Treadmill vs. Real Hill: Which is harder to run

April 12, 2025 10:14 PM

Steve Mould takes on the question... Treadmill vs. Real Hill: Which is harder to run - YouTube. Good arguments, good experiments, conclusion. Right or Wrong? Pick sides.
posted by zengargoyle (38 comments total)

I think he's still WRONG.
posted by zengargoyle at 10:14 PM on April 12

Isn't the radio car just a complicated treadmill dynamometer? I mean, kinesiology researchers probably already have treadmills with built-in dynos.
posted by Monday, stony Monday at 11:44 PM on April 12

I think the power that needs to be generated by the body in order to sustain a run on an inclined treadmill is, to a very good first approximation, the same as the power required to run up a hill with the same gradient at the same running speed with the same traction and the same airspeed. I think that even given its characteristically sloppy experimental design, Mould's demonstration that the powers involved are within 10% of each other is compatible with this view.

There's an argument that running up the hill causes the body to gain gravitational potential energy in a way that running up the treadmill doesn't, but I don't think it's valid. Gravitational potential energy is not an absolute measure, it's always a relative to some other height. For example, my GPE with respect to the ground I'm standing on right now is zero, but as any Road Runner fan well knows, Wile E Coyote's GPE with respect to the ground he's suddenly no longer standng on can be quite high.

Both hill and treadmill runners gain the same amount of GPE with respect to their starting heights, even though from the natural reference frame of an observer not on the treadmill, the treadmill runner's gain is due to their starting height sinking rather than their body rising.

It's a pity that Mould chose to obfuscate this point with a bunch of waffle about treadmill friction.
posted by flabdablet at 2:14 AM on April 13 [3 favorites]

I think that if someone's a popular YouTuber explaining something, it's a dead cert that they're an ignorant buffoon. I am a typical MeFite. (Also a fan of Steve Mould).
posted by ambrosen at 2:20 AM on April 13 [1 favorite]

This is wild timing for me. I have been a runner for over a decade. Last summer it was Too Damn Hot to run, so I got a gym membership and have been running on a treadmill since. Just yesterday the weather finally got just right for a run in my neighborhood outside in the real world. By the end I realized that running on the road is harder on your body/legs than a treadmill. Even when you elevate the treadmill, it still absorbs a lot of the shock of your footfalls. Roads don't. I am sore today, and I haven't had any leg soreness in years and years.
posted by zardoz at 3:39 AM on April 13 [9 favorites]

There's also the fact that one of the great glories of running is being outside in the real world and seeing and hearing cool stuff and getting away from the real world for a while, and the treadmill in a gym with Fox News on manages to cancel out all of that and just be a boring thing you're doing to your body.
posted by hydropsyche at 4:18 AM on April 13 [6 favorites]

For me the problem with treadmill running is that I can quit just by stepping off the treadmill. Boom. You're done. In real world running if you stop you have to walk home and that takes 2-3 times as long and you're dressed for running comfort rather than walking comfort. This ease of quitting makes treadmill running very psychologically taxing. There's also no great indicators of progress on a treadmill other than the data screen and watching the screen while running is like watching a clock at work.

That's an entirely different form of suffering than the suffering I am seeking to overcome when I run.
posted by srboisvert at 4:52 AM on April 13 [3 favorites]

This was more of a deep dive than I could commit to and noped out at “I’d like to pursue all the methods of answering this question” but I did still find it an interesting question. The “but you’re not moving your body UP” totally made sense to me. But still, I did a treadmill stress test a few years ago that ended with, like, a 5.5mph pace on a 10% incline, and while I couldn’t say for sure it’s exactly as hard as the same hill climb outside, it was sure HARD AS FUCK. I don’t think the stress test protocol would have persisted if it didn’t actually increase effort (lol assuming modern medicine only does stuff that works).

