The evaluation of football players: an in-depth look at the Expected Goal metric
研究聚焦足球预期进球（xG）指标，通过Understat数据识别当前模型并提出改进方法，帮助俱乐部、投资者和博彩公司更准确评估球员表现与市场价值。
Abstract The global sports industry is valued at $480 billion in 2023, projected to rise to $629.81 billion by 2028, with the betting sector alone generating €1.41 trillion in turnover in 2022, €730 billion of which was from football. Football attracts millions of fans worldwide, creating lucrative opportunities for advertising and marketing. In Europe, the Big Five leagues represent most of the European professional sports market. In this context, it is important to be able to evaluate football clubs and players. The value of a football player depends mainly on his performance, which depends on different factors, such as personal characteristics, attacking and defending ability on the pitch, mentality, and much more. This study focuses on attacking performance, specifically Expected Goals (xG), a critical metric for financial and strategic decisions in football, since it helps clubs, investors, and bookmakers to evaluate more accurately team performance and identify value bets. In addition, the xG metric can also significantly influence a player’s market value and transfer decisions. This study has two main objectives: first, we want to identify the current xG model through data from Understat, one of the primary data providers for shot characteristics; then, we want to propose an improved approach for more accurate xG predictions.