Yes.

As other posters have noted, the 1708 kj from Strava is a good estimate. The others are likely 69-75% high.

Some authorities use MET Metabolic Equivalent of Task which are supposed to scale with your weight. That's wrong for riders above their ideal weight because on flatter terrain most of your energy is going into overcoming aerodynamic drag which doesn't increase with weight. It's wrong because aerodynamic drag is less than proportional to weight.

The basic METs are also way off for some reason.

Gadgets based on heart rate can also be off by a factor of two.

Give some thought to what you are eating, more than calories. Pull out the carbs and sugars, replace with fats and proteins. I think you"ll see the trend reverse.

People obsess over weight a bit too much it seems.

It is just one metric of many that can be used to judge fitness.

Obsessing over 1 or 2 pounds is a bit silly.

Doesn't matter that much in the grand scheme of life.

That's exactly what I had done. Which is why the resulting gain was a surprise. When I cut back drastically on carbs I ended up feeling weak.

Just a quick question. I thought calorie burn was related to weight to some degree. At 190 lbs am I not going to burn more cals than my 125 lb gf for a given ride?

Yesterday 11 of us went off on a ride of 42 mi. There was 2600 ft. of climbing but 1600 ft. was in the first 18 mi. That amounted to 90 ft./mi. which is much more than I usually do. I worked my butt off. Most of the climbs were the long, slow kind that I suck at. So, when I'd get over a hill I had to continue to work to get back up to the group. Even the ride back was a bunch of rolling hills. Those weren't so bad 'cause I am actually good at powering over that kind of rise. Of course, by then, I was feeling kind of worn out. Point is that Strava says I burned 1495 Kj on that ride and over 1700 on the ride last week that I had referenced in this thread. Even though last week's ride was 6 mi. longer my body is telling me that can't be true.

Marginally. But if your goal is to lose weight, put your goal weight into the calculations.



42 miles? Using the 100 cal/5 km calculation, that's 1352 cal. That's what I would go with.

Strava appears to use speed. So if your average speed on this ride was, say, 18 km/h because of the climbing, Strava figures you haven't worked as hard as if you did a ride at, say, 22 km/h on a flatter ride.

But if you're trying to lose weight, just go with a lowish number of calories burned. Going low makes up for an extra calories you might have consumed. :)

I must be missing something. You're saying I should use 100 cal/5 km regardless of speed or elevation data?

Yes. Because the formula incorporates speed.

If you ride 20 kilometres in an hour, and you go out for a 1-hour ride, you've burned 400 cal in an hour.

If you ride 25 kilometres in an hour, and you go out for a 1-hour ride, you've burned 500 cal in an hour.


As for elevation ... usually, what goes up must come down. So you spend some time and energy getting up the hill, but then you coast down the other side and you don't expend much energy at all for the speed you're going.

Where can I see this formula? Thanks.

Machka's formula is simplifying things quite a bit; climbing definitely affects calorie burn, and the heavier you are the more that is true. Speed also affects calorie burn, because of aerodynamics; the faster you're going, the more calories you'll burn per mile. Having said that, her formula is a pretty decent one to use for weight-loss purposes, because it's better to underestimate the calories you're burning than to over-estimate.

You seem to be pretty focused on getting an accurate calorie count, but the reality is the only way to do so is with a power meter. Anything else is going to be just an estimate, there are just too many variables to factor in.

The number from Strava will be better than most other estimates, provided you understand it's limitations. You need to enter an accurate body weight and bike weight in your profile. You also need to understand that Strava's calculation doesn't factor in wind or drafting/pacelines. So if you're riding with a tailwind or in a group and getting a significant draft, Strava will overestimate calories.

For weight-loss purposes, your best bet is to get a power meter. Second-best, use the lower number between Strava's estimate and Machka's formula for a given ride. It's not about getting the highest calorie number possible; it's about making sure you don't eat too much relative to the work you're doing.

