Repositorio Institucional de la Universidad de Lima

El Repositorio Institucional Ulima centraliza, preserva y difunde en acceso abierto la producción académica, científica y cultural de la Universidad de Lima.

Repositorio Institucional de la Universidad de Lima

El Repositorio Institucional Ulima centraliza, preserva y difunde en acceso abierto la producción académica, científica y cultural de la Universidad de Lima.

Artículo

Predictive machine learning models for match outcomes in taekwondo based on competitive history

Fecha
2025
Altmetric
Resumen
Taekwondo is an Olympic combat sport where performance depends on speed, strength, and tactical precision. Although data-driven methods are advancing in sports science, predictive modeling in taekwondo remains limited. Most existing studies focus on physical metrics or more popular disciplines, leaving a gap in outcome prediction based on competitive history. In this study, we analyze the contribution of technical and contextual features to match outcomes, aiming to identify the most relevant predictors of success. We also propose a dual-structured dataset: one version models individual match sequences, and the other captures pairwise confrontations. This design allows evaluation under both temporal and head-to-head prediction frameworks. Using official data from the Peruvian Taekwondo Sports Federation, we trained and compared eight machine learning models. LightGBM achieved the highest F1-score (84.00%) in the sequence-based format, while XGBoost performed best (75.00%) in the pairwise version. Feature importance analysis revealed second-round actions—clean points and penalties—as key predictors. Our findings demonstrate that machine learning can effectively identify technical and contextual variables that influence match outcomes, offering valuable support for performance improvement, training optimization, and strategic planning in high-performance taekwondo.
Editor
SAGE
Cita bibliográfica

Velásquez Chávez, D. S., Utani Bendezú, X. N., & Escobedo Cárdenas, E. J. (2025). Predictive machine learning models for match outcomes in taekwondo based on competitive history. SAGE. https://hdl.handle.net/20.500.12724/24464

Velásquez Chávez DS, Utani Bendezú XN, Escobedo Cárdenas EJ. Predictive machine learning models for match outcomes in taekwondo based on competitive history. SAGE; 2025. Disponible en: https://hdl.handle.net/20.500.12724/24464

@misc{20.500.12724/24464,
	author = "Velásquez Chávez, Daphne Solange and Utani Bendezú, Ximena Nataly and Escobedo Cárdenas, Edwin Jonathan",
	title = "Predictive machine learning models for match outcomes in taekwondo based on competitive history",
	publisher = "SAGE",
	year = "2025",
	url = "https://hdl.handle.net/20.500.12724/24464",
}
  • dc.contributor.author
    Velásquez Chávez, Daphne Solange
    Utani Bendezú, Ximena Nataly
    Escobedo Cárdenas, Edwin Jonathan
  • dc.contributor.other
    Escobedo Cárdenas, Edwin Jonathan
  • dc.contributor.student
    Velásquez Chávez, Daphne Solange (Ingeniería de Sistemas)
    Utani Bendezú, Ximena Nataly (Ingeniería de Sistemas)
  • dc.date.accessioned
    2026-03-02T20:53:06Z
  • dc.date.available
    2026-03-02T20:53:06Z
  • dc.date.issued
    2025
  • dc.description.abstract
    Taekwondo is an Olympic combat sport where performance depends on speed, strength, and tactical precision. Although data-driven methods are advancing in sports science, predictive modeling in taekwondo remains limited. Most existing studies focus on physical metrics or more popular disciplines, leaving a gap in outcome prediction based on competitive history. In this study, we analyze the contribution of technical and contextual features to match outcomes, aiming to identify the most relevant predictors of success. We also propose a dual-structured dataset: one version models individual match sequences, and the other captures pairwise confrontations. This design allows evaluation under both temporal and head-to-head prediction frameworks. Using official data from the Peruvian Taekwondo Sports Federation, we trained and compared eight machine learning models. LightGBM achieved the highest F1-score (84.00%) in the sequence-based format, while XGBoost performed best (75.00%) in the pairwise version. Feature importance analysis revealed second-round actions—clean points and penalties—as key predictors. Our findings demonstrate that machine learning can effectively identify technical and contextual variables that influence match outcomes, offering valuable support for performance improvement, training optimization, and strategic planning in high-performance taekwondo.
  • dc.format
    html
  • dc.identifier.doi
    https://doi.org/10.1177/17479541251363562
  • dc.identifier.isni
    0000000121541816
  • dc.identifier.issn
    2048-397X
  • dc.identifier.journal
    International Journal of Sports Science and Coaching
  • dc.identifier.scopusid
    2-s2.0-105012840983
  • dc.identifier.uri
    https://hdl.handle.net/20.500.12724/24464
  • dc.identifier.wosid
    WOS:001543055100001
  • dc.language.iso
    eng
  • dc.publisher
    SAGE
  • dc.publisher.country
    GB
  • dc.relation.ispartof
    urn:issn: 2048-397X
  • dc.rights
    info:eu-repo/semantics/restrictedAccess
  • dc.subject
    Pendiente
  • dc.subject.ocde
    https://purl.org/pe-repo/ocde/ford#2.02.04
  • dc.title
    Predictive machine learning models for match outcomes in taekwondo based on competitive history
  • dc.type
    info:eu-repo/semantics/article
  • dc.type.other
    Artículo (Scopus / Web of Science)
ALICIA
Renati
La Referencia
GoogleScholar