I hate treadmill running so much that the mental effort makes it immensely more physically taxing for me. I’ve run 15 miles out on the road just soaking up the world, but even with the best entertainment possible, I don’t think I’ve ever been able to push past 2 miles on a treadmill.
posted by obfuscation at 5:28 AM on April 13 [2 favorites]

For me it's the opposite, when running outside my brain is constantly screaming "Where are you going, we're getting too far from home!" Every step you take doubles the total amount of distance you need to run, and calculating when I'm "half done" and need to turn around is a total mindfuck. But on a treadmill I can just run until I don't want to anymore and then hop off and I'll be done right away, which feels much less intimidating.

I'm mean I don't really run either way, running is the worst, but that's been my limited experience in vs out.
posted by grog at 5:31 AM on April 13

Yeah my personal experience does rely on the baseline assumption that I love running. :)
posted by obfuscation at 5:36 AM on April 13 [2 favorites]

I don’t get the argument that you’re not moving up on a treadmill. It seems to me that if you don’t run, it pushes you downward. If you do run, you are continually running against that downward movement which, being the opposite of down, is physically analogous to “up”
posted by anazgnos at 5:45 AM on April 13 [2 favorites]

Enh. On an inclined treadmill you have to lift your feet higher, but you don’t have to lift your torso. On a hill, you lift everything. Escaping a gravity well requires energy. Same deal with a stairmaster. You lift your legs but not your torso. It’s a workout but not the same one.
posted by seanmpuckett at 6:22 AM on April 13

On an inclined treadmill you have to lift your feet higher, but you don’t have to lift your torso.

Compared to what?

On the hill, you need to lift your whole self in order to gain altitude. On the treadmill, you need to lift your whole self in order to avoid losing it. The lifting effort is the same in both cases.
posted by flabdablet at 7:25 AM on April 13 [3 favorites]

Escaping a gravity well requires energy.

The reason that the very good first approximation is so very good is that the gravitational field strength measurable at the end of any practical hill run is so very very close to that measurable at the start.

I don't care how fit you are, you're not running up that hill at escape velocity.
posted by flabdablet at 7:28 AM on April 13 [1 favorite]

You don’t know me.
posted by obfuscation at 7:29 AM on April 13 [5 favorites]

Mrs Pancakes?
posted by flabdablet at 7:31 AM on April 13

As zardoz said: a real-world hill is not a perfectly angled slope for its entirety. Even running on a flat surface in the real world (outside of something like a track or the Utah Salt Flats) involves hopping over bumps, up and down curbs, avoiding bits of broken sidewalk, stepping over puddles in the road, avoiding other people, etc. It's never even and continuously level or angled like a treadmill. Running outside uses way more small balancing muscles in your ankles and legs. It's simply easier to keep your balance on a treadmill.

You can close your eyes and run an entire mile on a treadmill! Try doing that on a real hill somewhere. It's more of a challenge physically and mentally to run balanced and safe on any kind of real-world hill versus a stable treadmill because there are always obstacles to look out for and to physically account for in your stride.
posted by SoberHighland at 7:44 AM on April 13

> waffle/treadmill fiction

would read
posted by stevil at 9:20 AM on April 13 [1 favorite]

It's simply easier to keep your balance on a treadmill.

I find the opposite to be true. When walking or running out in the world, I can go along paying attention to everything around me without constantly having to correct for balance. On a treadmill, I never feel completely balanced, in fact I have to rest a hand lightly on one of the handles to help keep me upright. I've always wondered why that's so, but it's absolutely consistent.
posted by Greg_Ace at 10:14 AM on April 13 [1 favorite]

I didn't see that he addressed the most "common sense" way to think of it.

Real hill: you add +1 (arbitrary unit) of potential energy
Inclined treadmill: if you do nothing, you end up at -1. So you have to add +1 to stay where you are.
In both cases, you added +1
posted by ctmf at 10:58 AM on April 13 [1 favorite]

i.e. you ARE actually lifting yourself through gravity... from where you WOULD be if you didn't.
posted by ctmf at 11:00 AM on April 13 [1 favorite]

Previously
posted by Rumple at 11:35 AM on April 13

you ARE actually lifting yourself through gravity... from where you WOULD be if you didn't

just as is the case on the hill.

