Your one stop shop for all things Mission Critical.
NBA contracts are sometimes 9 figures long. It is in a team's best interest to have a good idea of the longevity of a player's productivity. This application uses historical data to find similar players and predict a selected player's likely longevity and success in the league.
- An entry field for the user to enter a current college or NBA basketball player
- A model that finds similarity between entered player and current and historical NBA players
- A display that shows the entered player and the similar player side-by-side with statistics and other determining criteria
- Create a predictive model for career longevity using historical player data (begin with a baseline!)
- Deploy model via Flask API to receive inputs (player names), and output predicted longevity in JSON format.
- User creation/login flow
- Upon logging in for the first time, a user sees a form to send data to the DS API to get a result back on. Work with the DS students to determine what should be included in the form
- When results come back the user can add a name to the player and save the results to their profile.
- User can edit the names of the players saved in their profile
- User can delete saved results from their profile
Stretch:
- Incorporate deeper data such as health data and other background information.
- Create visualizations that compare the player with other similar players and create a predicted "career arc" that shows not just longevity but statistical production projections.
- Find and Add more data from additional sources to improve your model's performance.
- Create a visualization for a player's predicted longevity.
- Create a way to rank the players saved to the user's profile in a way that highlights the best players. DATA SETS / DESIGN LINKS https://data.world/rvino88/1976-to-2015-nba-draft-data