Unveiling the Relationship Between Fan Sentiment and Player Performance - Assessing Players, Key Factors, and Field Position Biases

Abstract

In today's dynamic sports landscape, characterized by an abundance of data and the influence of social media, we venture into a comprehensive exploration of the complex connection between player performance and fan sentiment. Leveraging the data generated by fan sentiment expressed on Twitter, we seek to determine patterns that reflect how player performance is perceived by fans, contributing with fresh perspectives to the field of sports analytics. In this study, we developed and implemented a method for gathering Twitter data related to the 2022/2023 Premier League season, focusing on identifying the sentiments expressed in the opinions shared. Additionally, we created a player identification algorithm based on Levenshtein distance, enabling us to link specific tweets to individual players and subsequently calculate the sentiment ratio for each player in each game. Using this extensive dataset, we analyzed correlation studies between player performance and expert ratings from Fotmob and BBC. Furthermore, we employed multiple linear regression to identify the primary factors influencing fan sentiment, shedding light on what drives their assessments. Ultimately, we unveiled the relationship between players' on-field positions and the sentiments expressed by fans. The outcomes of this study reveal that fan sentiment, following the Wisdom of Crowds principle, can serve as a meaningful indicator of player performance. Moreover, we have identified the key determinants influencing fan responses to player performances. This innovative approach is a valuable tool for future player performance assessments, emphasizing the significance of considering fan opinions in this domain.