Cycling is dictated by the rigid and repetitive motion of turning the cranks, making an optimal interface between the body and bike crucial. When pedaling over 5000 times every hour, it truly becomes critical to understand as much as we can about the biomechanics of the pedaling motion. This is important not just for optimizing performance, but also to minimize the risk for injury.
We’ve said it before, but nothing beats a personalized bike fit by a professional who knows their stuff. However, even with an optimized bike fit, one hidden elephant in the room is that we are not mirror images of ourselves. Specifically, our right and left sides are not built equally.
First Amongst Equals
Externally, our right and left sides look pretty much alike. However, we are all familiar with the fact that each of us possesses varying degrees of ambidexterity, from our preferred eating and writing hands to whether we are right-handed batters or hockey shooters. Over time, the repetitive use of that one dominant hand or side logically must cause differences in the balance of the skeleton and muscles between the two sides.
Such is amply evident in very side-specific athletes, such as baseball pitchers, where the bone thickness and muscle strength can be dramatically higher in the dominant arm. Such marked side-dominance can eventually emerge in the form of injury through to muscle imbalances. Does the same side-dominance hold true for the lower body and legs, though?
On the one hand, we all have preferred kicking legs for sports such as soccer or for the pushing leg when skateboarding. However, unlike the arms and upper body, the legs are constantly used for keeping us upright and for walking, so it may be the case that, aside from the motor control issues involved in specific tasks or specific anatomical issues, the overall strength capabilities muscle development between the right and left lower body are actually minimal.
Therefore, the question that needs to be asked is: does a significant pedaling asymmetry exist in cycling? And if so, is this an important issue to address in terms of corrective exercises or risk of injury?
Carpes et al. 2007
You should know this is coming by now, but the previous paragraph was an obvious segue to a study in the scientific literature. This 2007 study was performed by a Brazilian/US research group, and introduced the rationale by discussing previous research reporting 35-45% variance in ground reaction forces (how hard the leg pushes off against the ground) in walkers and runners, and the possibility for 5-20% variance while cycling.
Some specific details of the study:
Six competitive but non-elite cycling-trained subjects were tested. Subjects were ~20 years old, 71 kg, relatively non-elite VO2max values of 56 mL/kg/min, but still were capable of an impressive peak power output of 400W.
The low number of subjects is a bit of a concern, as it greatly increases the possibility for one outlier subject affecting overall conclusions. This is compounded somewhat by the lack of information about whether any of the subjects had overt anatomical differences between their left and right sides.
The dominant leg was defined as the preferred kicking leg. One additional data I would have really liked to have seen is whether the subjects had a significant difference in strength between the dominant and non-dominant legs irrespective of cycling. Such a test could have easily been done for a “maximal voluntary contraction” or MVC on a neuromuscular testing apparatus common to many exercise science departments. That would have helped to see whether there is an easy way to detect leg asymmetry, and whether such tests matched cycling-specific findings.
Subjects were tested over the course of a 40 km time trial performed in the laboratory. Subjects were provided feedback about their speed, cadence, time, distance covered, etc. Importantly, they were not told about the specific purpose of the test being to analyze pedaling asymmetry, which is vital to ensure that subjects didn’t consciously or subconsciously altered the way they would ride.
The TT was performed on a SRM ergometer that could separately analyze torque from the right and left cranks. Cadence was freely chosen by the subjects, and the course was a simulated flat course to remove the confounding effects of climbing.
Subjects adopted their normal racing (no aero bars) position, and the simulated gearing was a 53×15.
One question that the authors faced was how to actually quantify whether a subject were asymmetrical or not, as no previous study has been able to establish an objective number or threshold for this. The authors ultimately settled on a threshold value of 10% difference in torque between the right and left legs for a threshold for significance.
What Does The Data Say?
Mean time for the 40 km TT was 61 min, ranging from just under 60 to just under 64 min.
In keeping with much of the literature in pacing strategy, the subjects adopted a J-shaped power profile over the 4×10 km segments of the time trial. That is, they started with high mean and peak torque values in the first 10 km, gradually decreased over the 2nd and 3rd segments, and then had an “end-spurt” of higher (actually the highest of the 4 segments) mean and peak torque in the final 10 km. this paralleled the rate of oxygen consumption, suggesting that the torque values were not anomalies.
All subjects demonstrated significant asymmetry (>10%) at some point during their TT. On average, segments 2 (13.5%) and 3 (17.3%) had significant asymmetry of >10% while having lower overall mean torques. Interestingly, the higher mean torque values in segments 1 (8.9%) and 4 (0.3%) featured lower and non-significant asymmetry values.
While cadence was freely chosen throughout, no data was provided on what were the average cadences throughout the 4 segments, making it difficult to correlate whether changes in pedaling dynamics were due to cadence changes or actual biomechanics. However, as the gear was “fixed” at 53×15, changes in speed can only occur through changes in cadence, so it is likely that the pattern for cadence would be similar if not identical.
For me, this study is the intriguing start of a potential series of studies examining the actual implications of pedaling asymmetry on cycling performance. One interesting finding is certainly that pedaling asymmetry exists, but at this stage it’s unclear just how much improving any imbalances can play in improving performance or decreasing the risk of injury.
What is interesting here is that, in a time trial setting anyway, lower torque and power outputs resulted in higher asymmetry. If it can be extrapolated to both real-life time trials and also road races, it would suggest that we do spend much of our riding time with the dominant leg performing a much greater percentage of the overall work of pedaling. But when the hammer drops and the torque/power output increases, the imbalances seem to disappear.
Overall, this leads to several applied questions about pedaling asymmetry that are more difficult to answer:
Is the asymmetry during lower intensity cycling a bad thing in the long term? Would having less asymmetry reduce overall stress on the dominant leg, and possibly decrease the metabolic effort of cycling and therefore improve efficiency?
Why does the asymmetry decrease with higher torques and power outputs? Is it because the non-dominant leg becomes more active, or is it because the dominant leg is so “fatigued” by it working harder at lower efforts that it cannot contribute any more when the hammer drops?
Even if it is hard to quantify, I feel that, logically anyway, a more balanced body and pedal stroke can only lead to improved performance and decreased injury. Personally, I have been using PowerCranks or variations since 2005 either continuously or for specific workouts, and I find they have been a useful system and excellent training tool. At the very least, I would encourage you to think about the way you pedal, and not just treat pedaling as a means to an end.
Have fun and ride safe!
Carpes, F.P., M. Rossato, I.E. Faria, and C. Bolli Mota. Bilateral pedaling asymmetry during a simulated 40-km cycling time-trial. Journal of Sports Medicine and Physical Fitness. 47:51-57.
Stephen Cheung is a Canada Research Chair at Brock University, with a research specialization in the effects of thermal stress on human physiology and performance. He can be reached for comments at firstname.lastname@example.org .