REF-262 · Emanuele, Horn & Denoth, 2012 · International Journal of Sports Physiology and Performance, 2012; · Paper
The relationship between freely chosen cadence and optimal cadence in cycling
Participants
n=7 experienced cyclists.
Protocol
Cycling on a treadmill under two conditions:
LGDP: level ground, dropped handlebar position
UHUP: uphill, upright position
Measurements: FCC (freely chosen cadence), Copt (the cadence at which blood lactate at a constant power output is minimised), Pmax (maximal external power output).
Key Results
| Condition | FCC (rpm) | Copt (rpm) | FCC vs Copt |
|---|---|---|---|
| Level ground, drops (LGDP) | 89.3 ± 10.6 | 87.7 ± 10.9 | not statistically different |
| Uphill, upright (UHUP) | 82.1 ± 11.1 | 81.5 ± 9.8 | not statistically different |
Most but not all experienced cyclists chose a cadence close to Copt, independent of the condition
Uphill/upright: FCC and Copt both significantly lower (FCC about 7 rpm lower than on level ground) and Pmax significantly higher (+2.0 ± 2.1%)
Authors' conclusion: most experienced cyclists choose a cadence near Copt to minimise peripheral fatigue at a given power, and a lower cadence with a more upright position is advantageous uphill
Mechanism
Inference, not tested: the authors frame Copt as the cadence minimising peripheral fatigue (lowest lactate at the same power); the lower uphill optimum and higher Pmax when upright are consistent with a higher-torque, lower-velocity muscle action on a climb. The abstract does not test ventilation, hip angle or gravity mechanics — those explanations are inference.
The 7-rpm drop is consistent with REF-258's intensity- and velocity-dependent optimum, but that link is the base's, not the paper's.
Practical Application
n=7 experienced cyclists on a treadmill: freely chosen cadence was 89.3 ± 10.6 rpm on level ground in the drops and 82.1 ± 11.1 rpm uphill in an upright position (about 7 rpm lower), and in both conditions it was not different from the cadence that minimised blood lactate at the same power (87.7 and 81.5 rpm); maximal power was 2.0 ± 2.1% higher uphill/upright — most riders self-select close to their metabolic optimum, and that optimum drops on a climb.
Do not prescribe a flat-road cadence on climbs: a drop of several rpm uphill is the optimum moving, not laziness, and the upright position on the climb is part of the same adjustment.
Trust an experienced rider's self-selected cadence, but not blindly — "most but not all" matched their optimum; correct it only for a specific reason (a neuromuscular target, a fatigue-related drift — REF-261).
Inference, not tested: on a trainer, ERG mode holds power but not cadence — lower the cadence deliberately on simulated climbs; SIM mode produces the higher-torque, lower-cadence stimulus on its own.
n=7, one gradient, smooth belt: the direction is robust, the exact rpm values are not a target.
Limitations
Only 7 cyclists — very small sample
Treadmill cycling — not real road/gravel terrain
Single gradient tested — no range of climb steepness
No gravel surface simulation — only smooth treadmill belt
How this entry was checked and changed
Content correction 2026-09-27
Added the abstract's Copt values (87.7 ± 10.9 and 81.5 ± 9.8 rpm) and the Pmax result (+2.0 ± 2.1% uphill/upright), which the entry had omitted.
"Experienced cyclists chose cadence near Copt regardless of condition" -> abstract: "most but not all"; lead and advice adjusted.
Mechanism (gravity/torque, ventilation, hip angle) and the ERG/SIM advice labelled as inference; cross-reference to REF-260 reworded to what that entry now says (gear ratio, not fixed-power cadence).