How many calories does rucking actually burn?

Nobody has measured how many calories rucking burns for you. Every figure you have been handed — by a watch, a calculator, or an article with a confident table in it — is the output of a prediction equation fed your bodyweight, your load and your pace. Some of those equations are nearly fifty years old. The honest answer is a range you build yourself, and the size of the pack's contribution to it is smaller than the marketing implies.

This article is the arithmetic. Where the numbers come from, which model produced them, how to run the back-of-envelope version on your own bodyweight, how big the load premium really is at recreational weights, and why two rucking calculators can hand you figures that differ by hundreds of calories for the same hour. If the activity itself is new to you, our rucking basics for beginners covers the setup this piece skips. And a caveat we mean literally: if you are pregnant, managing a cardiac or joint condition, or returning from injury, whether to carry load at all is a question for a doctor or a qualified coach before it is a question for a calorie article.

How many calories does rucking burn? Start with what the number is

It is a model output. That is not a technicality — it changes what you can reasonably do with the figure.

Measuring energy expenditure properly means indirect calorimetry: a mask, analysed expired air, oxygen uptake and carbon dioxide production recorded breath by breath. That happens in laboratories, on treadmills, on small samples of mostly young and mostly military-age adults. Nobody has done it on your Saturday loop with your pack.

So the field predicts instead. Feed body mass, load mass, speed, grade and terrain into an equation, and out comes a metabolic rate in watts, which converts to kilocalories. Everything downstream — your watch, the app, the calculator with the sliders — is a wrapper around one of those equations, or around a MET table that was itself built from a different set of laboratory measurements.

A prediction is only as good as the population and the conditions it was fitted to. That is the thread running through everything below. We extended this argument from the intensity side in rucking vs walking; here we are taking apart the calorie figure itself.

The Pandolf equation, and why a 1977 model still runs your calculator

The best-known prediction for load carriage is the Pandolf equation, published by Pandolf, Givoni and Goldman in the Journal of Applied Physiology in 1977, volume 43, pages 577 to 581. It estimates metabolic rate from body mass, external load mass, walking velocity, grade and a terrain factor. Nearly five decades on, it appears to still sit underneath many of the rucking calculators on the internet, though few of them disclose which model they run.

It also has documented limits. According to Wikipedia's summary of the equation's known constraints, it "is known to underpredict the metabolic rate for higher loads and speeds," underpredicts over prolonged periods, performs poorly for people in heavy protective clothing, and needs a separate correction factor to handle downhill walking at all.

Newer models exist. A 2022 paper in Medicine and Science in Sports and Exercise by Looney and colleagues, "Modeling the Metabolic Costs of Heavy Military Backpacking", built a replacement from treadmill trials at loads of 0, 22, 44 and 66 percent of body mass and speeds from 0.45 to 1.97 metres per second. Against those measurements the Pandolf equation carried a bias of minus 0.44 watts per kilogram — a systematic underestimate. Their new Load Carriage Decision Aid backpacking equation reported a bias of minus 0.01 watts per kilogram on internal validation and minus 0.08 on external validation across historical datasets.

The authors also state the underlying physics plainly: "walking energy expenditure rises nonlinearly with increased load relative to body mass," with "heavier loads incurring disproportionately greater costs." The same group extended the approach to vest-borne load in a 2024 paper in the same journal, using 20 military-age adults across vest loads from 0 to 66 percent of body mass.

Two things follow. The better equations are military tools fitted mostly to military loads, and we could find no consumer rucking calculator that had been updated to use them.

The MET method: the honest back-of-envelope, worked once

For most people the useful move is not a better equation. It is a rough estimate you can audit yourself.

The 2024 Adult Compendium of Physical Activities is the reference table behind a great deal of tracker maths. It defines one MET as 1 kcal per kilogram per hour, equivalent to the oxygen cost of sitting quietly at 3.5 mL per kilogram per minute. It lists brisk walking at 3.5 to 3.9 mph on a level firm surface as code 17200 at 4.8 METs.

