The intensity rather than the quantity of prior work determines the subsequent downward shift in the power duration relationship in professional cyclists
Participants
n=14 professional cyclists
Height 176.3 ± 7 cm, body mass 67.3 ± 5 kg, critical power 5.3 ± 0.21 W/kg
Protocol (from the abstract)
Power profile test = a 15-s, a 3-min and a 12-min maximal effort on a standardised climb, performed three times:
Fresh: first occasion, no prior work.
LIC (low-intensity continuous): after ~2000 kJ of work at <70% of CP.
HI (high-intensity): after a novel fatiguing protocol of 5 × 8 min at 105-110% of CP. Participants did significantly less total work in HI than in LIC — the HI protocol is shorter and smaller in kJ, not work-matched.
Key Results (from the abstract)
| Effort | HI vs LIC | HI vs Fresh | LIC vs Fresh |
|---|---|---|---|
| 1-s power | significantly lower after HI | — | — |
| 15-s power | significantly lower after HI | — | — |
| 3-min power | — | significantly lower after HI | not significantly lower |
| 12-min power | not different | not different | not different |
| Critical power (from the profile) | not different | not different | not different |
The abstract gives no watt values or percentages for any condition; only the significance pattern above.
Critical Finding: Despite doing LESS total work, the high-intensity protocol (5 × 8 min at 105-110% CP) shifted the short end of the power-duration relationship down more than ~2000 kJ of low-intensity riding did: 1-s and 15-s power were lower after HI than after LIC, and 3-min power fell versus fresh only after HI.
The 12-min power and CP did not change in either condition. Total work alone is therefore not sufficient to describe prior work — its intensity matters, especially for shorter efforts.
Mechanism
The paper establishes the pattern (short efforts suffer after supra-threshold work; CP and 12-min power do not), not its cause.
Inference, not tested here: supra-CP intervals draw on W'/anaerobic capacity and fast-twitch fibres, which is where sprint and 15-s power live; long sub-threshold riding spares them. The abstract reports no PCr, glycogen, neural or lactate measurements.
Application
In 14 professional cyclists, 5 × 8 min at 105-110% of CP lowered subsequent 1-s and 15-s power more than ~2000 kJ of riding below 70% CP did — despite being less total work — and only the high-intensity protocol lowered 3-min power versus fresh; 12-min power and CP were unchanged in both conditions (within-subject study, Spragg 2024).
Prior kilojoules alone do not predict how much a rider has left: an hour with hard intervals costs more sprint and 3-min power than a long easy day. Read "what was done above CP", not just the kJ total.
CP and 12-min power were robust to both protocols here, so a durability problem in the sprint and VO2max range does not imply the threshold engine has faded — the two ends of the curve fail separately.
Inference, not tested: place sprint-quality work before, not after, supra-threshold intervals within a session or a race plan, and expect the finishing sprint to be blunted by hard climbing earlier in the race more than by distance.
Not from this paper; the base's own practice: recovery spacing between supra-threshold sessions (roughly 48 h) is a house rule; this study did not measure recovery time and reports no recovery-hour figures.
Limitations
From the design as the abstract reports it: within-subject study in 14 professional cyclists; the fresh test was always the first occasion, and the abstract does not describe randomising or counterbalancing the two fatigued conditions.
The two fatiguing protocols were not work-matched (significantly less work in HI than in LIC), so intensity and total work are not separated by matching.
Efforts were done on a standardised climb; the abstract reports the significance pattern only, with no power values or effect sizes.
How this entry was checked and changed
Content correction 2026-09-27
"HI total work ~2000 kJ, SAME as LI" -> abstract: participants did significantly LESS work in HI than in LIC; the protocols were not work-matched.
Power profile "10 s, 15 s, 30 s, 1 min, 5 min, 20 min" -> 15 s, 3 min, 12 min on a standardised climb. Recovery periods (30 min / 60 min), "~5.5 W/kg × 180 min", "~400 W per 60 kg", "VO2max ~70" deleted: not in abstract.
Results table (2100/2050/1920 W, −2.4% … −16.8%, 5-min −6.5%, 20-min −2.6%, "5-6× larger") deleted: invented; abstract reports significance pattern only and says 12-min power and CP did NOT differ.
Mechanism/molecular sections (PCr, glycogen, CNS 15-20%, 48-72 h, 24-48 h) deleted or relabelled; "RCT, Level A-" -> within-subject n=14, Level B; Sport "cycling, sprint" -> cycling.
Cross-reference label: REF-141 is Jones & Kirby's resilience paper, not Laursen 2007.
Cross-references to duplicate stubs repointed to their canonical entries (REF-162->REF-306); the stubs were emptied on 2026-09-27.
Cross-reference labels pointed at the wrong entry or named the wrong author (checked against each cited entry's byline); repointed to the entry the label describes, or removed where the base holds no such paper.
Cross-reference descriptions checked against the cited entries' current content; wrong years, wrong entries and claims the cited entry does not make were corrected.
2026-09-28: added Limitations from the abstract (Crossref/OpenAlex abstract for the DOI; no PMID).
2026-09-28: journal issue/pages corrected from the DOI's Crossref record (24(8), 1113-1120 -> 24(4), 449-457).