Space-Time Analysis of Movements in Basketball Using Sensor Data

Space-Time Analysis of Movements in Basketball Using Sensor Data

Rodolfo Metulini, Marica Manisera, Paola Zuccolotto

This study, published in 2017, presents an innovative approach to analyze basketball players’ movements through spatial-temporal data collected through GPS devices. The main goal of the study was to provide coaches and analysts with advanced tools to better understand game dynamics, going beyond traditional statistics.

The study is based on a limited sample of players belonging to the same team; however, it is hoped that future applications of the methodology can be extended to larger and more representative samples, possibly including both teams on the court, and the analysis of more games and more real competitive contexts.

 

🎯 Purpose of the Study

The study aims to:

  • Segment a basketball game into time intervals characterized by homogeneity in players’ spatial configurations.
  • Identify and label phases of play (offense, defense, transitions) based on players’ positions and movement patterns.
  • Analyze transition probabilities among these phases to uncover recurring tactical dynamics.

 

🧪 Methodology

  1. Data Collection

Player movement data were captured using wearable GPS devices, allowing real-time tracking of player trajectories and positioning throughout full games.

  1. Clustering Analysis

The authors used a clustering technique to divide the game into segments, grouping moments with similar spatial formations of players.

  1. Phase Classification

Each cluster was classified as offensive, defensive, or transitional based on the location and relative positions of the players.

  1. Transition Matrices

Transition matrices were built to calculate the probabilities of moving from one game phase to another, revealing patterns in team strategies and momentum shifts.

 

📊 Key Findings

  • Game Phase Identification:
    The clustering method successfully isolated offensive, defensive, and transition moments with high accuracy.
  • Spatial Differences:
    Players were more compact and closely positioned during defensive phases, while spacing increased during offense and transition plays.
  • Transition Patterns:
    The transition matrices revealed frequent and structured sequences, such as the typical shift from defense to offense through a fast break.

 

🧩 Practical Applications

  • Tactical Analysis:
    Coaches can use these insights to improve spatial control, optimize formations, and adjust game strategies in real-time or during post-game analysis.
  • Player Performance Assessment:
    This method allows for deeper evaluation of player impact by analyzing off-ball positioning and movement, not just statistics like points or rebounds.
  • Real-Time Integration:
    The analytical approach can be embedded into performance monitoring systems, offering feedback during training or live matches.

 

 

CITATION

Metulini, R., Manisera, M., Zuccolotto, P. (2017), Space-Time Analysis of Movements in Basketball using Sensor Data, Statistics and Data Science: new challenges, new generations SIS2017 proceeding. Firenze Uiversity Press. eISBN: 978-88-6453-521-0

REFERENCE – ORIGINAL ARTICLE

 

Leave a Reply