How LPS Data Helps Handball Coaches Find Quality Scorers
Researchers used LPS data and machine learning to create a new model for predicting goals in handball. This model helps understand how players perform and plan strategies better, making game analysis and tactics more advanced.
The Expected Goals (xG) model aids coaches and analysts in evaluating how effectively players create and convert scoring chances, even under challenging conditions. This model assesses individual player performance, identifying those who consistently generate or capitalize on high-quality opportunities. By quantifying the quality of scoring chances, it provides a refined measure of offensive efficiency, highlighting players who exceed expected performance levels despite low scoring probabilities.
In this study titled, “Expected Goals Prediction in Professional Handball using Synchronized Event and Positional Data” the authors utilized a comprehensive single-season dataset of event and positional data, alongside machine learning techniques, to develop an Expected Goals (xG) model for handball. Key features included distances, angles, and game context.
The German Men’s Handball Bundesliga (HBL) collaborates with KINEXON Sports, a provider of UWB-based Location Positioning Systems (LPS), for positional data collection. KINEXON’s system, featuring 14 strategically placed anchors, calculates the positions of mobile devices attached to players and the ball using Time Difference of Arrival (TDoA) and Angle of Arrival (AoA) of radio signals. This system operates at 20 Hz for player positions and 50 Hz for the ball, with validated accuracy.
A Guide to Tracking Three Base Performance Metrics in Professional Handball
The study unveils groundbreaking insights into handball strategy and player performance, revolutionizing game analysis and tactical planning. These findings promise to elevate player training, sharpen team strategies, and deepen our understanding of handball dynamics.
The Expected Goals model shines a spotlight on individual excellence, pinpointing players who consistently generate or capitalize on high-quality scoring chances. This data-driven approach is pivotal for maximizing player contributions in offensive plays, offering a precise measure of offensive efficiency. Remarkably, some players consistently outperform expectations, showcasing superior skills even when their scoring odds are low.
If you’d like to find out more about performance metrics in handball, contact KINEXON Sports.
Start Benefiting From Live Handball Data to Increase Revenue and Fan Engagement
Expected Goals Model and LPS: Powerful Tools
The integration of an xG model into handball analysis, as demonstrated in this study, provides a holistic assessment of both teamwide and individual offensive performance, offering valuable insights into offensive and defensive prowess.
The authors say their research marks a pivotal advancement towards a data-driven approach in handball analytics, fostering objective evaluations and strategic decisions. The findings highlight the potential of machine learning in sports analytics, paving the way for exciting future research opportunities.
It also unveils groundbreaking insights into handball strategy and player performance, revolutionizing game analysis and tactical planning. These findings promise to elevate player training, sharpen team strategies, and deepen our understanding of handball dynamics.
The Expected Goals model shines a spotlight on individual excellence, pinpointing players who consistently generate or capitalize on high-quality scoring chances. This data-driven approach is pivotal for maximizing player contributions in offensive plays, offering a precise measure of offensive efficiency. Remarkably, some players consistently outperform expectations, showcasing superior skills even when their scoring odds are low.
If you’d like to find out which performance metrics matter most in handball, or how the LPS system works, feel free to contact us at any time.