cycling cadence

The Cycling Cadence Debate Renewed, Plus Workouts for a Smooth, Powerful Pedal Stroke

Share This Article

By Chris Carmichael,
Founder & Head Coach of CTS

Early in 2019 there were headlines saying “High Cadence Cycling Offers No Benefits to Amateurs” and asking “Is High Cadence Cycling Actually Slowing You Down?” They – and others – were referring to a Feb 2019 study in the International Journal of Sports Medicine that showed pedaling efficiency and muscle oxygenation decreased when cyclists rode at higher cadences. So, should we go back to mashing the pedals at 60 revolutions per minute?

No, we should not.

The study by Federico Formenti1 and colleagues does indeed show an increased metabolic cost and reduced tissue saturation index (an indicator of skeletal muscle oxygenation) in the vastus lateralis muscle of the thigh. However, the study involved just nine subjects, two of whom were regionally competitive triathletes, six who “regularly engaged in moderate and vigorous exercise”, and one was active occasionally. Researchers had subjects pedal at 40, 50, 60, 70, 80, 90rpm for 4 minute stages, each at ventilatory threshold.

Not surprisingly, the figure above shows pedaling faster at ventilatory threshold led to increases in heart rate, VO2, VCO2, and RPE; and a decrease in peak pedal force. Increasing cadence increased metabolic work to do the same mechanical work, which means a decrease in efficiency. Reduced tissue saturation index also meant muscle oxygenation declined, meaning muscles were using oxygen faster than freshly oxygenated blood could perfuse the muscle, as cadence increased from 40 to 50 to 90rpm.

Cycling media took this information and said amateur cyclists shouldn’t bother with high-cadence cycling. For me, the fact there were 9 subjects means there might be cause to conduct a similar study with a meaningful sample size of experienced cyclists, even if they’re not highly fit. It also indicates cadence is trainable. If most subjects don’t ride bicycles regularly, I would expect them to have trouble maintaining a 90rpm cadence for four minutes at their ventilatory threshold.

However, there’s a large body of research2 that shows trained cyclists freely select higher cadences (80-100rpm), even though the metabolic cost is higher. Although oxygen demand is lower at lower cadences (55-65rpm), research in trained cyclists (not just elite) shows their power outputs, time trial performances, and race performances are improved at the higher cadence range.

It is important to realize efficiency isn’t everything; there are advantages to periods of inefficient pedaling, like stomping on the pedals to accelerate from a slow corner or spinning a light gear to surge on a climb. You can create high or low power at high or low cadence, and there’s a time and place for each.

In addition to the gradual adaptation to more efficient pedaling at higher rpms, I recommend proactively including the following cadence-oriented drills into training.


It’s exactly what it sounds like, the caveat is to use a light gear to keep resistance low, but not so light there’s no resistance at all. Keep your upper body calm and pedal as fast as you can without bouncing in the saddle. Start with three to five 1-minute efforts separated by two minutes of moderate-cadence recovery. Progress to 3-minute efforts with five minutes recovery between them. Your heart rate will increase (as demonstrated in the study above), but RPE should stay at about 5-6 out of 10.

Tip: You already know how to push down on the pedals. To smooth out your stroke so you can move your feet faster without bouncing, think about kicking your foot forward over the top of the stroke and scraping it back through the bottom. As you improve these areas you can slightly extend the effective portion of your pedal stroke at all power levels.

Seated High Speed Sprints

Find a downhill section of road where you can sprint safely. In a medium gear (big chainring, middle of the cassette), get rolling to about 15-20mph. You want to be going fast enough to pedal about 90rpm with only light resistance. Stay seated and sprint for 20 seconds. If you spin out the gear, start in a bigger one the next time. Rest five minutes between sprints and complete 5-8.

Tip: These seated accelerations work in races, too. Standing up and swinging your bike around is a dead giveaway you’re attacking. If you can powerfully rev up your cadence while keeping your upper body calm, you may ride people off your wheel before they realize what’s going on.


At the opposite end of the cadence spectrum, accelerating against a heavy resistance helps reinforce the mechanics – particularly the kick over the top and scrape through the bottom – that help you smooth out your high cadence pedaling. They’re also great for developing high power to accelerate from slow speeds. Roll to a near standstill at 2-3 mph in a big gear. Focus on keeping your core stable and preventing your back from rounding, and accelerate as powerfully as you can without changing gear. Continue for 10 seconds or until the gear is spun out. Rest five minutes between PowerStarts and complete 5-8.

Perhaps the biggest takeaway from study above is that faster isn’t always better when it comes to cadence. The goal of cadence drills isn’t to have the fastest cadence possible; it’s to increase the range of cadences you can use effectively to produce power in different situations.


