Player Performance Prediction Project

Welcome to the Player Performance Prediction project! This repository contains a complete workflow for predicting baseball player statistics such as Wins Above Replacement (WAR), Hits, Home Runs (HR), and Stolen Bases (SB) using historical data.

The project includes:

• Data preprocessing and analysis
• Model training and evaluation
• A Flask web application for interactive predictions

Table of Contents

• Overview
• Project Files
• Data Analysis
• Modeling
• Web Application
• Installation
• Usage
• Future Work

---

Overview

The objective of this project is to forecast future baseball player performance metrics using machine learning models. A Flask web app provides an intuitive interface to select players and view predictions into the future.

Key features:

• Analysis of historical player performance data
• Predictions using multiple linear regression models
• Interactive web-based interface

---

Project Files

• notebooks/
  • Mini_Project_2.ipynb: The main notebook containing data exploration, modeling, and Flask app setup.
• data/
  • Player_Offense_by_Season.xlsx: The dataset used for training and testing.
• models/
  • WAR_model.pkl, H_model.pkl, HR_model.pkl, SB_model.pkl: Pre-trained models for forecasting.
• app/
  • app.py: Flask application code.
  • templates/: HTML templates for the Flask app.
• README.md: This documentation.

---

Data Analysis

The dataset includes player statistics across seasons, such as:

• WAR: Wins Above Replacement
• Hits, Home Runs (HR), Stolen Bases (SB)
• Additional performance metrics: Plate Appearances (PA), At Bats (AB), Games Played (G), Slugging Percentage (SLG), On-base Percentage (OBP)

Insights

• Linear regression was used to model relationships between player statistics and future performance.
• Exploratory analysis revealed trends and correlations among variables.

---

Modeling

Four machine learning models were trained to predict:

1. WAR
2. Hits
3. Home Runs (HR)
4. Stolen Bases (SB)

Model Performance

• WAR Model: R²: 0.8774, MSE: 0.6255
• Hits Model: R²: 0.9768, MSE: 57.0402
• HR Model: R²: 0.9013, MSE: 18.2825
• SB Model: R²: 0.4450, MSE: 69.7393

---

Web Application

A Flask web application provides:

1. Player Selection: Users can select a player from a dropdown.
2. Future Forecasts: Predictions for up to 5 future seasons.
3. Interactive Interface: Results displayed in a user-friendly table format.

Routes

• /: Home page with player selection.
• /predict: Displays forecast results.

---

Installation

Prerequisites

Ensure the following are installed:

• Python 3.x
• Flask
• Pandas
• Scikit-learn

Steps

1. Clone the repository:

   git clone https://github.com/chadb12/player-performance-prediction.git
   cd player-performance-prediction
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Run the Flask application:

   python app.py
   

4. Open the app in your browser:

   http://127.0.0.1:5000
   

---

Usage

1. Launch the Flask app.
2. Select a player from the dropdown menu.
3. View forecasts for WAR, Hits, HR, and SB for the next few seasons.

---

Future Work

• Incorporate additional metrics for richer predictions.
• Add clustering techniques to group players by performance.
• Enhance the web interface for better user experience.
• Transition to a production-grade deployment using services like AWS Elastic Beanstalk.

---

Acknowledgments

• Data sourced from StatHead.com