zoexiao0516 / lol2022

In this project, we used data science approaches, inference, prediction, and classification, to answer questions related to the LoL matches.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

League of Legends 2022: Secret Winning Tips from Top-Tier Players

League of Legends (LoL), developed by Riot Games, is the most widely played multiplayer online battle arena (MOBA) video game in the world. In 2020, LoL had more than 100 million monthly active players (https://activeplayer.io/league-of-legends/) and generated revenue of 1.75 billion U.S. dollars in a year (https://www.statista.com/statistics/806975/lol-revenue/). Due to its entertaining and fast-paced gameplay, LoL is one of the most popular eSports in the industry, with an international competitive scene composed of 12 regional leagues. Each year, more than 100 top-tier teams in professional leagues compete against each other in worldwide tournaments to win an enormous prize pool reaching several million dollars (https://www.statista.com/statistics/749024/league-of-legends-championships-prize-pool/). Besides the high prize money, winning the LoL world championship has been considered the highest honor and the ultimate goal for all eSports players. Driven by both incentives, it is crucial for professional teams to identify the determining winning factors and develop appropriate gaming strategies to maximize the probability of victory.

LoL’s gaming mechanics are deliberately designed to have a rigorous evaluation system while preserving unpredictability by inducing complex features across different game stages. In the game, two teams (red or blue) of five players battle in player-versus-player combat to occupy or defend their team’s part of the map. Each player controls a character, known as a “champion”, with unique abilities and different play styles. During the match, the champions collect experience points, upgrade skill levels, earn gold, and purchase items from the store to become more powerful. The goal is to destroy the other team’s base, also called “Neux”. Each base contains a set of resources, including a series of turrets on three lanes, waves of minions that constantly spawn from the inhibitors, monsters within the Jungle, and heralds and drakes. Experience points and gold could be earned from these resources, as well as from killing other champions of the opponent team.

With the hope of learning from LoL’s top-tier players about their keys to winning, this study aims to identify critical winning factors in professional matches of the game, using the League of Legends eSports match data in 2022. This report consists of three parts. In part one, we tested whether the team’s pre-game factor, being on the red/blue team, is independent of winning the match. In the second part, we looked at the contributing factors to predict the team's total gold, one of the commonly recognized winning factors. In the last part, given the gaming statistics, we developed machine-learning models to identify a winning/losing team.

Data source and data dictionary

The data used in the project and the data dictionary explaining all the variables can be found in here.

Code

We implemented our analysis in this jupyter notebbok.

Report

We demonstarted our results and our analysis of the results in the report.

About

In this project, we used data science approaches, inference, prediction, and classification, to answer questions related to the LoL matches.


Languages

Language:Jupyter Notebook 100.0%