A Bayesian Dual-Skill Framework for Roster-Based Cycling Race Outcome Prediction

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Professional road cycling is a team sport where cyclists serve in different tactical roles, yet most predictive models focus solely on individual performance. This paper introduces VeloRost, a Bayesian dual-skill framework that separately models cyclists’ capabilities as leaders and supporting helpers. Using the TrueSkill rating system, we develop three methods for quantifying helper contributions and aggregate them into a roster strength score combined with each cyclist’s leader skill to predict race outcomes. We evaluated our framework through direct ranking using the skill estimation and statistically enhanced learning across seven seasons of cycling data. Results demonstrate that modeling helper skills significantly outperforms state-of-the-art method, achieving NDCG@10=0.443, highlighting the important role of helpers in race outcomes.

Original languageEnglish
Title of host publicationSports Analytics - 2nd International Conference, ISACE 2025, Proceedings
EditorsJin-song Dong, Jing Sun, Xiaofei Xie, Kan Jiang
PublisherSpringer Science and Business Media Deutschland GmbH
Pages193-208
Number of pages16
ISBN (Print)9783032061669
DOIs
StatePublished - 1 Jan 2026
Event2nd International Sports Analytics Conference and Exhibition, ISACE 2025 - Shanghai, China
Duration: 26 Sep 202527 Sep 2025

Publication series

NameLecture Notes in Computer Science
Volume15925 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd International Sports Analytics Conference and Exhibition, ISACE 2025
Country/TerritoryChina
CityShanghai
Period26/09/2527/09/25

Keywords

  • Machine Learning
  • Recommendation System
  • Sports Analytics

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'A Bayesian Dual-Skill Framework for Roster-Based Cycling Race Outcome Prediction'. Together they form a unique fingerprint.

Cite this