REF-739 · Leirdal & Ettema, 2011 · Medicine & Science in Sports & Exercise, · Paper · PubMed

Evenness of power around the crank

Not how "effectively" force is aimed — was the only predictor of gross efficiency, and it accounted for 56% of the difference between cyclists

Blimited evidenceMeasured on highly trained athletesWritten before October 2026, being re-checked

Protocol

Twenty-one competitive cyclists cycled 10 min at approximately 80% V̇O₂max at a freely chosen cadence. Gross efficiency, force-effectiveness ratio and dead-centre size were computed from oxygen consumption and propulsive force recordings (pedal normal and shear forces).

Descriptives: mean work rate 279 W, mean freely chosen cadence 93.1 rpm, mean gross efficiency 21.7%. Force effectiveness 0.47 (0.79 after correcting for inertial forces); dead-centre size 27.3% (25.7% corrected).

Key Findings

Dead-centre size correlated with gross efficiency at r = 0.75, better than force effectiveness did (r = 0.50).
In multiple regression, dead-centre size was the ONLY significant predictor of gross efficiency (P = 0.001). Force effectiveness was not.
Dead-centre size and force-effectiveness ratio did not correlate with each other at all. They are two different things, and only one of them tracked energy cost.
The authors' conclusion: generating power evenly around the whole pedal revolution "may be an important energy-saving trait".

READ THE DIRECTION CAREFULLY — the name inverts the intuition

"Dead-centre size" here is the minimum pedal work rate as a proportion of the average work rate over the revolution. So a LARGER number means power dipped LESS at the top and bottom — a SMALLER dead spot in plain language. Larger DC size went with higher efficiency.

Anything that says "smaller dead centres are linked to greater efficiency" is using the colloquial sense of the words and the opposite of the paper's parameter. A widely circulated newsletter summary of this study carries exactly that inversion in its headline while stating the correct direction in its body. When citing this entry, say *evenness*, not *dead-centre size*, and the ambiguity disappears.

Practical Application

The coaching cue that matches this evidence is "make the power even", not "pull up on the backstroke". Force effectiveness — the metric behind "pedal in circles" and "pull up" — did not predict efficiency here, and REF-499 records Korff 2007 finding that a deliberate pulling-up technique improved mechanical effectiveness while *reducing* gross efficiency. Two independent lines pointing the same way.
Do not promise an efficiency gain from technique work. This study cannot support it (see Limitations). What it supports is a statement about what distinguishes efficient riders, which is a different sentence and an honest one.
If an athlete wants to work on it anyway, cadence variation is the defensible framing — exposure to a range of pedalling conditions as motor learning, not as a guaranteed efficiency intervention. That is a hypothesis, and the base's cadence entries (REF-084, REF-117, REF-259) are about training effects, not about moving dead-centre size.
Where efficiency actually matters, remember the tier. Gross efficiency separates riders when VO₂max and fractional utilisation are already similar (REF-541 shows the same at world-class level in skiing). For a Tier 1-2 athlete the engine is still the lever.

Limitations

Cross-sectional correlation, single session, n=21. It shows that efficient cyclists pedal more evenly. It does not show that pedalling more evenly *makes* a cyclist efficient, and the causal arrow could run the other way — an efficient muscular system may produce even power as a consequence, not a cause.
r = 0.75 means r² ≈ 0.56, and "explains 56%" is a statement about variance BETWEEN these 21 riders at one moment. It is not 56% of any individual's efficiency, and it is not a ceiling on what training could change.
One intensity, one duration, freely chosen cadence. Nothing here describes what happens at race intensity, in the fourth hour, out of the saddle, or at an imposed cadence.
Whether dead-centre size is trainable is untested here. Short-term technique interventions have generally failed to raise gross efficiency — that is context from the wider literature and from the summary that prompted this entry, NOT a result of this study.
2011, and 21 riders. A single-cohort correlation of this size is a strong hypothesis generator and a weak basis for prescription.
Requires instrumented pedals to measure at all. Dead-centre size is not something an athlete can see in a normal head unit, so this is a reason to be sceptical of technique claims rather than a metric to chase.
How this entry was checked and changed
Content correction 2026-09-27
Cross-references to duplicate stubs repointed to their canonical entries (REF-014->REF-084); the stubs were emptied on 2026-09-27.