Running and walking do not burn the same calories per mile. That claim has been repeated for decades, including in the previous version of this article, and the direct measurements contradict it. Running costs roughly 40% more energy to cover the same distance.
Almost everything else about this comparison is more interesting than the calorie arithmetic. Two large cohort studies have tracked runners and walkers separately for years, and their findings on joints in particular run against the folk belief. This article works through what has actually been measured.
The energy claim this article used to make
The old version stated a “1st Scientific Rule: A 150 lb person running 1 mile or walking 1 mile burns the same number of calories,” and listed roughly 90 calories per mile for a 150 lb person regardless of pace.
A 2004 study in Medicine & Science in Sports & Exercise measured this directly. Hall and colleagues used indirect calorimetry on 24 adults covering 1,600 m — a mile — walking at 1.41 m/s and running at 2.82 m/s. On the treadmill, running required 481 kJ and walking 340 kJ, about 115 versus 81 kcal. On an outdoor track the figures were 480 and 334 kJ. Running required significantly more energy on both surfaces.
Those are gross totals, which include the resting metabolism you would have spent anyway. Subtracting it widens the gap, because the walk takes roughly twice as long and accumulates twice as much resting expenditure inside the measurement. The “same calories per mile” rule was never right.
Per mile versus per minute
The comparison changes completely depending on which denominator you use, and this is the source of most of the confusion.
Per mile, running costs about 40% more. Per minute, running costs roughly two to three times as much, because you are moving at twice the speed. Per session, it depends entirely on what you actually do — an hour of walking covers more ground and burns more energy than a ten-minute run.
Which denominator matters depends on your constraint. If time is scarce, running is the efficient option. If joints, fitness or enjoyment set the ceiling, distance is the honest unit and walking simply takes longer to reach the same total.
What the large cohort studies found
The most useful data on this question come from two parallel cohorts run out of Lawrence Berkeley National Laboratory: the National Runners’ Health Study and the National Walkers’ Health Study.
In a 2013 analysis in Arteriosclerosis, Thrombosis, and Vascular Biology, Williams and Thompson followed 33,060 runners and 15,945 walkers for an average of 6.2 years. Expressed per metabolic equivalent hour per day, running was associated with 4.2% lower risk of incident hypertension, 4.3% lower hypercholesterolemia and 12.1% lower diabetes; walking with 7.2%, 7.0% and 12.3%. The differences between the two were not statistically significant except marginally for cholesterol. The authors concluded that equivalent energy expenditures by moderate and vigorous exercise produced similar risk reductions.
These are observational associations in self-selected exercisers, not evidence that either activity prevents anything. What they do support is the exchangeability idea behind the Physical Activity Guidelines for Americans, 2nd edition, which count one minute of vigorous activity as two minutes of moderate — 150 to 300 minutes of moderate activity, or 75 to 150 minutes of vigorous, plus muscle-strengthening twice a week.
Body fat is the exception. In a 2012 analysis of 33,374 runners in PLoS One, Williams found that per MET-hour per day, running was associated with larger declines in BMI and waist circumference than non-running exercise at the same nominal energy expenditure. He argued that the exchangeability assumption fails for adiposity specifically. That finding is unexplained and, being observational, is open to the possibility that people who run differ from people who do not in ways the questionnaire did not capture.
Joints: the folk belief and the data
The received wisdom is that running wears out knees and hips and walking spares them. The evidence points the other way for recreational running.
A 2013 analysis in Medicine & Science in Sports & Exercise followed 74,752 runners for 7.1 years and 14,625 walkers for 5.7 years. Among runners there were 2,004 reports of physician-diagnosed osteoarthritis and 259 hip replacements. Compared with running less than 1.8 MET-hours a day, running more was associated with 15% to 18% lower osteoarthritis risk and 35% to 50% lower hip replacement risk, with most of the reduction arriving by 1.8 MET-hours a day. Adjusting for baseline BMI cut those associations substantially, which suggests much of the effect ran through body weight. The reductions did not differ significantly between runners and walkers. Other, non-running exercise was associated with higher risk, independent of BMI.
A 2017 systematic review and meta-analysis in the Journal of Orthopaedic & Sports Physical Therapy pooled 17 studies covering 114,829 people. Prevalence of hip and knee osteoarthritis was 3.5% in recreational runners, 10.2% in non-running controls and 13.3% in competitive runners. Recreational running sat at the bottom; sedentary living and elite-level volume sat at the top.
