Skip to main content
Hero Light

​
Welcome to the NBA Statistics Data Platform

This data platform aggregates comprehensive NBA statistics from multiple authoritative sources, providing historical and current season data covering the 2014-15 through 2024-25 NBA seasons. Built with Python, the platform combines web scraping, API integration, and automated data pipelines to deliver consistent, structured datasets for advanced basketball analytics.

Quick Start Guide

Get up and running in minutes with installation and your first data collection

Data Sources

Explore the three major data sources: NBA.com, Basketball Reference, and pbpstats

API Reference

Browse the complete collection of data extraction scripts and utilities

Data Schema

Understand the structure of output datasets and available metrics

​
Key Features

​
Multi-Source Data Aggregation

The platform intelligently combines data from three complementary sources:
  • NBA.com Stats API: Official league statistics including tracking data, shooting metrics, and advanced analytics
  • Basketball Reference: Historical totals, per-possession stats, and player identification
  • pbpstats.com API: Play-by-play derived metrics, on/off data, and advanced possession statistics
Each source provides unique data points that are merged using player and team identifiers to create comprehensive datasets.

​
Historical Coverage (2014-2025)

Access 11+ seasons of NBA data organized by:
  • Regular Season: Complete data for all regular season games
  • Playoffs: Separate datasets for postseason performance
  • Year-over-year structure: Data organized in year-based directories (2014/, 2015/, …, 2025/)

​
Comprehensive Statistical Categories

  • Speed and distance metrics
  • Touches and time of possession
  • Drives and paint touches
  • Hustle statistics (deflections, loose balls, screen assists)
  • Shot quality by defender distance (Very Tight, Tight, Open, Wide Open)
  • Dribble-based shooting splits (0, 1, 2, 3-6, 7+ dribbles)
  • Catch & shoot vs. pull-up shooting
  • Shot zone frequency and efficiency
  • Assists by type (3PT, at-rim, mid-range)
  • Potential assists and assist conversion rates
  • Secondary assists
  • Bad pass turnovers and steals
  • Opponent field goal percentage by distance
  • Rim protection metrics (frequency, accuracy)
  • Defensive matchup data
  • Team defensive tracking
  • Player salaries by season
  • Team cap holds and dead money
  • Contract options (player/team)
  • Cap space analysis

​
Automated Data Pipeline

The platform includes Python scripts that:
  • Make authenticated API requests with proper headers
  • Respect rate limits with built-in delays
  • Handle pagination and multi-year data collection
  • Automatically merge new data with historical datasets
  • Generate unified master CSV files for easy analysis

​
Data Output Format

All data is stored as CSV files with consistent schemas: 
workspace/source/
├── 2014/                    # Season directories
│   ├── defense/
│   ├── player_shooting/
│   ├── tracking/
│   └── playoffs/
├── 2015/
├── ...
├── 2025/
├── hustle.csv              # Master files (all seasons)
├── passing.csv
├── defense_master.csv
├── player_shooting.csv
└── tracking.csv
Master CSV files contain consolidated data from all seasons, while year-specific directories contain individual season data.

​
Technology Stack

The platform is built with:
  • Python 3.x: Core scripting language
  • pandas 1.5.3: Data manipulation and CSV operations
  • requests 2.32.3: HTTP requests to APIs
  • BeautifulSoup4 4.12.3: HTML parsing for web scraping
  • nba_api 1.6.1: Official NBA.com API wrapper
  • plotly 5.23.0: Data visualization (optional)

​
Use Cases

This data platform supports:
  • Player evaluation: Compare players across multiple statistical dimensions
  • Team analytics: Analyze team offensive and defensive tendencies
  • Historical trends: Track how the game has evolved from 2014-2025
  • Predictive modeling: Build models using comprehensive feature sets
  • Scouting reports: Generate detailed performance profiles
  • Contract analysis: Evaluate player value relative to salary

​
Getting Started

Ready to dive in? Head over to the Quick Start Guide to:
  1. Set up your Python environment
  2. Run your first data collection script
  3. Query and analyze the resulting CSV data

View on GitHub

Explore the source code and contribute to the project

Build docs developers (and LLMs) love

Get started for freeTalk to us