Abstract:This paper presents a study on the prediction of outcomes in matches of the electronic game League of Legends (LoL) using machine learning techniques. With the aim of exploring the ability to predict real-time results, considering different variables and stages of the match, we highlight the use of unpublished data as a fundamental part of this process. With the increasing popularity of LoL and the emergence of tournaments, betting related to the game has also emerged, making the investigation in this area even more relevant. A variety of models were evaluated and the results were encouraging. A model based on LightGBM showed the best performance, achieving an average accuracy of 81.62\% in intermediate stages of the match when the percentage of elapsed time was between 60\% and 80\%. On the other hand, the Logistic Regression and Gradient Boosting models proved to be more effective in early stages of the game, with promising results. This study contributes to the field of machine learning applied to electronic games, providing valuable insights into real-time prediction in League of Legends. The results obtained may be relevant for both players seeking to improve their strategies and the betting industry related to the game.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to predict the results of "League of Legends" (LoL) games in real - time. Specifically, the authors use machine - learning techniques to explore the ability to predict game results at different points in the game by analyzing different variables and stages of the game. The focus of the research is on using unpublished data as an integral part of this process to provide comprehensive and up - to - date game analysis. In addition, the research also aims to identify the best technique or combination of techniques, especially when using these unique data sources. The paper evaluates the effectiveness of multiple models and tests them at different moments in the game, ensuring a dynamic and up - to - date approach in predicting real - time results in "League of Legends". ### Main Objectives of the Paper: 1. **Real - time Prediction**: Develop machine - learning models capable of predicting the results of "League of Legends" games in real - time. 2. **Data Utilization**: Use recent and unique data to provide comprehensive and up - to - date game analysis. 3. **Technique Selection**: Identify and evaluate the machine - learning techniques or combinations of techniques that are most suitable for predicting game results. 4. **Dynamic Evaluation**: Evaluate the effectiveness of models at different stages of the game to ensure prediction accuracy. ### Research Background: - **Introduction to "League of Legends"**: "League of Legends" is a multiplayer online battle arena (MOBA) game developed by Riot Games. Since its release in 2009, it has become one of the most popular games in the world. - **Importance of Match Prediction**: With the popularity of "League of Legends" and the development of e - sports, match prediction has become increasingly important. It is not only helpful for players to formulate strategies, but also of great significance to the game - related gambling industry. ### Methodology: - **Data Collection**: Use the official API of Riot Games to collect data, covering 64,556 games from January 12, 2023 to May 18, 2023. - **Data Pre - processing**: Convert time values into minutes, divide the data into four scenarios (20% PET, 40% PET, 60% PET, 80% PET) according to the percentage of time elapsed in the game (PET), exclude games with a duration of less than 5 minutes, convert boolean variables into binary values, and remove redundant variables. - **Feature Selection**: Based on correlation analysis, select variables highly correlated with the victory of the blue team, such as the number of kills, the number of dragon kills, first blood, etc. - **Model Building and Evaluation**: Use multiple machine - learning algorithms (such as logistic regression, random forest, naive Bayes, gradient boosting, XGBoost, LightGBM, neural network and Bagging) to build prediction models, and evaluate model performance through cross - validation techniques. ### Results: - **Model Performance**: The LightGBM model performs best in the middle stage of the game (60% - 80% PET), with an average accuracy rate reaching 81.62%. Logistic regression and gradient - boosting models perform well in the early stage of the game and have a high prediction accuracy. ### Conclusions: - **Contributions**: This research provides valuable insights into the application of machine - learning in video games, especially in terms of predicting the results of "League of Legends" games in real - time. - **Application Prospects**: The research results are not only helpful for players to improve their strategies, but also of great significance to the game - related gambling industry. Through these efforts, the paper provides a new perspective and technical support for predicting the results of "League of Legends" games.

League of Legends: Real-Time Result Prediction

Machine learning models for DOTA 2 outcomes prediction

The Application of Machine Learning in Predicting the Results of Popular eSports Games: Win Rate Prediction in MOBA and FPS Games

Predicting Football Match Outcomes with Machine Learning Approaches

Machine Learning for Soccer Match Result Prediction

Prediction of esports competition outcomes using EEG data from expert players

Machine Learning in Football Betting: Prediction of Match Results Based on Player Characteristics

On predictability of rare events leveraging social media: a machine learning perspective

Betting the system: Using lineups to predict football scores

Incremental Learning for Football Match Outcomes Prediction

In-game soccer outcome prediction with offline reinforcement learning

Modelling Competitive Sports: Bradley-Terry-Élő Models for Supervised and On-Line Learning of Paired Competition Outcomes

Predicting Outcomes in Video Games with Long Short Term Memory Networks

Win Prediction in Esports: Mixed-Rank Match Prediction in Multi-player Online Battle Arena Games

Outcome Prediction Of Dota2 Using Machine Learning Methods

On Predicting Soccer Outcomes in the Greek League Using Machine Learning

Match predictions in soccer: Machine learning vs. Poisson approaches

Predicting Events in MOBA Games: Prediction, Attribution, and Evaluation

Analysis and Prediction of Cricket Match Using Machine Learning

Descriptive and Predictive Analysis of Euroleague Basketball Games and the Wisdom of Basketball Crowds

Predicting soccer matches with complex networks and machine learning