SNAPP

SNAPP

SNAPP is an AI-powered football analysis database that analyzes player statistics to predict the outlook of current college players and visualize comparisons between professional football players.

Created on 11th February 2024

SNAPP

SNAPP

SNAPP is an AI-powered football analysis database that analyzes player statistics to predict the outlook of current college players and visualize comparisons between professional football players.

The problem SNAPP solves

Fans of football and college football players alike can use SNAPP to access the statistics of their favorite football players in a clean, fresh interface. Features such as the player statistics hexagonal overlays allow users to visualize comparisons between a player and other similar players in their position. These similar player projections allow scouts and fans to predict the potential future path of college players in the NFL, based on the history of players that they are similar to. While football statistics databases already exist, SNAPP differs from these with its comparison and prediction features, as well as its overall visual cleanliness.

Challenges we ran into

Frontend:
On the frontend, we ran into many challenges when dealing with the specific formatting of CSS. Due to the many moving parts in the project, the CSS was often finicky and required a lot of attention when building each page of the website. Connecting the frontend to the backend also posed a slight challenge as we then noticed that it was necessary to edit the setup of a couple components.
Backend:
The backend posed a great deal of challenges in terms of being able to web scrape the data and utilize it for our model. While the website that we wished to web scrape data off of originally seemed very easy to web scrape, we discovered over the course of the web scraping process that the html of the website had many inconsistencies that kept leading to errors in the web scraping process. These flaws in the web scraping process decreased the amount of time we were able to work with our models and also decreased the quality of the data that we were ultimately able to work with. However, we ultimately decided that we should be able to make some minor compromises on the quality of the data that we use due to the large number of past football players that we ended up scraping even with some issues web scraping.

Tracks Applied (2)

Sports

The project analyzes football player data and predicts how college football players would perform professionally. It als...Read More

Best Use of MongoDB Atlas

We used MongoDB Atlas to store player data, and we used its vector search capability to determine player similarities am...Read More

Major League Hacking

Discussion

Builders also viewed

See more projects on Devfolio