There would be a difference in the amount of work done between treadmill and hill scenarios if the strength of the gravitational field were different at the end of the hill run from what it was at the beginning, but for real-word runs that don't involve falling off things, this difference is so small as to remain well below the experiential noise floor.
posted by flabdablet at 11:43 AM on April 13 [1 favorite]

Running on treadmills makes me want to poke my own eyes out. It’s fucking boring, plus I never feel really balanced or connected to the “road”. If I didn’t have a nice park to run around or was a super marathoner needing to put in 10 or 20 miles in the summer heat, I’d probably use one, but boy do I hate it. I just run 2 miles every other day and if it’s 90 F or 30 F or raining I just go anyway. It’s fun to check the creek levels !
posted by caviar2d2 at 12:15 PM on April 13 [3 favorites]

I hate that a treadmill makes me have to run a consistent constant speed instead of whatever natural variation I feel like at the moment. I guess that would be useful in training for sport, but it's annoying. On the other hand, if I had a treadmill at home I could just run any time without the overhead of, eh, what's the weather like, get dressed, plan a route, lock the house, carry my jingly keys, watch for cars, etc. Even if I go to a park with a track or trail, now it's a car trip to get to my running route. Convenience trade-offs either way, and it doesn't take much sometimes for me to say eh, some other day.
posted by ctmf at 12:22 PM on April 13

flabdablet: "I think the power that needs to be generated by the body in order to sustain a run on an inclined treadmill is, to a very good first approximation, the same as the power required to run up a hill with the same gradient at the same running speed with the same traction and the same airspeed. I think that even given its characteristically sloppy experimental design, Mould's demonstration that the powers involved are within 10% of each other is compatible with this view.

There's an argument that running up the hill causes the body to gain gravitational potential energy in a way that running up the treadmill doesn't, but I don't think it's valid. Gravitational potential energy is not an absolute measure, it's always a relative to some other height. For example, my GPE with respect to the ground I'm standing on right now is zero, but as any Road Runner fan well knows, Wile E Coyote's GPE with respect to the ground he's suddenly no longer standng on can be quite high.

Both hill and treadmill runners gain the same amount of GPE with respect to their starting heights, even though from the natural reference frame of an observer not on the treadmill, the treadmill runner's gain is due to their starting height sinking rather than their body rising.

It's a pity that Mould chose to obfuscate this point with a bunch of waffle about treadmill friction."

Will students of history say this was the moment we reached peak flabdablet? I say this with fondness and respect
posted by ginger.beef at 2:07 PM on April 13

I guess I'm convinced. More so by the comments here than by the video, but I guess I'm convinced. I haven't done much running for a decade or more now, but I've participated in a few endurance events (up to 96km) as a walker/jogger. Why I say I guess I'm convinced is that running/jogging/walking in real life has all sorts of obstacles like bumps and rocks and slippery patches and all sorts of things, so I do think it's at least a little harder in terms of just being able to stay on track and not fall flat on your face and it's much harder to focus on binge-watching series on Netflix when you're outside running than it is on a treadmill. It's perhaps not that a human exerts more effort in climbing the hill vs the treadmill, but any kind of rough surface uphill is a lot tougher to traverse than the equivalent surface on the flat or downhill.

SoberHighland: "You can close your eyes and run an entire mile on a treadmill! Try doing that on a real hill somewhere. It's more of a challenge physically and mentally to run balanced and safe on any kind of real-world hill versus a stable treadmill because there are always obstacles to look out for and to physically account for in your stride."
I have discovered, accidentally but repeatedly, that you can in fact walk and/or jog while sleeping, as long as the surface is reasonably smooth (eg the middle of the road in the middle of fuck-knows-where at 2 am when it's easy to convince yourself of just how absolutely fucking stupid you are for voluntarily doing this in the first place). However, it's not possible to do that in a straight line. It's not the road camber or anything like that, because each individual will veer off to the same side every time. That's how we knew the person in front had fallen asleep and wrung some small amusement from wondering how far they would drift before waking up or falling on their face in the road.
posted by dg at 5:52 PM on April 13 [1 favorite]

I think part of the confusion is that thinking about the problem in terms of potential energy is not actually a very helpful way of thinking about human exertion.

