Spearman Correlation and Entropy Weight Method Based Best-of-n Competition Momentum Prediction

Abstract

This paper builds a novel model for predicting momentum in competitive settings, particularly in Best-of-n competitions, using a feature extraction method. The study focuses on quantifying various momentum indicators using relevant match data to capture the dynamic nature of momentum shifts. Key indicators include fundamental momentum and psychological factors such as player confidence, opponent pressure, and unforced errors, as well as offensive momentum. These elements are integrated into a comprehensive assessment framework that allows for a holistic understanding of competitive dynamics. The model employs advanced statistical techniques and algorithms to enhance prediction accuracy, ensuring reliable outcomes. Through further evaluation, the approach is visually demonstrated in predicting momentum trends, making it applicable across diverse competitive environments. By leveraging these insights, stakeholders could make informed decisions, improve strategic planning, and enhance performance in various competitive arenas. Ultimately, it contributes to a deeper understanding of the intricacies of momentum in sports and other competitive fields.