NBA Career Predictor and Similar Players

The Vision

We needed to create an MVP in a week. This MVP needed to contain a working frontend, backend, and database to predict how long an NBA player will play in the league while also finding the most similar player.

Data Science Team

This was an interesting build. The data consists of players from 1976 to 2018. We used a Postgres DB and wrote SQL queries as endpoints to communicate with the frontend. Using an xgboost regressor and training, validating and testing on web scraped data, we were able to create an application that can predict a NBA player’s time in the NBA by + or - 1.44 years.

The End Result

The NBA Career Predictor is a working application that predicts how many years a player will stay in the NBA and the most similar player. Here are some pictures:

If you would like to check out the website, message me on socials for a temporary login! Thanks for reading!