PhD, Associate Professor M.I. Galyautdinov¹
PhD L.R. Galyautdinova²
PhD, Associate Professor R.M. Khusnutdinov^1
1Volga Region State University of Physical Culture, Sport and Tourism, Kazan
²Lebedev Russian State University of Justice, Kazan branch, Kazan

Keywords: sports analytics, big data, machine learning, expected goals (xG), tactical analysis, injury forecasting.

Introduction. Modern sport is undergoing a data-driven revolution. Subjective observations are being replaced by sophisticated mathematical models that not only describe events but also predict them.
The aim of the study is to identify the features of applying modern methods of advanced statistics in sports analytics.

Methods and Research Design. The study is based on an analysis of open scientific publications and case studies from leading sports leagues. Several approaches dominate modern analytics. Regression analysis and clustering are widely used to identify hidden dependencies and segment players. Specialized metrics play a particularly important role, such as Expected Goals (xG) in football [1, 2], which assesses the quality of scoring chances, and Player Efficiency Rating (PER) in basketball, providing a comprehensive assessment of a player's effectiveness. Machine learning methods are becoming increasingly significant, used for predicting match outcomes and assessing injury risk based on workload data. This picture is complemented by tracking analytics, which provides detailed spatiotemporal data on player movement.

Results and Discussion. The implementation of advanced methods has led to a reassessment of many established concepts in sports. For example, analysis based on xG has shifted the focus of teams from simple ball possession to creating high-quality chances, which has proven effective in the long term. In team management, comprehensive metrics allow for an objective assessment of a player's contribution, minimizing the role of subjective opinion during contract negotiations. Significant progress has been made in injury prevention, where algorithms analyzing data from trackers help identify at-risk players and optimize their workloads. New game strategies are also being developed based on tracking data, as happened in basketball with the prioritization of three-point shots and attacks at the rim. Despite its effectiveness, the implementation of analytics faces resistance from specialists of the "old school" and requires high costs, as well as competent data interpretation to avoid erroneous conclusions.

Conclusions. Advanced statistics have become an essential tool for gaining a competitive advantage in modern sports. It provides powerful means for tactical optimization, roster management, and risk reduction. Success increasingly depends on an organization's ability to work effectively with data, making sports analytics a key discipline of the future.

References

1. Robberechts P. How Data Availability Affects the Ability to Learn Good xG Models [Electronic resource] KU Leuven Sports Analytics Blog. 2020. URL: https://dtai.cs.kuleuven.be/sports/blog/how-data-availability-affects-th...
2. Sumpter D. Soccermatics: Mathematical Adventures in the Beautiful Game. London: Bloomsbury Publishing, 2016. 256 p.