Assessing Player Contributions in League of Legends Matches: An Analytical Approach

This paper presents a comprehensive study on the contribution of various factors in “League of Legends” (LoL) matches. The research focuses on multiple aspects, including match analysis, data reduction strategies, predictive models, and exploratory data analysis (EDA). We predict match outcomes with significant accuracy using machine learning techniques such as Binary Logistic Regression (BLR). Additionally, we employ Principal Component Analysis (PCA) and Gradient Boosting Regressor (GBR) for dimensionality reduction to simplify the complex interactions among game variables. Our exploratory data analysis identifies key patterns, trends, and relationships within the game data, helping to optimize gameplay strategies. The models demonstrate high accuracy and robustness through rigorous evaluation, and the findings provide valuable insights for players, researchers, and the eSports industry, highlighting the potential of real-time data and machine learning models in enhancing game performance and strategic decision-making.