REF-744 · Rønnestad BR, Hansen J, Hollan I, Spencer M, Ellefsen S — 2016 · International Journal of Sports Physiology and Performance · Paper
Elite cyclists who stopped strength training for the first 8 weeks of the competition season, after 25 weeks of heavy lifting, lost squat-jump height, peak aerobic power (Wmax) and 30-s Wingate power relative to controls (ES 0.49–0.84)
The gains of a whole preparation period decayed within two months
Protocol
25-week preparatory period: EXP = endurance training + heavy strength training (3 sets × 4–10 RM, 1–2 sessions/week); CON = endurance training only.
8-week competition period: both groups endurance training only — strength cessation for EXP.
Tests at each phase boundary: maximal isometric half-squat force, squat jump, Wmax, power at 4 mmol/L lactate, 30-s Wingate mean power; analysed with magnitude-based inferences.
Key Findings
Preparation period (25 weeks): EXP gained more than CON in maximal isometric half-squat force, squat jump, Wmax, power at 4 mmol/L lactate and 30-s Wingate mean power (ES 0.46–0.74).
Cessation period (8 weeks in season): EXP showed a reduction in squat jump, Wmax and Wingate mean power compared with CON (ES 0.49–0.84).
The decline appeared within a normal racing block — 8 weeks — not after an off-season away from the gym; the authors call it a rapid decline of adaptations on termination of strength training.
Practical Application
Keep some heavy lower-body work through the racing season rather than stopping (not from this paper; the base's own practice: about one session every 7–10 days). 1–2 sessions a week (3 × 4-10 RM) was the dose that built the gains here, and stopping for two months in season measurably reduced peak aerobic and sprint power — the currency of starts, bursts and finishes. Not from this paper; the base's own practice: place it 48 h or more from the key race.
The failure mode is cessation, not under-dosing: reduce, do not eliminate (the same logic as REF-217 for deloads).
For disciplines decided by repeated short high-force efforts — cyclocross and XCO (REF-741, REF-742), criteriums — this is the most consequential place to lose eight weeks.
Limitations
n = 7 per group, elite men; magnitude-based inferences rather than null-hypothesis tests, so the effect sizes carry wide uncertainty.
8-week window: whether the decline plateaus or continues over a longer season is not shown.
Wmax and Wingate are laboratory measures; race performance was not the endpoint. The size of the loss in amateurs who lifted less to begin with is unknown.
How this entry was checked and changed
Content correction 2026-09-27
The in-season maintenance dose ("one session every 7–10 days") and the 48-h spacing were written as findings → they are not in the abstract (the study tested 1–2 sessions/week, then complete cessation) and are now labelled as the base's own practice.