Functional Data Analysis of the Power-Duration Relationship in Cyclists.

Publication/Presentation Date

10-1-2025

Abstract

PURPOSE: To extract, prioritize, and model the highly conserved variations in mean-maximal power (MMP) data in cyclists utilizing functional principal component (FPC) analysis.

METHODS: A 3-parameter model (F3 model) was derived from the first 3 functions identified by FPC analysis of a large MMP data set. The F3 model was assessed for goodness of fit to a reserved out-of-sample partition of the MMP data set. Post hoc external validation was used to test the sensitivity of the second FPC to sprint and endurance bias within published data.

RESULTS: The first 3 FPCs accounted for 97% of the variation in the MMP data. The FPCs were interpretable as gain, sprint-endurance bias, and W' analog functions, respectively. The F3 model showed excellent out-of-sample goodness of fit. FPC2 discriminated between sprint- and endurance-biased data in a post hoc analysis.

CONCLUSION: FPC analysis is a powerful tool to statistically identify functions that describe the principal modes of variation in MMP data. The identified functions and resulting F3 model show great promise for performance prediction, as well as revealing novel insights into the mechanistic determinants of exercise performance.

Volume

20

Issue

10

First Page

1331

Last Page

1340

ISSN

1555-0273

Disciplines

Business Administration, Management, and Operations | Health and Medical Administration | Management Sciences and Quantitative Methods

PubMedID

40829766

Department(s)

Administration and Leadership

Document Type

Article

Link to Full Text

Share

COinS