Energy (calories) from climbing is dead simple. m*g*h. Kilograms times gravity times meters climbed equals joules. By a mathematical and physiological quirk, kilojoules is about the same as nutritional calories used.

This is additional due to climbing, plus whatever you'd have burned if it was flat. Hard as it may be to believe, simple as that.

The body going into starvation mode is known to be more or less a myth nowadays. The dip in metabolism people are concerned about is transitory. You're just hungrier and eat more, due in part to an increase in some hormones, and the base metabolic rate is slowed somewhat immediately after severe dieting.

Despite all the theories and math, converting distance, speed, climbing, etc to calories is only possible if you're looking for ballpark numbers. That's true even with a power meter.

There are too many variables, and while a power meter will address some of them such as varying wind drag, is can't address one big one, which is that human engines vary greatly in efficiency. So the calories needed to produce a given output (if it's measured correctly) will vary among people, and even in the same person with changes in temperature or other conditions.

So, rather than sweat trying to pin calories down precisely, accept them as a rough guideline based on simple formulas (or a power meter) and let the long term average things out for you.

Very true, and even kilojoules to calories is very ballpark, a few percent off in fact.

But using math again can put the ballpark into perspective. Our expected metabolic efficiency is given at 22% plus or minus 4%. So bottom line if some formula says "25 calories" it might really be 24 calories, or 26 calories. I think it's pretty useful for that level of precision, though it's true you'll never get an exact number this way.

There's a rule of logic that you can't draw specific conclusions from generalities. The efficiency conversion number itself is an approximation, and while it may get into a fairly tight band over time and numbers of people, it's actual variation is greater. So, you're still back to calculations that are only as accurate as the size of the database, or time interval allows.

It's like the old story of the statistician who drowned crossing a river with an average depth of 3'.

Of course everybody has his own approach and/or belief in numbers, but I'm an old school (before computers made complex calculations practical), and prefer to step pack and look at the big picture.

Watch weight over time for gain or loss. Don't fret if it seems to be 3 steps forward and 2 back, you only care about the trend. At the same time track your riding (also in a big picture sense, ie 100miles/week).

After a while you'll get a picture which will tell you if, on balance, the riding you do, however you tend to do it, is burning more or less than your calorie input supports. The rest is common sense and there are plenty of guides about what to eat, when to do it, and so on. Meanwhile, if you decide you need to shave 1,500 calories/per week (for example) you know if you can do it by cutting out deserts, portion control, giving up starches or whatever -- your call.

The results (over time) will prove the estimates, and you'll have to make your own decisions about sustainability.

In a way, this kind of thing is like a pointillist painting. You can focus on each of the dots or you can step back far enough to let the big picture emerge.

I take objection to that. The statistician isn't the person who demands certainty and precision and will be more aware of the risk of deeper water than the other. Practically all medical science is predicated on probabilities and confidence intervals.

Every person is not going to have metabolic efficiency between 18% and 26% but the odds are very strong that a given individual will. Especially if we're willing to account for extreme outliers, and reasonably consider healthy adults having decent medical care, reasonable diet, at least moderate levels of physical activity etc.

The assumption that it is all about metabolic efficiency is yet another generality that fails.

100 cal/5 km.

Or if you prefer ...

33 cal/mile.

How so? I'm interested in why the energy used climbing (additional to the bike ride if it were flat) is about more than mgh and metabolic efficiency.

I think that this description is precisely correct, not a generality at all, but I've been wrong now and again. So seriously, what's your reasoning here?

Note that sprince puts out just shy of 2000 watts when he rides ... like Mark Cavendish does when he is in a full-on sprint for the line. :)


Post 6 - http://www.bikeforums.net/18876308-post6.html

Sorry. I still don't get it. It doesn't look like a "formula" to me. It looks more like a result of a computation. Where does that take into account weight, gender, age, speed, climbing, etc? There's no difference between a 30 mi. ride at 16 mph and 3,000 ft. of climbing and a 30 mi. ride at 10 mph with 1,000 ft of climbing? Anytime you ride you burn 33 cal/mile?

