Date of Submission

Spring 2022

Academic Program

Computer Science

Project Advisor 1

Sven Anderson

Project Advisor 2

Kerri-Ann Norton

Abstract/Artist's Statement

League of Legends (LoL) is the one of most popular multiplayer online battle arena (MOBA) games in the world. For LoL, the most competitive way to evaluate a player’s skill level, below the professional Esports level, is competitive ranked games. These ranked games utilize a matchmaking system based on the player’s ranks to form a fair team for each game. However, a rank game's outcome cannot necessarily be predicted using just players’ ranks, there are a significant number of different variables impacting a rank game depending on how well each team plays. In this paper, I propose a method to predict rank game outcomes based on several different variables that would be collected 14minutes into the games. Using three different machine learning algorithms: Logistic Regression, Random Forest, and Support Vector Machine, I found that game outcomes can be predicted with 79.46% accuracy with all the collected data and newly added variables. I also use the model to show the importance of each variable in helping players make winning decisions. The results show that the difference in gold between the two teams has the most impact on the final result of the games. The skill difference in the three groups can be further examined with extra variables that were not explored during the study.

Open Access Agreement

Open Access

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

This work is protected by a Creative Commons license. Any use not permitted under that license is prohibited.