Here is one worked example, and it is an example — not your number.

Take an adult of 180 pounds, or 81.6 kg, walking one hour at a brisk 3.5 to 3.9 mph on flat firm ground. Unloaded, at 4.8 METs, that is 4.8 multiplied by 81.6 kg multiplied by 1 hour: about 392 kcal.

Now watch the first crack appear. The other standard form of the same calculation — MET multiplied by 3.5, multiplied by body mass in kg, divided by 200, to give kcal per minute — assumes 5 kcal per litre of oxygen and returns 6.85 kcal per minute, or about 411 kcal for the hour. Same person, same walk, same table, roughly 19 kcal apart before a single pound of load has entered the sum. Swap your own bodyweight and duration into either version and the structure holds.

How big is the rucking premium, really?

This is where the honest answer disappoints people, and where the published figures disagree with each other badly enough to be worth showing.

An ACE-commissioned study run by Len Kravitz and James McCormick at the University of New Mexico's Exercise Physiology Laboratory put 13 untrained women aged 18 to 55 on a treadmill at 2.5 mph across grades of 0, 5, 10 and 15 percent, wearing vests at 10 and 15 percent of bodyweight. Their finding on flat ground is the one to sit with: the 10 percent vest "did not create a significant increase in VO2," and the 15 percent vest was needed to produce one — worth "a 12 percent increase in the number of calories burned over not wearing a vest." On a 5 or 10 percent grade the 10 percent vest bought about 13 percent more calories, and going up to 15 percent added nothing further. Two caveats we would not skip: the sample was 13 untrained women at a single slow pace, and those were vests rather than rucksacks — where the load sits is its own comparison, and we run it in rucksack or weighted vest.

Apply that 12 percent to the worked example above and the loaded hour lands near 439 kcal, against 392 unloaded. The pack's entire contribution is roughly 47 kcal for the hour.

Now the disagreement. The 2024 Compendium's own codes imply something very different. Walking at 2.5 mph on a firm level surface is code 17170 at 3.0 METs. Treadmill walking at the same 2.5 mph and 0 percent grade carrying a 40 lb load is code 17434 at 4.8 METs — a 60 percent premium at identical speed. The same table also lists "walking with a day pack, level ground, assumed in the city" at 3.5 METs and "backpacking, hiking with a daypack" at 7.8 METs. Two daypack entries, more than double apart.

We cannot reconcile these, and we are not going to pretend otherwise. The loads differ, the populations differ, the source studies differ, and the non-linear scaling Looney and colleagues describe explains some but nowhere near all of a 12-versus-60-percent gap. What survives across all of it is the direction and a rough ceiling: at recreational loads the premium is a modest percentage on top of the same walk, not a multiplier. Choosing the load itself is a separate question with its own published ranges, and we handle it in how much weight to ruck with. For what else the load is doing while it is not doing much to your calorie count, see the benefits of rucking.

Is rucking good for weight loss?

Directly: it can contribute to fat loss, and it is not a shortcut around energy balance.

Start with the size of the effect you are adding. In the worked example, the load bought about 47 kcal in an hour. Set that against the measurement error in the device you would use to track it. A 2022 systematic review in the Journal of Medical Internet Research by Germini and colleagues, covering 65 articles on wrist-worn trackers, reports that "for EE, the MAPE was greater than 30% for all the brands, showing poor accuracy across devices," and concludes that "none of the tested devices proved to be accurate in measuring EE." Thirty percent of a 392 kcal walk is about 118 kcal. The uncertainty in the reading is more than twice the effect you are trying to read.

Then the wider literature on exercise and bodyweight. A 2011 systematic review and meta-analysis in the American Journal of Medicine by Thorogood, Mottillo and colleagues identified 14 randomised controlled trials of aerobic exercise without dietary change, covering 1,847 patients, and pooled the six-month and twelve-month programmes separately. The pooled differences were about 1.6 kg at six months and 1.7 kg at 12 months in overweight and obese populations, and the review's conclusion was that "isolated aerobic exercise is not an effective weight loss therapy in these patients," though it "may still be an effective weight loss therapy in conjunction with diets."

