Current knowledge is that riding at a higher cadence is good for long-term aerobic efforts. Does that mean that sprinting at lower cadences would be better? I have certainly experienced myself that I seem to be able to sprint faster when pushing a fairly tough gear. It feels like I can put more power out and I am exerting myself more.

Or is it that I'm just bad at riding at high cadences and I should learn how to do so? Professional sprints all seem to be at high cadences.

  • 1
    How long is your sprint? A short few-seconds to get on someone's wheel is different to a leadout minute at the end of a race, vs an "effort" of several minutes.
    – Criggie
    Commented Apr 25, 2021 at 1:33
  • 3
    Not an easy answer - everyone has different physiology, training makes a big difference, and perceived vs actual effort can vary considerably. Best way to know is a power meter to remove the reliance on 'feel'. Higher cadence also reduces chance of injury.
    – mattnz
    Commented Apr 25, 2021 at 1:38
  • 1
    @Criggie We would be talking about an all-out sprint for 5-15s, such as what you see at the end of a crit race or something. For several minutes’ worth I’d choose to spin faster.
    – MaplePanda
    Commented Apr 25, 2021 at 16:13
  • 3
    The reason why this is a difficult question to answer is because cadence is a result of your sprint, not a determinant of it. That is, cadence is a dependent or endogenous variable, not an independent or exogenous one. Because of that, sprint cadence depends on what your situation was before you launched and how long you must sustain it. That said, under ideal conditions, max power is attained near half of max cadence and half of max pedal force -- but in a sprint you may not have that choice.
    – R. Chung
    Commented Apr 25, 2021 at 17:31

3 Answers 3


I'll add a few things in addition to @Criggie's answer:

  1. Power meter. It was mentioned in the comments, but you really don't know if you produce more power pushing a bigger gear with more force at lower cadence than you produce with a smaller gear at a higher cadence unless you can measure it.

  2. Expand your usable torque range. Everyone knows about things like anaerobic intervals of 3-7 minutes each to train your ability to ride anaerobically for several minutes or 10-to-30-second all-out sprint intervals. You can do the same for cadence, too. For example, 3x7x3 intervals at 100-110 rpm where you do 3 sets of 7 minutes of riding at 100-110 rpm with three minutes of rest between intervals. Or a 20-minute interval at low-to-moderate power where every 30 seconds you spin out, maxing your cadence to the fastest you can do seated. If you've never worked on expanding your effective cadence range, you'll be surprised at how painful those types of rides can be even though your power level or heart rate never really gets too high. If you're doing power-based training, the power level on these rides will be pretty low. But you'll hurt anyway. The high cadence can cause your HR to elevate over what you'd otherwise expect for the effort level, though.

To expand your ability to generate power at lower cadences you need to build muscle strength. For me, the most effective means of doing that were 5-10 minute intervals at 50ish RPM and high z3/low z4 power levels.

Leg strength for shorter time periods for me was best improved in the weight room.

Really short-term (under 5 sec) and standing-start power at low and even medium cadence is all about pure strength. It's also almost useless in cycling except for track cycling standing starts. Heavy leg presses, leg extensions, leg curls, and squats if you want to go that far worked for me here. This is where how much weight you can max out at matters. But again - it's almost irrelevant for cycling. (And you'll add muscle mass without adding to your ability to generate power aerobically, which is detrimental to your climbing...)

For medium-length sprint effort low-cadence strength, I found that rapid sets of squats with a light weight worked well. Pick a weight you can squat about 20 times in a single set. Now, do sets of 10 reps. With only ten seconds between sets. Do 4-6 sets, or until you fall over. That workout will kick your butt and seriously improve your sprints in the 20-40 second range at any cadence, but especially at lower cadences. Because you're using a light weight there's no real long-term negative effect on your back. Just don't ever do that workout except in a real squat rack with safety bars to catch the weight or with at least two spotters that you trust.

  • why would you use a power meter instead of a speedo, surely speed's what matters
    – minseong
    Commented Apr 25, 2021 at 17:44
  • 2
    @theonlygusti Because speedometers are completely useless in cycling. Are you going uphill? Are you going downhill? Do you have a headwind? Do you have a tailwind? Are your drafting someone? Are you out in the wind, either solo or pulling the pack? The only thing that matters is who gets across the finish line first, and how fast you go in absolute terms doesn't matter one bit - it's where you cross the finish line compared to everyone else. What matters? How aero are you? How much do you weigh? How much power can you put out and for how long? Commented Apr 25, 2021 at 18:37
  • Everything is about getting more aero, weighing less, and putting out more power. Commented Apr 25, 2021 at 18:38
  • So the point is to look at power output instead of speed because you might face variable headwinds and variable inclines during your sprint
    – minseong
    Commented Apr 25, 2021 at 18:53
  • The risk of doing weights is that muscles bulk up, adding mass. A structured weights plan could be advisable.
    – Criggie
    Commented Apr 26, 2021 at 0:16

Sprinting is usually a (very) short time effort combining high power and maximum speed, per se excluding each other in normal cycling. The idea is to combine a very high cadence and the highest wattage on the pedals in a totally anaerobic effort lasting seconds or 200-250m.

  • 2
    Right, because wattage is torque * RPM. Just because I am mashing hard at low cadence does not mean I am actually making much power.
    – MaplePanda
    Commented Apr 25, 2021 at 16:15

I suspect it depends on your personal ability to change gears while sprinting. That has two-parts.

  1. Can your bike's transmission drop the chain from one cassette cog to a smaller cog while under high load?
    With brifters this is much more accessible, but back in the day noone would have changed gear with a downtube shifter during the sprint.
  2. Can your muscular "sequencer" still generate the same power after as before a gear change?

In short, whatever works best for you is what you should do.

Personally I have a weak sprint with relatively small acceleration - I'm easily able to hit the change lever on the right-hand brifter and any functional rear mech can handle an upshift to a higher/harder/bigger gear or smaller cog.

However if I change too early then its hard to keep the effort going - believe the phrase for this is "Crank Inertial Load". This is the same odd sensation you can get while riding, finding the gear a bit hard, change down for a few seconds, then change back up to the original gear and its okay again.
If I change too late then I've already lost something from it being too hard just before changing.

There are several exercises you can try out to help clarify your own internal understanding.

  • Red light sprint - you're stopped at a red light, and when it goes green you take off and work through your cassette and keep the power consistent until you're out of gears, where you keep holding the same effort for as long as possible.
    If you have a speedo, you need to sprint up to a speed that is at least equal to your normal cruising speed, or ideally faster.
    When I do this, I might only get one or two crank revolutions before changing gear, and mine are only 9 or 10 speed cassettes.
  • Green light sprint - You're approaching an intersection, and you can see it's green in the distance. Put the effort in to reach that green light. Do keep it safe though.
    This tends to lead to the occasional aborted sprint which can be annoying, so have a drink and then do the previous.
  • High gear take-off. From stopped, pull away in one of the harder gear selections, all the way to a shorter high power effort. I don't do this one by choice, I worry about my bike taking damage.

So do try those and see which one you personally feel gets you going faster sooner. You can use perceived effort, and compare that with something like strava ride recording.

  • I believe there is an ancient forgotten technology called "bike computer" or "speedometer" that can provide accurate and immediate speed readings at sight.
    – ojs
    Commented Apr 25, 2021 at 18:22
  • @ojs true - And they're cheap at $30ish compared to fancy pricey head units. But often when sprinting, you're barely able to glance at the display. Strava etc is good for post-sprint analysis, along with flybys if you were racing someone else at that moment.
    – Criggie
    Commented Apr 26, 2021 at 0:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.