Holding a difficult yoga or gymnastics position will change nothing about your body's potential energy, but it will still be incredibly tiring and require great exertion, because maintaining muscle tension isn't "free" even if your 'energy' isn't going into potential or kinetic energy. Contrary, most of us can stand for relatively long periods with much less effort.

From a mechanics perspective, the person running on a treadmill is not losing and then gaining gravitational potential energy. What they are doing is applying a consistent force downwards, which they have to move their limbs to maintain. The same person running up a hill is gaining potential energy, but they are also just maintaining the same upwards force.

It might be counter intuitive, but a person standing on a hill and a person running up a hill at a constant speed are both pushing downwards with the same force (neglecting air resistance). They are both maintaining a constant velocity against a gravitational force, so they are both applying an equal downwards force to balance it out to net zero. It's just that biomechanically maintaining that force while running up an incline is hard and maintaining that force while standing is easier.
posted by Ktm1 at 7:50 PM on April 13 [1 favorite]

My personal experience is that running on a treadmill is SO FUCKING BORING that it feels much much MUCH harder than running outdoors. I can never run the same pace on a treadmill as I do outdoors without completely exhausting myself.

Y'all do you. I know my truth.
posted by suelac at 9:02 PM on April 13 [3 favorites]

Great point about the gymnast's effort, Ktm1. But I think Mould's "intuitive" sense that sparked the question was, all other things being equal, the only difference is really the ending altitude. And that potential energy shouldn't come for free. But then he thinks it through a bit, and changes his mind, now he thinks it SHOULD be the same energy expended. Then he tries it and gets a result that's neither solidly proves one answer or the other. Spoiler alert: all other things aren't equal, but we can't tell from his experiment what's causing the (very small) difference, or if there was too much error in his method.

I mean I appreciate his demonstration of the process and the curiosity about the world - I have a question and a guess, let's test my guess! More people in the world could stand to do that rather than just assume they're so smart their guesses must be right all the time.
posted by ctmf at 9:05 PM on April 13

Will students of history say this was the moment we reached peak flabdablet?

History is a treadmill and its peaks are illusory.
posted by flabdablet at 9:42 PM on April 13

I suppose my response was that Mould's experiment doesn't really tell us anything about a person on a treadmill, that the whole methodology is not especially helpful, that you can't reason about what is physically harder for a person from first principles. People are not magic, and you may be able to draw a wide envelope of physical possibilities, but we operate well inside that envelope and our difficulties do not correspond to simple physical properties. A toy car does not move like a person moves or use energy in the same way.

In the YouTube comments, someone claims that sports scientists have found that you need to incline the treadmill up a few degrees to equate to running on level ground, because the treadmill imparts backwards motion to people's feet, making it easier to perform the 'cycling' motion on a treadmill than on a road. I don't know if this is true, but it is incredibly obvious that if Mould actually wanted to learn about the physics involved it is physiologists that he should be talking to.
posted by Ktm1 at 9:59 PM on April 13 [2 favorites]

I appreciated the journey, I guess. And then the comments here.

Mould clearly made a few errors. The null and test hypotheses were not clearly stated in the video before the results were given. As he retroactively explains it, the result was closer to his test hypothesis (energy equal) than to the null hypothesis (ramp easier by the amount of potential energy at a higher elevation) but the order of presentation (and maybe this mirrored the order of his thought) and the flat earth tangent made it seem like he was in fact put in the position of "being a flat earther" by not coming up with exact equality. Second, of course, the main factor he identified: different road materials. Finally, a third thing that occurs is both that the slope they built from plywood might not be exactly straight (it's likely a catenary, but by how much?) and that the treadmill's running surface might not exactly parallel what is under it, or that the treadmill might further displace the plywood "fixed slope" due to its mass.

