ABSTRACT

To date, a number of research studies have examined sport leagues for potential evidence of teams strategically losing games on purpose. Following tournament theory, it is believed sport teams will engage in such practices, often called tanking, in order to gain rewards in the form of better draft picks. Where prior research typically focused on detecting evidence of underperformance by teams, the present research analyzed one possible tanking strategy – the resting of healthy players. Specifically using data from National Basketball Association regular season games from the 2006–07 to 2017–18 seasons, we develop a count model of the number of players who are rested by teams. Furthermore, we utilize a natural experiment to consider whether teams eliminated from playoff contention rest more players. Poisson regression estimates found that eliminated teams will rest more players than others, and that the number of players rested by eliminated teams will increase as the competition for draft picks increases. As such, this study is one of the first to show how teams are able to purposefully lose games, with the strategy being instituted through managerial decisions rather than shirking by workers.

摘要

迄今为止，许多研究已经检查了体育联盟，以寻找球队故意输掉比赛的潜在证据。根据锦标赛理论，人们相信运动队会采用这种做法，通常称为坦克，以便以更好的选秀权形式获得回报。之前的研究通常侧重于检测球队表现不佳的证据，而目前的研究分析了一种可能的坦克策略——让健康球员休息。具体使用 2006-07 至 2017-18 赛季美国国家篮球协会常规赛的数据，我们开发了球队休息球员数量的计数模型。此外，我们利用自然实验来考虑从季后赛竞争中被淘汰的球队是否让更多的球员休息。泊松回归估计发现，被淘汰的球队会比其他球队休息更多的球员，并且被淘汰的球队休息的球员数量将随着选秀权竞争的增加而增加。因此，这项研究是第一个展示团队如何故意输掉比赛的研究之一，该策略是通过管理决策而不是工人推卸责任来制定的。