REF-237 · MacIntosh et al., 2000 · Medicine & Science in Sports & Exercise, Vol. · Paper · PubMed

Cadence, power, and muscle activation in cycle ergometry

Blimited evidenceMeasured on a mixed groupWritten before October 2026, being re-checked

Participants

Sample size and training status are not given in the abstract (full text; not checked against the abstract). Surface EMG over seven muscles considered important in cycling.

Protocol

Systematic matrix: 4 power outputs (100, 200, 300, 400 W) × 5 cadences (50, 60, 80, 100, 120 rpm). These power outputs represented efforts of up to 32% of the subjects' peak power output. EMG amplitude averaged across the seven muscles for each combination; "optimal cadence" = the cadence with the lowest EMG amplitude at a given power.

Key Results

Power Output	Optimal Cadence (lowest EMG)
100 W	57 ± 3.1 rpm
200 W	70 ± 3.7 rpm
300 W	86 ± 7.6 rpm
400 W	99 ± 4.0 rpm

EMG amplitude rose proportionally with power at every cadence; a second-order polynomial fitted the EMG:cadence relation very well at each power (r² = 0.87-0.996)

Critical finding: at a given power, muscle activation varies substantially with cadence, and the minimum shifts to a higher cadence as power increases

Mechanism

Based on the resistance-rpm relationship of cycling, which resembles the muscle force-velocity relationship (the authors' hypothesis): at low cadence, higher pedal force per revolution → more motor-unit recruitment; at high cadence, contraction velocity raises the activation cost. The optimum balances the two, and the authors note implications for the sense of effort and for the preference for higher cadences as power rises.

Practical Application

Across 100-400 W and 50-120 rpm, the cadence with the least muscle activation (seven-muscle surface EMG) rose with power: 57 rpm at 100 W, 70 at 200 W, 86 at 300 W, 99 at 400 W. Fixed watts do not fix the neuromuscular demand — the same power at a different cadence recruits muscle differently, which is why effort feels different and why riders spin faster as power goes up.

ERG mode implication: ERG holds watts, not muscle activation. Inference, not tested: the same 250 W at 75 vs 95 rpm is a different motor-unit recruitment pattern and a different training stimulus even though the trainer reports identical power.

Interval design: choose cadence deliberately in ERG intervals — low cadence for a force/recruitment stimulus, high cadence for lower per-stroke force at the same power. Not from this paper; the base's own practice.

Zone 2 cadence: by interpolation of the table (arithmetic on the abstract's values), a Zone 2 power around 200 W sits near a 70 rpm EMG minimum — lower than most riders' self-selected cadence; the EMG minimum is not the same as the metabolically or preferentially optimal cadence (see REF-238).

Limitations

Published 2000 — older EMG methodology. Surface EMG captures gross activation, not individual motor-unit behaviour. No metabolic measures (VO2, lactate) at each condition in the abstract. Sample size and training status not reported in the abstract.

How this entry was checked and changed

Why this level

regraded 2026-09-27 from B+: cross-sectional measurement study, sample size and training status not stated in the abstract

Content correction 2026-09-27

Participants "trained cyclists" and Tier 2-3 -> the abstract gives no sample size or training status; now stated as unknown. Level B+ -> B (small within-subject measurement study).

Added the abstract's "up to 32% of peak power" and the polynomial fit (r² 0.87-0.996); the 250 W example and the Zone 2 "~65-80 rpm" figure -> labelled inference / arithmetic on the table (200 W → 70 rpm).

Cross-reference REF-014 (duplicate of REF-084) removed.