Critical Power: An Important Fatigue Threshold in Exercise Physiology
Key Findings (concepts)
The hyperbolic power-duration relationship holds robustly and is highly conserved: it appears across species, exercise types, and single muscles or muscle groups.
For modalities such as cycling it resolves to two parameters: Critical Power (CP) — the power asymptote — and W' — the work doable above CP. Together they predict the tolerable duration of exercise above CP.
CP integrates the respiratory, metabolic and contractile physiological profiles into one framework: setting intensity relative to CP unifies the systemic and intramuscular responses, whereas calibrating to metabolically distant parameters such as the lactate threshold or VO2max does not.
Above CP: W' is expended, VO2max is attained and intolerance follows — duration is predictable from CP and W'.
Below CP: the physiological responses can be stabilised. CP is therefore a "fatigue threshold": the boundary between intensity domains where responses can and cannot stabilise (heavy vs severe).
The concept gives insight into the principal loci of fatigue (central vs peripheral) at different intensities, and into cardiovascular and metabolic control and its modulation by O2 delivery.
Practical Application
Conceptual review (Poole, Burnley, Vanhatalo, Rossiter & Jones 2016): critical power is the asymptote of the hyperbolic power-duration curve; together with W' it predicts how long exercise above CP can be tolerated, and it marks the boundary between intensities where physiology stabilises and where it does not — so prescribe intensity relative to CP rather than to lactate threshold or VO2max. CP is the anchor for the heavy/severe boundary; how far above it you work, and for how long, is governed by W'.
Zone anchoring: use CP as the heavy/severe boundary for interval prescription; VO2max-targeted intervals belong above CP (severe domain), where W' is the budget; sustained "threshold" work belongs at or just below CP, where a steady state is possible.
Individualization: CP and its relation to LT and VO2max differ between athletes, so CP should be measured, not assumed from FTP or a percentage of VO2max (REF-038).
Not from this paper: typical CP/LT percentages of VO2max by training status, specific interval prescriptions (e.g. 30/15 s at 120% CP, 4×4 min at 105-110% CP), and re-test intervals — these were house figures with no source in this review.
Limitations
Conceptual review — no new data; the two-parameter model is stated for modalities "such as cycling", and running / rowing / cross-country skiing CP models carry more measurement noise
W' recovery below CP is not instantaneous and varies between athletes; the review does not give recovery time constants (τW') — earlier house figures (90-600 s) removed
CP shifts with heat, hypoxia and glycogen state, so protocol standardisation matters; magnitudes are not in this review
Some athletes sustain > CP longer than the model predicts (glycolytic-aerobic interaction)
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: conceptual narrative review from the leading group
Content correction 2026-09-27
Tier "mixed→Tier 4-5" -> N/A: conceptual review with no study population.
Key Figures table (LT/GET 50-65% vs 70-80% VO2max, CP 70-80% vs 80-90%, 20-40 min at CP), τW' 90-600 s, CP shift 5-10%, interval prescriptions (120% CP, 105-110% CP), re-test every 6-8 weeks -> not in this paper; removed or marked "Not from this paper".
Key concepts and lead bullet rewritten from the abstract (hyperbolic power-duration relation, CP + W' predict tolerable duration, CP as fatigue threshold, prescribe relative to CP).
Cross-reference descriptions checked against the cited entries' current content; wrong years, wrong entries and claims the cited entry does not make were corrected.