Related Experiment Video
Updated: Jul 18, 2026

A Real-World High-Intensity Interval Training Protocol for Cardiorespiratory Fitness Improvement
Published on: February 22, 2022
Prediction of one-hour running performance using constant duration tests.
François Xavier Gamelin1, Jéré My Coquart, Nicolas Ferrari
1Laboratoire d'Etudes de la Motricité Humaine, Faculté des Sciences du Sport, Université, Ronchin, France. fxgamelin@hotmail.com
Critical velocity (CV) estimation helps rank runners, with the exponential model showing the strongest correlation to 1-hour performance. However, no current CV models accurately predict actual race performance for coaches or athletes.
Area of Science:
- Sports Science
- Exercise Physiology
- Biomechanics
Background:
- Critical velocity (CV) is theoretically the maximum sustainable velocity without fatigue.
- Accurate CV estimation is crucial for predicting endurance performance in runners.
Purpose of the Study:
- To compare five mathematical models for estimating CV.
- To determine which CV model best correlates with 1-hour running performance.
- To assess the accuracy of CV models in predicting 1-hour performance.
Main Methods:
- Twelve trained male middle- and long-distance runners participated.
- Participants completed constant duration tests (6, 9, 12 min) and a maximal velocity test.
- CV was estimated using two linear, two nonlinear, and one exponential model.
- Actual 1-hour track performance was measured.
Main Results:
- All CV estimates showed significant correlation with 1-hour performance (r > 0.80).
- The exponential model demonstrated the highest correlation (r = 0.93, p < 0.01).
- No model accurately predicted 1-hour performance velocity, with significant bias observed.
Conclusions:
- CV estimation effectively ranks middle- and long-distance runners' endurance capabilities.
- Current CV models, including the exponential one, do not provide precise performance predictions for training guidance.
Related Concept Videos
Run Charts
Interpreting Run Charts
Distance Problem
Prediction Intervals
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
The...
Wald-Wolfowitz Runs Test I
The test works...