Weight is not that important.
Gender is not that important.
Age is not that important.
Speed is incorporated into the formula.
Climbing is not that important (what goes up must come down).


Yes, there is a difference between the two rides.
Yes, you burn 33 cal/mile.

If you cover the 30 miles at 16 mph, that means you were out there for 1 hour and 52 minutes. It took you 1 hour and 52 minutes, to burn 990 calories.

If you covered the 30 miles at 10 mph, that means you were out there for 3 hours. It took you 3 hours to burn 990 calories. In this case, if you decided to call it a day at 1 hour and 52 min ... you'd only have burned approx. 650 calories in 1 hour and 52 minutes, instead of 990 calories.

It's an approximate rule of thumb, which is what I think he's looking for in the question. It works better at low speeds (particularly running speeds) and flat. Higher speeds will take more than the 33 calories per mile (power goes up with speed cubed) and climbing is added on top of it (which is why I brought up m*g*h). Coming down after a climb will be at the higher speeds where wind resistance predominates, requiring more power over the ride, which means more calories burned. Or perhaps you're coasting down using no power on that portion and it all depends on what you burned going up. It's all approximate.

Don't hold my breath in other words.

Yes, unless you're hooked up to a power metre (and I'm not even positive about those things ... I don't know) ... it's all approximate.

The 100 cal/5 km thing is a quick and easy calculation a person can do while riding. And we've discovered it's reasonably close to what Rowan's strava says. So we go with it.

Give it some time. Especially after a life time of ingesting lots of carbs daily it may take months/years to re-regulate. Switching to fat burning vs. carb/sugar burning takes time and I suspect your body is still hording fat. Yes, no doubt there is a drop off in immediate energy. I recall days I went to the gym and felt so lethargic I left after 10 minutes. The pay off in the long run [pun -- endurance efforts], is totally worth it. :)

Thanks. This is pretty much what is happening to me.

I think this is what has been confusing me. What you are using is an estimate based on your experience. Nothing wrong with that. But, when you called it a "formula" I was looking for an actual formula. And, since over the years I've spent some time researching how to estimate calorie burn, and all of the results include things like gender, age, weight, speed, etc. I couldn't understand how you got to where you got to. Much clearer now. Thanks.

My issue is, if I went by the 33kcal/mi thing, for it to jive with my PM data I'd have to be over 35% efficient. This morning was 2,692kj for 51.6 miles, ~52.2kj/mi-- and that was my usual mixed-Z2 intensity. I average much closer to 50kcal/mi (from ~4 months of PM use,) which implies that by rule of thumb, I'm using 50% more energy than the average person? I don't buy it.

As has been said a zillion times, without a PM, we're all just guessing.

I put your formula 100 cal/5km to the test today...I went for a 5 km ride to do some errands and according to my bike computer I burned 65 calories on a 5km ride...There are a lot of variables, I rode slow and easy but if I went hard and fast I could of easily burned around 150++ calories in those 5 km.

The amount of energy used by a person on a given climb is dependent on their efficiency on the bike, the weather, the weight of the bike, the clothing worn, and it goes on and on. Two people of identical weight, build, body fat, and fitness level could use vastly different amounts of energy to complete the same climb in the same time. Even a power meter can only provide part of the picture. As to weight loss, it adds another dimension with more variables that have to be factored into the calculations. So you might find that the variation for all humans is within a certain range for metabolic efficiency, but chain that in series with another variable and the end result could be exponentially different from a one dimensional formula.

Nope. The additional amount due to the climb (as we specified) does not depend on any of those factors, other than weight (the "m") and their efficiency (the metabolic efficiency). There will be zero difference in additional energy due to those factors.

But thanks for the explanation of what you were thinking.