Designing a Competitive Point System for Among Us
Overview
Among Us is a 4-15 player social deduction game in which players on a spaceship are secretly assigned the roles of Crewmates or Impostors. Crewmates work collaboratively to complete tasks and identify Impostors, while Impostors aim to eliminate Crewmates without being detected. Since players share a team-based outcome (a win or a loss), all members of a team are credited equally regardless of individual contribution, making it difficult to determine which players on either team had the best performance. As a result, Among Us functions primarily as a social game rather than a competitive one. This project explores how Among Us could be transformed into a competitive experience by developing an algorithm that ranks players based on gameplay performance statistics. To track such data, the game would require a system capable of recording player actions. Since the original Among Us source code is not publicly available, the Among Us game was recreated from scratch with a custom statistics-tracking system using the Godot game engine. A multi-agent simulation was added to automate the generation of large amounts of Among Us gameplay data. This data was used to design a scoring algorithm that rewards actions correlated with team success and penalizes actions that reduce a team’s chances of winning. Using principles from game theory, the scoring system also addresses edge cases that could encourage unintended, undesirable strategies. Simulation results were used to iteratively adjust the scoring system to ensure balanced rewards across roles and gameplay tasks. The results show that a casual social deduction game such as Among Us can engage players in more objectively measured competitive environments through the introduction of a point system based on player performance.
Awards (1)
- Association for the Advancement of Artificial Intelligence: Honorable Mention (do not read aloud). Winners receive a student level membership. Information is included separately in the SAO Portal.
Competition history
- ISEF 2026
Resources
Related projects
ISEF · 2021
Video Game Reward Systems and Their Impact on Players' Behavior in the Iterated Prisoner's Dilemma
ISEF · 2022
Gamifying Learning by Repetition
ISEF · 2021
Dynamic Paired Comparison Prediction Using Modified Elo Ratings
JSHS · 2024
Autonomous Over-The-Board Chess Solution for Players in Isolated Conditions
ISEF · 2024
A Novel and Convenient Go-Games Win-Rate Estimation and Misplay Detection Model Based on Convolutional Neural Networks
ISEF · 2016
Developing and Analyzing Strategies in a Two-Player Domino Tiling Game
ISEF · 2023
An Autonomous Over-the-Board Chess Solution for Players in Isolated Conditions
ISEF · 2020
Novel Perspectives in Academic Practice Strategy: Designing and Exploring Gamification Classroom Practice Approaches that Strategically Target Cognition, Student Motivation, and Engagement to Improve the Academic Performance of Underperforming Elementary School Students
Closest projects by meaning, across every fair and year in the corpus.
Browse more like this
Source: Regeneron International Science and Engineering Fair