PezCycling News - What's Cool In Pro Cycling : Optimal Cadence Revisited

Now On Pez
Daily Distractions!
Tech N Spec
Lake Tahoe - USA - wielrennen - cycling - radsport - cyclisme - Brand Lucinda (Netherlands / Rabobank Liv Women Cycling Team) Vos Marianne (Netherlands / Rabobank Liv Women Cycling Team) pictured during  training of the the Amgen Breakaway From Heart Disease Women’s Race Lake Tahoe - photo Anton Vos/Cor Vos © 2016
Pez Videos
<iframe id="ytplayer" type="text/html" width="280" height="150" src="" frameborder="0"></iframe>
Readers' Rigs
Geraardsbergen - Belgium - wielrennen - cycling - radsport - cyclisme - Niki Terpstra (Netherlands / Team Etixx - Quick Step) pictured during  Eneco Tour stage -7 - UCI World Tour) from Bornem to Geraardsbergen - photo Tim van Wichelen/Cor Vos © 2016
Optimal Cadence Revisited
Thanks to Lance, we have seen a “revolution” in prevailing thought about optimal cadence in cycling time trials from a rouleur rolling big gears to spinning high cadences. But should we all aim to adopt the same high rpm TT style? What does science say about cadence, efficiency, and performance?

This is certainly not the first Toolbox article where I have tried to address the issue of optimal cadence or pedaling efficiency. The topic remains one of high scientific interest judging by the number of studies being published every year. In this article, I’ve updated a 2005 Toolbox article on optimal cadence by adding a new study by a Spanish lab on elite amateur cyclists. Make sure you check out the links to other articles and our series on the PowerCranks down at the bottom!

Optimal cadence may mean a lot of things to a lot of people depending on the context. Certainly there are multiple reasons underlying cadence selection, ranging from the requirements of the event or situation, maximizing power production and/or efficiency, minimizing fatigue, or even comfort. They are all valid reasons, and there is certainly no optimal cadence for all conditions in all cycling events. For example, the acceleration required by track sprinters lead to super-high rpms, whereas a “comfortable” cadence for long endurance rides seem to lead to a self-selected cadence of 90-105 rpm on the flats for most cyclists.

In time trials, the prevailing wisdom has always been to roll bigger gears at lower cadences of 80-90 rpm until Armstrong came along. And hands up who amongst us has not tried to replicate his high cadence time trial (110+) and climbing (90+) style? That’s what I thought, and we’re not alone. Most of the pros have also attempted to replicate this high rpm style, with varying degrees of success. The question remains – is Lance truly a unique specimen, or is there scientific basis for all of us to aim for shifting TT cadence upwards?

Bringing the Road into the Lab
Science and lessons learned from the pros go hand in hand, so it’s no surprise that there’s been a renewed interest in optimal time trial cadence in the past six years. The prevailing wisdom from older scientific studies was that the optimal cadence, in terms of efficiency (mechanical power output as a percentage of total energy turnover) was shaped like an inverted U, with the peak cadence at approximately 50-60 rpm, FAR lower than the typical cadences employed by most cyclists.

Last year, I wrote about two articles (2004 and 2005) from the Norwegian group of Foss and Hallen (1, 2), which shed a few interesting ideas into the mix. Specifically, they postulate that the reasons for the findings from the older studies was that their tests were: 1) short (< 10 min), 2) at relatively low power outputs (~125 W), and were constant load tests (i.e., ride at the same intensity throughout) as opposed to being a true time trial where performance (i.e., highest power output achievable) was the goal.

Foss and Hallen 2004
Foss and Hallen’s 2004 paper attempted to address some of these limitations by having elite cyclists perform an incremental test to exhaustion (4-7 min) with cadences of 60, 80, 100, and 120 rpm, and found that performance was highest at 80 rpm compared to the other cadences. This elevates optimal cadence closer to the typical one employed by cyclists, and brings about their interesting idea that the optimal cadence increases with increasing power outputs.

Foss and Hallen 2005
In their 2005 study, Foss and Hallen attempted to further address some of the problems of prior studies, namely the lack of a true performance measure and also the short effort times. Again using national level Swedish and Norwegian cyclists, the protocol involved trying to complete a set amount of work in as short a time as possible, with the amount of work individually adjusted to approximate a 30 min time trial and the subject able to self-adjust the power output. Again, cadence was fixed at 60, 80, 100, and 120 rpm, with an additional trial where subjects could freely choose cadence.

