Statistical Methods and Computation for Sports Analytics

随着人们对体育的兴趣与日俱增，数据的规模和复杂性也越来越高，在体育分析中进行令人兴奋的工作的机会从未像现在这样大。体育分析中最普遍和最重要的问题之一是球员评估。在职业体育中，球员评价对奖项、名单选择和工资管理都有影响。球员评估是本研究方案的主要目标之一。在许多运动项目中，球员评估传统上依赖于无数的统计数据，这些数据并不都能准确地反映球员对这项运动的贡献。今天发生了变化的是海量数据的可用性，包括球员跟踪数据。例如，在一些职业足球联赛中，在90分钟的比赛中，球的位置和球场上的22名球员每秒被记录10次，外加额外的时间。这些庞大的数据集让研究人员能够调查这项运动的细微差别，并更准确地描绘出运动员贡献的全部。我计划研究的两项运动是正在增长的(但仍是次要的)泡泡球运动和世界上最受欢迎的运动--足球。在Pickleball中，我们有一个来自双打比赛的大型数据集，其中的搭档定期更换。使用比赛比分作为数据，我们计划提供准确的球员实力排名，其中比赛结果与球员的个人贡献相混淆。具体地说，我们感兴趣的是泡泡球在多大程度上是一项强(弱)环节的运动；强(弱)环节的运动是强(弱)方对比赛结果有更大影响的运动。*在足球运动中，我们将开发球员评估方法，利用详细的球员跟踪数据和足球对“场外”活动的洞察。在体育运动中，分析事件数据--即在球附近发生了什么--是很诱人和方便的。考虑没有发生的事情，描述为什么没有发生，并在这些情况下归因于价值，这是我的意图。这个想法是为了复制和加强教练/经理在密切关注特定球员的活动时所推断的内容。分析中的一个技术方面涉及通过聚类方法确定类似的阵型，以便根据许多类似阵型的结果将价值分配给运动员的行动。*由于我们与加拿大国家和省级体育组织的联系，研究活动对于为加拿大球队提供竞争优势非常重要。这些方法也是新颖的，并被移植到自然科学和工程的其他领域。例如，我们提出的聚类方法涉及一个新的时间方面，即团队编队是动态的。

With the growing fascination in sport and the increased availability of data in both size and complexity, the opportunity to do exciting work in sports analytics has never been greater.******One of the most widespread and important problems in sports analytics is player evaluation. In professional sport, player evaluation has an impact on awards, roster selection and the management of salaries. Player evaluation is one of the primary objectives in this research proposal.******Player evaluation in many sports has traditionally relied on a myriad of statistics, not all of which provide an accurate picture of a player's contribution to the sport. What has changed today is the availability of massive amounts of data including player tracking data. For example, in some leagues of professional soccer, the location of the ball and the 22 players on the field are recorded 10 times per second over 90 minute matches plus added time. These massive datasets allow researchers to investigate nuances of the sport and paint a more accurate picture of the entirety of player contribution.******Two sports that I plan on studying are the growing (but still minor) sport of pickleball and the most popular sport in the world, soccer. In pickleball, we have a large dataset from doubles events where the partners regularly change. Using match scores as data, we plan on providing accurate rankings of player strength where match outcomes are confounded by the individual contributions of players. Specifically, we are interested in the degree to which pickleball is a strong (weak) link sport; a strong (weak) link sport is one where the stronger (weaker) partner has a greater influence on the match outcome.******In the sport of soccer, we will develop player evaluation methods that take advantage of detailed player tracking data and soccer insight involving activity "off the ball". It is tempting and convenient in sport to analyze event data - that is, what has happened in the vicinity of the ball. The consideration of what did not happen, characterizing why it did not happen and attributing worth in these circumstances is my intention. The idea is to replicate and enhance what a coach/manager infers when the coach closely follows the activities of a particular player. One of the technical aspects in the analysis involves the determination of similar formations via clustering methods so that worth can be assigned to player actions based on the outcomes from many similar formations.******With our connections to national and provincial Canadian sports organizations, the research activities are important in providing a competitive advantage to Canadian teams. The methods are also novel and translate to other domains in the natural sciences and engineering. For example, our proposed clustering methods involve a novel temporal aspect in the sense that team formations are dynamic.***********