π VolleyViz: A volleyball performance analysis system using ML + NCAA stats to simulate games π
VolleyViz is a volleyball performance analysis system that leverages NCAA college stats and machine learning to simulate games. Whether you're a fan wanting to engage with the game on a deeper level or a coach looking to analyze matchups, VolleyViz offers an interactive and user-friendly web app to visualize and explore team and player stats.
π Features
- Game Simulation: Use machine learning models to simulate volleyball games based on NCAA stats.
- Interactive Dashboard: Explore team and player statistics in a visually engaging way.
- Matchup Visualization: Visualize potential matchups and see how teams are forecasted to perform against each other.
- User-Friendly Interface: Designed for both fans and coaches with easy-to-navigate features.
πΊοΈ Navigation
VolleyViz consists of several key pages to help you engage with and analyze volleyball stats:
Rankings Page: View and explore team rankings based on various metrics.
Team Selection Page: Select teams to view detailed statistics and compare their performance.
Stat Dashboard Page: Access a comprehensive dashboard displaying team and player stats.
Matchup Output Page
a. Matchup Page: Visualize and analyze simulated matchups between teams.
b. Win Prediction Component: Predicts the winning team and the forecasted number of sets.
c. Forecasted Statline: Provides a detailed statline for the matchup as well as a computed metric for comparing the teams' performance.
d. Offensive and Defensive Advice: Offers strategic advice for the losing team to improve both offensive and defensive play.
π» Technologies
- Frontend: React.js
- Backend: Flask
- Data Processing: Pandas
- Machine Learning: Scikit-learn, Keras
π€ Contributing
Below are the key roles and team members involved in this project:
- Project Manager: Arnav Akula
- Project Members: Ruba Thekkath, Ankita Khatri, Nikhil Karthikeyan
- Media Member(s): Lauren Lee
- Business Member(s): Akshay Raj
π― Areas of Improvement
-
Tailoring to UC Davis Volleyball Teams: Customizing the application specifically for UC Davis volleyball teams to enable coaches to use it as a training tool.
-
Optimizing Stat Prediction Model: Reducing the computational cost and memory footprint of our stat prediction model to improve performance and scalability.
-
Enhancing Simulation Aspect: Expanding upon the game simulation feature to make it more engaging and interactive for volleyball fanatics.