Display and aggregate information about games, teams, players, coaches, and owners from the National Basketball Association (NBA).

Although one can view recent statistics on NBA's website, there are use cases (for both developers and end users) where this project might be useful. In particular, having an queryable API and a (basic) database design gives developers a starting point to develop more visualizations with greater flexibility, write complex SQL commands to aggregate information in interesting ways, etc. End users could then use this new information to gain more insight into the data.

1. Clone or download the project
2. Install all required packages using requirements.txt (ideally within a virtual environment)
3. Run python manage.py runserver to activate the server from a shell
4. View application on port 8000: http:https://localhost:8000

Despite the fact that datasets can becomes quite large (upto ~100MB per season depending on the data we are interested in logging), we provide a small populated (SQLite) datatbase for the 2015-16 NBA season, only to be used locally for developmental purposes.

It is recommended, however, to use the scripts provided to automatically populate a client-server database (e.g. PostgresSQL, MySQL) in a cloud-based environment (e.g. Amazon RDS). See database/ folder for more details.

To populate a barebones version of the database (see design below), we programatically query endpoints provided by the NBA (e.g. for teams and players).

The Teams table records information about teams. In particular, we log the team id (primary key), home city, team name, which conference they play in, their current rank with respect to the other teams in that conference, and number of wins and losses (up to this point in the season).

Each team consists of players, coaches, owners, each of which are referenced via foreign keys in their respective table. In addition, every season, each team plays in a certain predefined amount of regular season games.

The Players table records basic information about each player (active or inactive). We save the player id (primary key), their name, their age, which team they play on (by team id), and their specific role within that team.

It is important to note that not every player plays in every game for the team (e.g. players can be injured, not get put into the game by the coach). As such, we have a seperate DB table (aptly named PlaysIn) which handles which player played in what game.

The Coaches table records all the (head, assistant, developmental) coaches. For each coach, we have a unique id (primary key), their name, their age, and the team they are associated with (referenced by the team id foreign key).

The Owners table records the majority and minority stakeholders who collectively own a team. Each owner has their own unique id (primary key), their name, their age, their networth, and which team they have a stake in (referenced by the team id foreign key).

Note that there can be multiple owners for a single team, and that (legally) a person can only have a stake in one team at a time.

The Games table records all the stats from each regular/post season game in the specified season. Here we record the date the game occurred, the teams playing in that game, and the winning team (either home or away). Since each team can play multiple times within a season, the primary key is the combination of both team ids and the date when they played.

Other information about the game (e.g. total points, assists, rebounds, etc.) logged by each team is stored in a seperate DB table (GameScore).

• Main page
  • Display info about games from latest date (or last day of the season).
  • Include score, team names, team logo, venue, etc.
  • Panel for daily leaders, e.g. pts, assists, rebounds, etc.
• Teams overview
  • Two column (Eastern and Western conference)
  • Teams ordered by rank, wins/loss record, last 10
• Player overview
  • List players by alphabetical order (grouping by A-Z)
• Games overview
  • Toggle for calendar views (day, week, month, year)
  • Display less/more information about games depending on which is chosen

The dashboard was developed by Rahul Pokharna, Sibi Sengottuvel, and Ayush Karnawat.