Real Performance is Messy
Divergence uses machine learning and predictive modeling to analyze athlete-level responses to critical stimuli to guide training and predict performances.
Machine learning built around cycling physiology
Environmental Modeling
Understand how heat, elevation, and changing conditions impact performance.
Fatigue Dynamics
Move beyond simple freshness metrics toward modeled fatigue and recovery interactions.
Predictive Outputs
Generate scenario-based insights rather than relying solely on retrospective analysis.
From data collection to decision support
The objective is not more dashboards. The objective is actionable interpretation: understanding what changes performance, by how much, and under which conditions.
Built around real-world physiological data
Divergence is designed for native integration with modern sensor ecosystems, enabling predictive models to incorporate richer physiological signals alongside traditional ride data.
Development currently emphasizes metrics derived from core body temperature and respiratory physiology, including workflows designed around CORE and Tymewear data streams. These inputs help extend modeling beyond power files alone toward environmental response, fatigue dynamics, and performance prediction.
