The field of sports analytics is rapidly evolving, with a growing focus on the development of innovative machine learning approaches to improve predictive performance and gain insights into team and player dynamics. Recent work has emphasized the importance of incorporating individual player attributes and team-level composition to enhance predictive models, offering new perspectives on player synergy, strategic match-ups, and tournament progression scenarios. Furthermore, there is a increasing interest in comprehensive soccer understanding, with researchers proposing holistic frameworks that integrate rich domain knowledge and multimodal data to enable knowledge-driven reasoning. Another area of research is the development of methods for action spotting and precise event detection in sports videos, with a focus on improving accuracy and efficiency in temporal action localization and event detection. Noteworthy papers in this area include:

• A paper that proposes a machine learning framework for predicting FIFA World Cup match outcomes, which highlights the importance of incorporating individual player attributes and team-level composition.
• A paper that introduces a multi-agent system for comprehensive soccer understanding, which achieves robust performance and provides a thorough analysis of optimal formations and players.
• A survey that provides a comprehensive overview of datasets, methods, and challenges in action spotting and precise event detection in sports, highlighting critical open challenges and promising research directions.