REF-848 · Cangley et al., 2010 · International Journal of Sports Medicine, Vol. · Paper · PubMed

The Effect of Variable Gradients on Pacing in Cycling Time-Trials

B+moderate evidenceMeasured on trained athletesWritten before October 2026, being re-checked

Twenty experienced cyclists rode a 4-km undulating time-trial course four times: twice holding a constant power (average 253 W) and twice varying power with the gradient (average 260 W). With times normalised to 255 W, the constant-power trials took 411±31.1 s and the variable-power trials 399±29.5 s, a 12±8 s (2.9%) saving (P<0.001).

Riders found it hard to hold constant power on the undulating course, which reduced the time the constant strategy could gain. The authors conclude that a variable power strategy can improve time-trial performance where the gradient is not constant.

Participants

n=20 experienced cyclists (abstract: "20 experienced cyclists"). Sex, age and fitness values are not given in the abstract.

Protocol

Field time trial over a 4 km undulating course.
Target: a constant power of 255 W vs a variable power that averaged 255 W, varied in response to gradient.
Four trials each: 2 at constant power (achieved average 253 W) and 2 at variable power (achieved average 260 W).
Completion time normalised to 255 W to compare the strategies.

Key Results

StrategyAchieved average powerTime normalised to 255 W
Constant power253 W411±31.1 s
Variable power (with gradient)260 W399±29.5 s
Variable power reduced completion time by 12±8 s (2.9%), P<0.001.
Participants had difficulty holding constant power over the undulating course, which reduced their time gain.

Practical Application

In 20 experienced cyclists on an undulating 4-km field course, varying power with the gradient at the same average (normalised to 255 W) was 12±8 s (2.9%) faster than trying to hold constant power (P<0.001). On a rolling time trial, push on the rises and ease on the dips: this is the field evidence that the model predictions hold on a real road.
Holding constant power on undulating roads is itself hard; riders who try tend to miss it, so plan where on the course the harder sections go rather than aiming at one number.
Inference, not tested here: the saving was measured over about seven minutes at a modest power; on a longer rolling course the direction should hold, but the size cannot be read from this study, and larger swings cost more physiologically (REF-252).
Not from this paper; the base's own practice: rehearse the variable plan in training on the course or a similar road before racing it.

Limitations

Short course (4 km) and a modest fixed power; the result may not scale to longer time trials or harder efforts.
The achieved averages differed (253 vs 260 W), so times were normalised to 255 W; the normalisation method is not described in the abstract.
Trial order, conditions (wind) and participant characteristics are not in the abstract.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

within-subject comparison, n=20; trial order not described in the abstract