GridX

Inspiration

We wanted to combine football with our interests in finance and prediction modeling.

What it does

GridX analyzes football play-by-play data and player stats to generate insights and predictive models - to help marketing teams to see when and what to post to social media. It helps them explore team performance, player impact, and potential outcomes in a structured, data-driven way.

How we built it

Python for data processing and analysis Pulse Mock API (from PrizePicks) and nflverse dataset for raw data React (Expo) for the frontend NumPy, Pandas, Matplotlib, Scikit-learn for modeling and visualization

Challenges we ran into

Getting charts to render properly in the React web app Interpreting the data and aligning different datasets into a consistent format

Accomplishments we’re proud of

We built a modular analysis pipeline (analysis.py) that can take any game file, process it, and output:

A player-level time series performance index A team-level time series performance index All stored in a clean dictionary-of-matrices format that makes further analysis simple.

What we learned

Deeper understanding of football statistics How to build and connect a full-stack project Working with data visualization in React

What’s next for GridX

Incorporating more granular data, including offensive line and linebacker stats Expanding beyond basic metrics into advanced analytics Improving clarity and usability of the data for both analysts and casual fans