Baseball Monte Carlo Simulation Project

1. Introduction

This project implements a Monte Carlo simulation for baseball games to predict win probabilities between two teams (Chicago Cubs vs. Chicago White Sox). The simulation uses player statistics (Batting Average, On-Base Percentage, Slugging Percentage for batters; ERA for pitchers) loaded from CSV data files. It runs a large number of simulated games (e.g., 10,000) to derive statistical outcomes.

The primary outputs include:

• Text-based win percentages for each team.
• A comparative bar chart showing the score distribution for both teams.
• A line chart illustrating the cumulative win percentage for each team as more simulations are run.

This project was developed as part of the "Discussion 8 | Monte Carlo Simulation" assignment, emphasizing code organization, data management, and project management practices, including collaboration with AI (Gemini).

2. Project Structure

The repository is organized as follows:

• baseball-simulation-project/

  • README.md (This file)
  • documentation_and_project_management_artifacts/
    • Activity_List.md
    • Functional_Specs.md
    • Product_Backlog.md
    • Roadmap.md
    • Status_Log.md
    • Work_Breakdown_Structure.md
  • functional_code/
    • baseball_simulation.py
  • prepared_data/
    • cubs_standard_batting_clean.csv
    • cubs_standard_pitching_clean.csv
    • whitesox_standard_batting_clean.csv
    • whitesox_standard_pitching_clean.csv
  • results/
    • comparative_score_distribution.png
    • cumulative_win_percentage.png

• /functional_code/: Contains the Python script (baseball_simulation.py) that runs the simulation.

• /prepared_data/: Contains the cleaned CSV data files for player and pitcher statistics used by the simulation. These files were sourced from the pitch-by-pitch-pro repository.

• /documentation_and_project_management_artifacts/: Houses all project planning and documentation. See links below for details.

• /results/: Stores the visual outputs (plots) generated by the simulation.

• README.md: Provides an overview of the project (this file).

3. How to Run

1. Prerequisites: Ensure you have Python installed, along with the pandas, matplotlib, and numpy libraries.

   pip install pandas matplotlib numpy

2. Clone the repository (if you haven't already):

   git clone https://github.com/hongyu-liao/baseball-simulation-project
   cd baseball-simulation-project

3. Ensure Data is Present: The prepared_data folder should contain the four necessary CSV files:
   • cubs_standard_batting_clean.csv
   • cubs_standard_pitching_clean.csv
   • whitesox_standard_batting_clean.csv
   • whitesox_standard_pitching_clean.csv
4. Run the Simulation: Navigate to the functional_code directory and execute the Python script:

   cd functional_code
   python baseball_simulation.py

5. View Results:
   • Text output showing win percentages will be printed to the console.
   • Image files (comparative_score_distribution.png and cumulative_win_percentage.png) will be saved in the results directory (which will be created in the functional_code directory if it doesn't exist there, or you can adjust the save path in the script to point to the root results/ folder).

4. Project Management & Documentation

Detailed project management artifacts provide insight into the planning, execution, and collaboration involved in this project.

• Functional Specifications: Describes user stories, requirements, and acceptance criteria.
• Work Breakdown Structure (WBS): Outlines the major tasks and subtasks.
• Product Backlog: Lists prioritized features and tasks.
• Status Log: Tracks project progress, decisions, and issues over time.
• Activity List: Provides a timestamped log of development activities.
• Roadmap: Shows a high-level timeline of milestones.

5. AI Collaboration

Generative AI (Gemini) was utilized as a co-developer in this project to assist with:

• Initial code structure and boilerplate generation.
• Drafting and refining project management documentation.
• Brainstorming solutions and debugging code.
• Generating code snippets for specific functionalities (e.g., plotting).

Details of this collaboration are reflected in the Status Log and Activity List.

---

Group Members:

• Hongyu Liao
• Yiwei Li
• Ziyang Huang