Neither study can prove causation, and both face an obvious objection: people whose knees already hurt do not take up running. That selection effect is real and probably inflates the apparent benefit. What the data will not support is the opposite claim — that recreational running is harder on joints than not running.
Injury: where running does cost you
The joint findings should not be mistaken for a clean bill of health. Running injures people at a rate walking does not approach.
A 2015 systematic review and meta-analysis in Sports Medicine pooled prospective studies and reported 17.8 running-related injuries per 1,000 hours of running in novice runners and 7.7 per 1,000 hours in recreational runners. Individual studies ranged from 2.5 to 33.0 per 1,000 hours.
Most of those are soft-tissue overuse problems rather than joint damage, and most resolve. But the pattern is consistent: novices are injured at more than twice the rate of experienced runners, which makes the transition period the dangerous one. That is the strongest practical argument for the old article’s one genuinely good instruction — build a walking habit first.
The limits of this evidence
Almost everything above is observational, and the cohorts were recruited through running and walking publications. People who answer a runners’ health questionnaire are not a random sample of anyone.
Exposure was self-reported: usual distance, usual pace, usual weekly time. Outcomes were self-reported too, including physician-diagnosed osteoarthritis, which depends on who sought a diagnosis. Runners see doctors for different reasons than non-runners do.
Reverse causation is the persistent worry. Lower BMI may lead to more running as easily as running leads to lower BMI, and Williams’ own adjustment for BMI removed much of the osteoarthritis association. No randomized trial has assigned people to years of running or walking, and none is likely to.
What survives all that is modest and worth having: at matched energy expenditure the two activities look broadly similar for most measured outcomes, recreational running does not appear to be joint-destroying, and running is injurious in a way that walking is not.
How to choose
There is no winner here, only a fit to circumstances.
| Your situation | The better default |
|---|---|
| Currently inactive | Walking. Build the habit and the tissue tolerance for a month or two before adding running. |
| Time-constrained | Running, or intervals. Per minute it is the efficient option. |
| Existing joint pain or high body weight | Walking, cycling or swimming, with strength work. Discuss impact loading with a clinician. |
| Fat loss is the priority | Either, but total weekly volume matters more than the choice; intake matters more than both. |
| You dislike one of them | The other. Adherence beats every physiological difference in this article. |
A 2014 analysis of 55,137 adults in the Journal of the American College of Cardiology is worth knowing here: across 15 years of follow-up, benefits appeared at very low doses — under 51 minutes a week, under six miles, one to two times a week — and did not increase much with more. Whatever you pick, the first small amount is where most of the value sits.
For how this fits into a week, see our article on the physical activity dose; for the time-poor case, short high-intensity sessions; and for why neither will do much to the scale on its own, exercise and appetite.
Common questions
Does running ruin your knees?
The available evidence does not support that for recreational running. The 2017 meta-analysis found hip and knee osteoarthritis in 3.5% of recreational runners against 10.2% of controls, and the 2013 cohort analysis found lower osteoarthritis and hip replacement rates at higher running volumes. Competitive-level volume was a different story, at 13.3%.
How many calories does a mile of walking burn?
In the 2004 calorimetry study, a mile of walking at about 3.2 mph cost 340 kJ, roughly 81 kcal, averaged across 24 adults. Energy cost scales with body mass, so heavier people burn more and lighter people less. Treat any single figure as an estimate with wide error bars.
Is walking enough on its own?
For the aerobic half of the guidelines, yes, if you do enough of it — 150 to 300 minutes a week at a brisk pace. It does not cover the muscle-strengthening recommendation, which needs resistance work twice a week.
Should I run every day?
Injury rates in novices are high enough to argue against it. Most of the association with lower mortality in the 2014 cohort appeared at one to two runs a week, and adding volume added little. Alternating running with walking days is a reasonable structure.
The bottom line
Running costs about 40% more energy per mile and two to three times as much per minute, so it is the efficient choice when time is short. At matched energy expenditure, the large cohorts found broadly similar associations for both. Recreational running does not appear to be harder on joints than not running, but it injures novices at roughly 18 events per 1,000 hours, which is the real argument for starting with walking.
Neither activity is a nutrition strategy, and no drink changes what is written above. If a greens blend is a convenient way to add plant food to an active week, that is what Greens Plus is for.
About the Author