The woman who repeats the experiment should LFMF, where M is Mould.
posted by the antecedent of that pronoun at 9:01 AM on April 14

> it still absorbs a lot of the shock of your footfalls. Roads don't. I am sore today, and I haven't had any leg soreness in years and years.

Yes a lot of it is the irregularity of the surface and all that. You need to gain strength and flexibility in your lower extremities - as well as everywhere else - and there is a very significant balance and coordination thing that goes on as well.

Little bit like running on a track vs cross-country, although maybe even more so.

Bicycling indoors/outdoors is much the same.

One year we managed 90 straight days of snow here, so I went cross-country skiing every day rather than my usual indoor cardio and such. End of that time I did an annual thing where we spend the whole day walking around on marble floors in bad shoes etc etc and usually by the end of that, my feet would be killing me.

At the end of that 90 days of cross-country skiing outside, though, finished that whole day and felt good as new.

You can ride a nordic track hours every day if you like - it just doesn't have the same effect as balancing your way across real, very uneven and slippery, ground.
posted by flug at 1:41 PM on April 14

The fact that running on a treadmill at 10 degrees is much, much harder than running on a treadmill at 5 degrees (or flat) indicates that something is going on.

In the YouTube comments, someone claims that sports scientists have found that you need to incline the treadmill up a few degrees to equate to running on level ground, because the treadmill imparts backwards motion to people's feet,

A slight incline is suggested, but the given reason is all wrong. The argument for a slight incline is that it compensates for the lack of air resistance on a treadmill. This Runner's World article, which confirms my biases and thus is correct, says that distance runners are going to use 2-4% of their energy overcoming air resistance (for cyclists it's a rather stunning 70-90%). 2-4% doesn't sound like that much, but if you are running a mile, 4% of your 10 minute (say) pace is 24 seconds and that sounds like a hell of a lot.
posted by It's Never Lurgi at 3:08 PM on April 14 [1 favorite]

At the end of the run, the person on the hill will have gained a net amount of gravitational potential energy and the treadmill runner will not. Where did the energy go? The treadmill runner actually loses the energy gained after every step when they fall back a bit on the treadmill, such that their net energy gain is zero. But the runner does no work when they lose the potential energy on the treadmill, so total work done by the runner on the hill and the treadmill are the same.

If you think about it, this is exactly the same reason for why the treadmill runner is climbing an incline but gains no height, and ends up at the same height where they started. They gain and lose equal amounts of both height and potential energy with every step.
posted by grog at 12:45 PM on April 15

Another way to think about it is if the hill runner goes up a circular ramp, such that they end up where they started but with additional height. Then they jump off the end of the ramp, doing no further work, and end up exactly how they started. The treadmill runner does exactly the same thing, except they break the "total fall" up into a bunch of little falls, one per step.

Thinking this way it occurs to me that maybe the treadmill is somewhat harder, because the hill runner can ignore the fall (from a height that would almost certainly kill them), but the treadmill runner needs to absorb the energy of that fall, albeit over time.
posted by grog at 1:44 PM on April 15

the runner does no work when they lose the potential energy on the treadmill

Within a reference frame where the height of the treadmill apparatus as a whole is fixed, the runner still does work on each step, with much of that work being applied to the treadmill belt and the rest lost as heat due to friction between belt and shoes. The work applied to the belt will be, to a very good first approximation, the same as the work done on a comparable hill runner's body in order to increase its gravitational potential energy.

It would be pretty easy to set up a demonstration of that: just measure the total electrical energy consumption of a flat treadmill vs the same treadmill on an incline, each with the same runner maintaining the same belt speed for the same experiment time.

My confident prediction is that inclining the treadmill will always make it consume less electrical energy than running it on the flat, some of the energy required having been supplied instead by the runner in the inclined case. I would further predict that the amount by which the inclined case's electrical energy consumption is less will be a close match to the gravitational potential energy gain expected for a hill runner on the same simulated incline. The second prediction is based on the assumption that the treadmill's assorted sources of internal energy dissipation won't be affected much by the incline it's operating on.
posted by flabdablet at 3:56 PM on April 15