New AI models H-VAEP and H-xT enhance handball player evaluation

Researchers have developed new frameworks, Handball-xT (H-xT) and Handball-VAEP (H-VAEP), to better evaluate player performance in professional handball. These models adapt existing football analytics techniques, Expected Threat (xT) and Valuing Actions by Estimating Probabilities (VAEP), to handball's specific dynamics. The H-xT model uses a handball-native court zoning system, and H-VAEP is optimized to provide stable and intuitive player ratings by considering build-up play and minimizing team-identity leakage. The researchers have released the code to enable professional clubs to implement these advanced valuation models. AI

IMPACT These models offer a more sophisticated approach to player performance analysis in handball, potentially influencing team strategies and scouting.

RANK_REASON The cluster describes a new academic paper introducing novel AI models for sports analytics. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Expected Threat (xT)
• handball
• Handball Bundesliga
• Handball-VAEP (H-VAEP)
• Handball-xT (H-xT)
• H-VAEP
• H-xT
• Valuing Actions by Estimating Probabilities (VAEP)

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →