Rock-Paper-Scissors Tournament Analysis

This repository contains the full analysis for the MT220-01 Group Project (Spring 2025), where we explore strategic behavior in a Rock-Paper-Scissors tournament dataset collected during class.

🧠 Team Members

• Alejandro De La Torre
• Isabel Nold
• Brian Tobin
• Matt Schwartz
• Meredith Kendall

📁 Directory Structure

├── data/                    # Original and reformatted datasets
│   ├── competition_last.xlsx
│   ├── competition_last_long.xlsx
│   └── competition_last_long_lagged.xlsx
├── report/
│   ├── tables/              # CSV tables and markdown reports
│   └── figures/             # Visualizations and plots
├── scripts/                 # R scripts for parts 1–4
│   ├── part2_player_summary.R
│   ├── part3_standings_strategy.R
│   └── part4_independent_analysis.R
├── README.md

📜 Project Components

Part 2 – Player Behavior Summary

• Computes hand counts and win/loss/draw outcomes by player.
• Separates behavior by early vs. later match rounds.
• Tests for statistically significant strategy shifts.

Part 3 – Final Standings & Strategy Analysis

• Game and match-level ranking of players.
• Strategy summaries for each team member.
• Comparison of top 5 and bottom 5 players.
• Early round outcome comparison across rankings.

Part 4 – Independent Modeling & Exploration

• Reformats dataset into long panel structure.
• Adds lagged variables (previous throw/result).
• Builds multinomial logistic regression model.
• Visualizes predicted probabilities for each throw.

🔧 How to Run the Code

Each script is self-contained and reproducible. To run the project:

1. Clone this repository.
2. Open in RStudio or VSCode with R support.
3. Run the scripts in this order:
   • scripts/part2_player_summary.R
   • scripts/part3_standings_strategy.R
   • scripts/part4_independent_analysis.R

All outputs will be saved in report/tables/ and report/figures/.

📦 Dependencies

Make sure the following packages are installed:

• readxl
• dplyr
• tidyr
• ggplot2
• writexl
• nnet
• rpart
• rpart.plot

🔄 Git Workflow Tips

• Always Pull before making changes.
• Commit with clear, descriptive messages.
• Push regularly to update your team.

🏁 Final Notes

This project exemplifies applied statistics and collaborative research. For additional details, see generated reports and figures under the report/ directory.