Analytics, Designed For You
Data Dribble is an NBA analytics hub designed to deliver clear, insightful breakdowns of players and teams. Featuring interactive filters and clean visualizations, it helps analysts, coaches, and fans explore player and team performance effectively. If you love the tool, please consider suppporting Data Dribble's run costs below!
💙 Support Data Dribble 💙Our Features
The Data Dribble app will be iterated and improved over time. Currently it has 6 different tools:
Dashboards
A season snapshot of any player or team since 2017. View offense, defense, and impact data.
Lights System
A visualization system using a traffic-light style to highlight team strengths and weaknesses.
Comparisons
Directly compare players and teams across multiple statistical categories.
Player Trajectory
Tracks career progress between 2017 and 2025 using advanced impact metrics.
Scatter Plots
Create custom scatter plots for players or teams using any stat combinations.
Stat Tables
Detailed player/team stats with filters for season, minutes played, and more.
More on Data Dribble
Data Dribble is preceded by a number of awesome NBA analytics tools that I have used extensively over the years. DARKO, Cleaning the Glass, FiveThirtyEight, Dunks & Threes, Crafted NBA and BBall Index to name a few. This project would not have happened without them paving the way.
All player metrics used within Data Dribble are models I have built myself - many are pre-existing metrics such as Regularised Adjusted Plus Minus - so the output may differ with others based on modelling decisions.
Use the sidebar to access and edit all filters. This will give control of displayed Season, Player, Team and more.
Estimated Player Impact (EPI)
The primary metric used in the Data Dribble tool is Estimated Player Impact (EPI). This is a hybrid metric which uses a combination of Regularised Adjusted Plus Minus (RAPM) methodology and box-derived stats. This metric aims to find a happy medium between the strength of North-Star metrics like RAPM which aim to capture pure player scoreboard impact, while also acknowledging the value of player production.
The RAPM component of EPI is created using a Bayesian Prior (A value which is an assumption of a player's impact before the model observes game data). In this case the prior is a Statistical Plus Minus model which includes Offensive (Points, FT Rate, OREB%, 3P Rate, AST%, TOV%, TS%, Rim Rate, USG%, ORTG) and Defensive (Steals, Blocks, DREB%, Matchup AST:TOV, Matchup TS%, Matchup Difficulty, DRTG) stats.
The box component of EPI is similarly split between offense and defense. On offense, considers the offensive portion of Game Score (PTS + 0.4 * FG - 0.7 * FGA - 0.4*(FTA - FT) + 0.7 * ORB 0.7 * AST - TOV) at both per game and per minute levels. The defensive box component takes the defensive components of game score (0.3 * DRB + STL 0.7 * BLK - 0.4 * PF), as well as Matchup Difficulty, and Matchup Efficiency.