None of that is an argument against rucking. It is an argument against buying a pack as a weight-loss device. Our own reading, flagged as ours: the variable that actually moves is adherence. A modest per-session premium repeated four times a week for a year is worth more than a large premium abandoned in March, and a pack that makes a walk feel like training is a legitimate adherence tool. If you want the sessions structured rather than improvised, a beginner rucking training plan is the place to start.

Why rucking calorie calculators disagree with each other

Because they are running different equations, and because those equations disagree even under ideal conditions.

The cleanest demonstration we found is a 2021 study in the Journal of Applied Physiology by Weyand, Ludlow, Nollkamper and Buller. They took seven participants over a 6,415 m outdoor course with gradients from minus 3.0 to plus 5.0 percent, across asphalt and grass, unloaded and carrying 30 percent of bodyweight, and compared measured energy expenditure against four laboratory prediction equations. Every one of them overpredicted the unloaded trials: the Minimum Mechanics model by 4 percent, the ACSM equation by 13 percent, Pandolf by 17 percent and Looney by 20 percent. In the loaded trials — where only two of the four equations can take a carried load at all — the errors ran from minus 2 percent for Minimum Mechanics to plus 13 percent for Pandolf.

Read that carefully, because it cuts both ways. Pandolf underpredicts against laboratory treadmill data at heavy loads and overpredicts against real-world field walking. The direction of the error depends on the conditions, which means no single correction factor fixes it.

So a calculator's answer depends on which equation it wrapped, which correction it applied, what terrain factor it assumed, and whether it silently swapped to a MET table. Those choices are rarely disclosed. The spread between two reputable tools for the same inputs is not a bug in one of them — it is the honest width of the current science, hidden behind a single bolded number.

What we would do instead of chasing the figure

Our position is that the calorie number is the least useful output of a ruck, and the easiest to overfit.

• Treat any figure as a range, not a value. If a tool says 520, read it as somewhere between roughly 400 and 650 and get on with the session.
• Compare against yourself, not against other people. The same tool used consistently on the same route is a usable relative signal even when its absolute accuracy is poor.
• Turn pace before you turn load. The Compendium's own walking codes climb from 3.0 METs at 2.5 mph to 4.8 METs at 3.5 to 3.9 mph — a bigger jump than the flat-ground load premium in the ACE data, and it costs nothing.
• Judge the session by duration and consistency. Minutes accumulated are the variable you control directly and the one the weight-loss literature keeps pointing back to.
• Do not eat back an estimate. With tracker error above 30 percent, treating a predicted burn as licence is the most reliable way to erase it.

The honest answer

How many calories does rucking burn? More than the same walk unloaded, by a percentage rather than a multiple at recreational loads, and by an amount nobody has measured on you specifically. The best available number is one you build yourself from a MET table, understanding it as an estimate with wide and largely unquantified uncertainty around it.

What the evidence does not show is equally worth naming. It does not show that rucking burns calories at running's rate. It does not show that the load premium scales in a straight line either: Looney and colleagues describe external load as having "a multiplicative, non-linear effect on walking metabolic rate with light loads causing small increases and heavier loads incurring disproportionately greater costs," which means the steep returns arrive only at loads far above beginner ranges. And it does not show that adding load to walking, on its own, produces meaningful weight loss; the pooled trial data on aerobic exercise alone is modest and consistent about that.

What it does support is narrower and more useful. A pack raises the intensity of a walk you were going to do anyway, at low impact, without needing a gym. If that makes the walk happen more often, the arithmetic in this article stops mattering very much — which is, in our experience, the point. For the full on-ramp, pack fit and progression, the complete rucking guide has the rest of it, and what the load actually changes covers the intensity side we only touched here.

Then stop calculating and go for the walk. The estimate will still be an estimate when you get back.