A 30-kg child, a 60-kg adult, and a 90-kg adult go on a bike ride.

The 60-kg adult uses a cyclocomputer, which knows the weight of the adult—and gender and age, as well as the weight of the bike.

At the end of a ride, the cyclocomputer declares that the trip used 1000 calories.

Would it be accurate for the 60-kg adult to declare to the child and to the other adult that—if we were to ignore gender + age differences, as well as the difference in bike weights, which may well be serious omissions—that the child has just spent roughly 500 calories and the adult 1500 calories?

Two cyclocomputers may not even agree on the estimation of calories, and so this question only seeks a very rough "ball-park" confirmation whether the relation is linear.

In short, do cycling calories scale near-linearly with weight?

Since older children are typically quoted as needing 1500 calories per day, compared to an adult's 2000, before factoring in exercise, might we use this to expect that a child is using 75% of the calories during exercise as well?

  • 1
    Ask Google for a random number, and multiply the calorie count by that. The result will be at least as accurate as anything else you might do. – Mark Mar 23 at 17:59
  • 1
    Playing around with bikecalculator.com will help with an understanding of first order approximation how varying different parameters affects power needed/calories burned. When cruising at speed on level ground body weight matters very little compared to wind resistance, bicycle tire pressure, etc. – Affe Mar 23 at 19:27
  • @Affe Ah! So that's why the power meter is so crucial. The device knows my altitude (from GPS, not a barometer). But there is no way to feed it information about whether one rider or or another is nearly upright sail-like against the wind, or is bending way down. Hence the power meter is useful as a round-about way to capture the all-important wind resistance. Is this thinking on the right track? – Sam Mar 24 at 0:51
  • @Affe Conversely, we can say that not only extrapolation is impossible, but the very number provided by any cyclocomputer (lacking a power meter) is a nearly useless estimate—for the 60-kg cyclist in this example. The heart rate also does not provide useful information as a parameter to the cyclocomputer to estimate calories, since the heart rate depends on the fitness level of the rider. Is this about right? – Sam Mar 24 at 0:56

No, calorie burn does not scale linearly and can vary based on a great many factors.

Calorie consumption is very closely tied to power output. Power in watts x4 is usually considered a good estimate.

If the entire ride was on a steep climb using tyres with virtually no rolling resistance, then in this case you would be correct. However on a real ride other factors come into play (rolling resistance and aerodynamic drag).

In terms of rolling resistance, there is a large variance between tyres - or even between the same tyres at a different pressure. This will also change (but not linearly) with rider weight.

Increases in body mass also do not lead to a linear increase in aerodynamic drag. Further to this, power output can change a lot due to a riders position and clothing.

  • 1
    Speaking of power, the OP mentioned that the bicycle had a computer but not a power meter. Without actual power data, any estimate the computer produces should not be that accurate. – Weiwen Ng Mar 23 at 9:14
  • 1
    "Power in watts x4 is usually considered a good estimate" which handily makes work done (kJ) == energy burnt (kCal). – Holloway Mar 23 at 18:38

AndyP is right - the scaling is far from linear with weight. Anyway your reading has a fairly large margin for error.

But you can approximate your calorie burn by calculations (which is likely to be less inaccurate than a cyclocomputer if that doesn't take into account ascent, e.g. if solely wheel-driven). Even so the surface is important and is ignored by anything I can find. If you're on road or very smooth gravel, this will be good enough.

Easy option: upload to a fitness website

You could upload to route to Strava (free, but you have to register) for your weight, and then change the weight to get the values for other riders (you need to delete the ride, change the weight, and re-upload for it to update). I believe Komoot and RideWithGPS will also give you this information but I use those only for planning. This is what I would do, and the estimates are OK for casual riding, touring etc.

Tedious estimation option

You can estimate without Strava though. I'll assume a perfectly flat ride for now, then discuss how to account for a bit of climbing.

Using bikecalculator.com you can plug in your weight, riding position, tyres, etc. and the details of the ride. You'll get a number out for calories burn that will hopefully bear some resemblance to what your computer says. Then you can change the rider and bike weights to estimate for other people. If you choose, you can use these to scale your recorded figure, but I'd only do that if using a bike computer with altitude (ideally barometric; GPS altitude is poor in many conditions).

If the ride isn't completely flat, break it down into sections that are consistent within themselves:

  • If you freewheeled down the hills, you can call that zero calories for the downhill stretches.
  • Add up all the flat sections, and plug in your flat speed
  • Add up the climbing. You may have to do this in multiple sections as not only do you want to get the right total, but also to use sections with a consistent speed. So perhaps a long draggy climb would be one section, and a couple of short steep bits another.
  • Estimate separately, and add up.

Bikecalculator's metric trip calculator allows up to 6 segments with different speeds/gradients and the same rider/bike information. Note that it's not clear whether the calculator takes into account a difference in frontal area for smaller (lighter) riders; I would assume not.

Another easy option for the future Use a mobile phone bike computer app for each rider. Plenty don't need a data connection and will run on old phones (e.g. relegated to kids' use). Giving the same details to an app and your computer will also give an idea of the spread in the estimates.

Gold standard Buy a power meter for every bike, along with a suitable head unit (or phone app), again per bike. Stupidly expensive and unnecessary even for many of us who ride quite seriously (but don't race), let alone for a family ride.

  • 1
    In a strange coincidence, my bike computer phone app and Strava estimate exactly the same calorie burn (1498) for this morning's ride (long commute) even though Strava varies almost 10% from one ride to to another on repeated rides of an identical route (not quite the one I did today, but equivalent) – Chris H Mar 23 at 11:33
  • Do you use a heart rate monitor? – Paul H Mar 23 at 15:47
  • @PaulH I have done in the past, but only a standalone one. Coincidences happens - the first ride on this phone had the climb within 2m (in 1600m) of Strava's prediction, again by chance. – Chris H Mar 23 at 16:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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