The primary results of the 2005 study support their 2004 findings, in that the optimal cadence in terms of efficiency was again at 80 rpm, though there were only minor decrements at 100 rpm and indeed a slightly higher energy turnover (i.e., performance) at 100 rpm. Freely chosen cadence averaged about 90 rpm, with the results pretty much similar to 80 and 100 rpm. So it appears that, at more realistic power outputs and with longer efforts typical of time trials, efficiency and performance can be achieved anywhere within the 80 – 100 rpm range.

Mora-Rodriguez et al. 2006
In a study published this month, a Spanish group (3) pretty much replicated the fundamental design of Foss and Hallen 2004. 9 elite amateur cyclists (racing as elite amateurs for at least 3 years) performed an increasing load test to exhaustion, beginning at 175 W and then increased 25 W every 3 min. This is similar to what you might do if you were tested for your lactate threshold in an exercise science lab.

Subjects repeated this test at a constant cadence of 80, 100, and 120 rpm, the same ones used by Foss and Hallen. Subjects also performed an initial test at their freely chosen cadence, which ended up averaging 89 rpm.

One of the primary results of this study demonstrated that gross efficiency (a basic measure of the total energy required) at 275 W was similar across the three cadences. In other words, it took the same amount of energy to pedal at 275 W at 80, 100, or 120 rpm. Heart rate was higher with increasing cadence though.

On the other hand, what did become impaired at 120 rpm compared with 80 and 100 rpm were the peak wattage at the point of exhaustion, the wattage at ventilatory threshold (where your breathing becomes laboured and inefficient), and a trend towards a lower wattage at lactate threshold.

So while cadence didn’t seem to impair performance at a high but submaximal wattage, it did seem to cause earlier “fatigue” when at maximal sustained effort. The authors hypothesized that this may be due to: 1) fatigue and greater energy required simply to move the legs at the higher speed (remember, while you may be pushing against less resistance with each pedal stroke, it costs more energy to move 120 times compared to 100 times even against no resistance), and 2) the higher speed of movement caused greater recruitment of fast-twitch muscle fibres, which are less efficient aerobically and produce more lactic acid.

Hitting the Road Again
So then should we all be trying to emulate Lance and continue to strive for revving up our rpms? Several comments and caveats:

• If the trend observed by Foss and Hallen are correct, there seems to be a continuing upward trend of optimal cadence with increasing power outputs, and thus Lance’s incredibly high power outputs may also explain his higher cadence. Regardless, unless we can also generate these incredibly high power outputs, it may be that a higher cadence may not be necessary.

• Your friend might drive you crazy because he always seem to walk faster than you, but each individual has a walking stride length and frequency that is the most optimal and efficient for them, based on things like leg length, body size/height, etc. Indeed, it takes more energy to walk slower or faster than your ideal speed, because you either have to slow your body or speed it up past its natural preference. Similarly, optimal cadence is also highly individual based on factors such as individual preference, perception of effort, and crank length. In the end, it appears that you can be fairly similar in efficiency and performance between the range of 80 – 100 rpm, so a higher cadence may not be as big a deal as it seems.

• If you do attempt to change your typical cadence, it will require a long-term commitment and cannot be a one-month or even possibly a one-year affair. Regardless of how efficient your current cadence is, your body’s system: neural, muscular, cardiovascular metabolic, etc. has become adapted to it through continous training. Therefore, it takes an enormous amount of time for all of your body’s system to re-adapt and become as efficient as possible at a new typical cadence. This is the fundamental difficulty of research into this area!

Check out these tasty links:
Josh Horowitz’s Season 3 intro to PowerCranks, with links to all our other PowerCrank articles

We examine Lance’s improvements in pedaling efficiency from 1993 through his first Tour victory

We analyze another Spanish study that finds pedaling efficiency much more important than what your VO2max is

An article by Bruce Hendler on pedaling efficiency and training drills

1. Foss O and Hallen J. Cadence and performance in elite cyclists. Eur J Appl Physiol 93: 453-462, 2005.
2. Foss O and Hallen J. The most economical cadence increases with increasing workload. Eur J Appl Physiol 92: 443-451, 2004.
3. Mora-Rodriguez R and Aguado-Jimenez R. Performance at high pedaling cadences in well-trained cyclists. Med Sci Sports Exerc 38: 953-957, 2006.

About Stephen:
Stephen Cheung is an Associate Professor of Kinesiology at Dalhousie University, with a research specialization in the effects of thermal stress on human physiology and performance. His typical cadence dropped to <70 rpm in his initial riding with PowerCranks but is now happily spinning at a preferred cadence of 90+. He can be reached for comments at


Pez Comments

Related Stories