  1. Formenti, Federico, et al. “The Effect of Pedaling Cadence on Skeletal Muscle Oxygenation During Cycling at Moderate Exercise Intensity.” International Journal of Sports Medicine, 2019.
  2. Vercruyssen, Fabrice, and Jeanick Brisswalter. “Which Factors Determine the Freely Chosen Cadence during Submaximal Cycling?” Journal of Science and Medicine in Sport, vol. 13, no. 2, 2010, pp. 225–231., doi:10.1016/j.jsams.2008.12.631.

Share This Article

Comments 10

  1. The match sprint game has changed considerably based on similar studies and data. It used to be back in the 60-early 80’s that cadence was king. Track sprinters would ride 94 inch gears and 150 plus rpm. What has been found since is that riders many times put out max watts at a much lower cadence. Why have a gear that allows pedaling at 160 rpm if maximum watts happens at 140 rpm. Now, track sprinters are riding gears as high as 104 or more and kilo/200 meter times are dropping.

  2. To try to smooth out my pedal stroke at a high cadence while n a trainer, I had success trying to push my shins forward on the down stroke.

  3. From personal experience I agree that higher cadence uses more oxygen, and heart rate increases, even at the same power output. What isn’t mentioned is muscle fatigue. At lower cadences I find my quads get tired more quickly. So I agree with Chris that a range of cadences should be used to improve effectiveness in different situations. For example, if I’m racing a 20K time trial then I’m going to target a cadence under 90 rpms so as to maximize both my quad strength and cardio. I don’t have to worry about muscle fatigue for such a short effort. However, if I’m racing the Leadville 100MTB then I going to use a cadence over 90 rpms (well, maybe not on Powerline cuz I don’t have gears that go that low!). The higher cadence for long events helps to limit quad fatigue over long hours of pedaling, and despite the slightly higher heart rate it’s not an issue since i’m deliberately targeting lower HR zones because the event is so long, i.e. proper pacing. I would think another consideration is muscle strength. I have strong quads from years of weightlifting, so it’s easier for me to mash the pedals with lower cadences, e.g. 75-85 rpms, which helps to keep my HR from going into to the red zone. All this being said, overall I’m almost always riding at a cadence of 90-95. Great article!

    1. Exactly so and well said.

      I’d like to mention that after 20+ years of experimenting with FastPedal drills, I’ve found the most effective thing is to do one long interval of up to 45′, no breaks. I use a gear which enabled me to spin 115-120 when I was in shape and still stay in HR zone 2. My max cadence w/o bouncing has become somewhere over 150. I do a long interval starting at 15′ once a week, staying in HR zone 2, and over the weeks, interval length, cadence, and therefore power would gradually come up, i.e. oxygen cost goes down. I start these drills in October every year and continue into summer, northern hemisphere. Now 74, I can only hold 112 anymore and in a lower gear. 117 was in my 60s. Best advice I had was to pedal with the shoe uppers, leaving a cushion of air between my foot and insole.

      Cadence is definitely trainable – I also do low cadence drills.

  4. “The goal of cadence drills isn’t to have the fastest cadence possible; it’s to increase the range of cadences you can use effectively to produce power in different situations.” Well put!

  5. Experiments like the one cited make me hang my head in despair.

    How do people get paid to do rubbish science like this? And what sort of journal publishes the results?

    1. How do you come to the conclusion that the science is rubbish? Just because you don’t agree with the results doesn’t mean they are wrong.

      The problem here appears to be the conclusions drawn by media. According to what Carmichael sites, the result of the study was ‘pedaling efficiency and muscle oxygenation decreased when cyclists rode at higher cadences’ and that appears to be supported by the data in the chart. The headline “High Cadence Cycling Offers No Benefits to Amateurs” appears to be a conclusion drawn by the media source, not by the scientist. Don’t blame the scientist for the ignorance or ineptness of the media.

      One piece of information that seems to be missing from the analysis is power output. It would be interesting to see how the power changes at the different cadences.

      The other thing to keep in mind is that a 4 minute span doesn’t represent the results you would have from a 30 minute or 2 hour ride. It’s important to understand the results of the data and not to make assumptions about what it means.

        1. One research subject can be a very meaningful sample. People get confused about the relationship between sample size and statistics. If you’re trying to see who will win a tight election you need a large sample size. You can see very conclusive trends in one single person though and if you confirm similar conclusive trends in 9 fairly random people, it’s strong evidence that it’s likely that way in most people, especially if we’re talking about physiology, and not sociology. Anyway, if you don’t believe tis conclusion, you haven’t cycled much and haven’t read much. There’s no doubt that lower cadence is more efficient, none. That doesn’t mean you won’t tire out fast though. Energy expenditure isn’t the only factor in fatigue.

Leave a Reply

Your email address will not be published. Required fields are marked *