endurance) and anthropometric (body mass and height) data were obtained. Hedge's effect sizes were calculated to assess the influence of selection-level and competition on these physical attributes, with logistic regression models constructed to discriminate Australian Football League Drafted and National Championship players. Rule induction analysis was undertaken to determine a set of rules for discriminating selection-level.
Effect size comparisons revealed a range of small to moderate differences between State-level club players and both other groups for all attributes, with trivial to small differences between Australian Football League Drafted and National Championship players noted. Logistic regression models showed multistage fitness test, height and 20 m sprint time as the most important attributes in predicting Draft success. Rule induction analysis showed that players displaying multistage fitness test scores of >14.01 and/or 20 m sprint times of <2.99 s were most likely to be recruited.
High levels of performance in aerobic and/or speed tests increase the likelihood of elite junior Australian football players being recruited to the highest level of the sport.
建立一种生理表现和人体测量属性模型，以预测澳大利亚足球联赛选秀情况。
横断面观察研究。
从2010年至2013年的三项18岁以下澳大利亚足球比赛中获取数据（n = 4902）。根据球员在少年足球最后一年的最高选拔水平，将其分为三组之一（澳大利亚足球联赛选秀球员，n = 292；全国锦标赛球员，n = 293；州级俱乐部球员，n = 4317）。获取生理表现数据（垂直跳跃、敏捷性、速度和跑步耐力）和人体测量数据（体重和身高）。计算赫奇斯效应量，以评估选拔水平和比赛对这些身体属性的影响，并构建逻辑回归模型来区分澳大利亚足球联赛选秀球员和全国锦标赛球员。进行规则归纳分析，以确定区分选拔水平的一组规则。
效应量比较显示，州级俱乐部球员与其他两组在所有属性上存在一系列小到中等程度的差异，而澳大利亚足球联赛选秀球员和全国锦标赛球员之间存在微小到小的差异。逻辑回归模型显示，多级体能测试、身高和20米短跑时间是预测选秀成功的最重要属性。规则归纳分析表明，多级体能测试得分>14.01且/或20米短跑时间<2.99秒的球员最有可能被招募。
在有氧和/或速度测试中表现出色，会增加澳大利亚优秀少年足球运动员被招募到该运动最高水平的可能性。
