If you have ever been genuinely depressed and someone has suggested you go for a run, you will know the very particular flavour of rage that produces.
It’s not that the person is wrong, exactly. It’s that they’ve missed the entire nature of the problem. Depression is not a shortage of good ideas. You already know that exercise is supposed to help. You knew that before they said it. The difficulty is that the illness’s central symptom is an inability to do the things that would help, and being told to do them is a bit like advising an insomniac to fall asleep.
So I want to try to write about this properly, because I think the standard version of the conversation is useless in both directions. On one side you have people saying exercise fixes depression, which overstates it to the point of cruelty. On the other you have people dismissing the whole idea as toxic positivity, which throws away something that genuinely helped me and quite a lot of people I know.
The truth is somewhere in the middle and it’s more interesting than either camp.
What the evidence actually says
Broadly: exercise has a real, measurable effect on mood, and the effect is bigger than most people expect. In mild to moderate depression it performs surprisingly well in trials — comparable, in some analyses, to what you’d expect from therapy or medication. For anxiety, the effect is also solid and appears to work through somewhat different mechanisms.
That’s a genuinely remarkable finding for something free with no side effects to speak of.
But there are two important asterisks, and they’re the ones that get dropped when this gets turned into a headline.
The first is that the studies are mostly on people who did the exercise. Which sounds obvious, but the entire practical difficulty lives in that gap. An intervention that works brilliantly for the people who can start it, and is nearly impossible to start for the people who most need it, is not as useful as its effect size suggests.
The second is that “comparable to medication” is a claim about *averages across mild to moderate cases*. It does not mean exercise is a substitute for treatment in severe illness, and I want to be unambiguous about that because someone somewhere will read the encouraging version and stop taking something they need. If you’re seriously unwell, this is an adjunct. A useful one. Not a replacement, and not something to negotiate with yourself about instead of seeing a doctor.
With that said, the effect is real, and it’s worth understanding why.
Why it works, as best anyone can tell
Nobody has a single clean mechanism, which is usually a sign that several things are happening at once.
The endorphin story you’ve heard is probably the least important part, and “runner’s high” is a genuinely rare experience that most people never have. More promising are the changes in brain-derived growth factors — exercise appears to promote the health and growth of neurons, particularly in regions involved in mood regulation, and some of the same changes are seen with antidepressants. There’s also a decent case around inflammation, since chronic low-grade inflammation shows up in a lot of depression and exercise reduces it
But I suspect the mechanisms that matter most day-to-day are the unglamorous behavioural ones.
Exercise regulates sleep, and disturbed sleep both causes and worsens almost every mental health problem there is. It gets you outside, into daylight, which does things to your circadian rhythm that matter more than people credit. It’s an interruption to rumination — it is genuinely hard to spiral about a conversation from Tuesday while you’re out of breath. It puts you around other people, incidentally, without requiring conversation. And it provides evidence, at the end of a day when you may have no other evidence, that you did a hard thing and completed it.
That last one shouldn’t be underrated. Depression is very good at constructing an argument that you are useless. A completed session is a small, boring, factual counterexample, and it’s difficult to argue with.
Anxiety and depression are different problems
One thing that’s rarely separated out: exercise seems to help these two in quite different ways, and knowing which you’re dealing with changes what’s useful.
For anxiety, a lot of the benefit seems to come from something like exposure. Anxiety involves a body in a state of physiological alarm — racing heart, shortness of breath, adrenaline — and a mind that interprets those sensations as evidence that something is terribly wrong. Hard exercise produces the exact same physical sensations in a context where they’re obviously benign. You get a racing heart and shortness of breath and nothing bad happens, repeatedly, and over time the association weakens.
That’s a real mechanism and it’s why hard, breathless exercise often works better for anxiety than gentle exercise does.
For depression, it’s more or less the opposite. Intensity is not the point and can be counterproductive, because the barrier is activation, not fear. What matters is doing something at all — the smallest possible thing, frequently. Gentle, outdoors, ideally rhythmic and undemanding. A walk beats a session you dread and skip, by an enormous margin.
If you’ve tried exercise for low mood and it didn’t help, it’s worth asking whether you tried the version that fits the problem. Plenty of people have set themselves a punishing regime while depressed, failed at it, and added the failure to the pile.
The activation problem, seriously
So: you know it helps, and you can’t make yourself do it. This is the actual question and everything above is preamble.
I don’t have a clean solution, because I don’t think one exists. But a few things are true.The bar has to be humiliatingly low.** Not “go to the gym.” Not “go for a run.” Put your shoes on and stand outside the front door. That’s the entire commitment. If you then come back inside, that counts, and it counted. This sounds like a trick and it is, but it works often enough to be worth using, because the resistance is nearly all located at the start of the sequence rather than distributed through it.
**Waiting to feel like it doesn’t work here either.** In ordinary circumstances, motivation follows action rather than preceding it. In depression that’s even more true — the feeling of wanting to do things is often the *last* thing to return, arriving well after the doing has resumed. If you’re waiting for the desire to come back before you start, you may be waiting for the outcome to arrive before the cause.
**Outsource it to someone else.** The single most effective intervention available is a friend who’s coming at eleven whether you like it or not. Depression will beat your own intentions in a straight fight almost every time. It’s much worse at beating a social obligation. If you can arrange for one person to expect you, that will do more than any amount of resolve.
**Lower the target, not the frequency.** Ten minutes, most days, beats an hour on Sunday. The point is the interruption and the rhythm, not the training effect.
**Be sceptical of the guilt loop.** The genuinely unhelpful thing about this whole conversation is how easily it converts into another thing you’re failing at. If you can’t do it, you can’t do it, and that’s a symptom rather than a moral failure. Adding self-criticism to depression is like adding petrol to a fire — it’s the one thing guaranteed to make it worse, and it’s what a lot of well-meant advice accidentally does.
## What it is and isn’t
I want to be honest about my own experience because I think the overstated version does harm.
Exercise didn’t cure anything for me. The bad period I had in my early thirties ended because of a combination of things — time, a change of job, talking to someone qualified, and eventually a proper night’s sleep — and training was one input among several. On the worst days it did nothing perceptible at all. I’d walk for an hour and come back feeling exactly as flat as when I left.
What it did do was raise the floor slightly, over months. The bad days were marginally less bad. There was usually a window of an hour or two afterwards where things were quieter in my head. That’s not a cure and I’d never sell it as one. But when you’re in it, a marginally less bad day is not nothing. It’s actually a great deal.
And it did something else that took me longer to notice, which is that it gave me a relationship with my body that wasn’t adversarial. Depression and anxiety both tend to make the body into either a problem or an irrelevance — something that’s letting you down or something you’re dragging around. Training gradually made mine into something that could do things. That shift is hard to quantify and I think it mattered more than any of the neurochemistry.
The thing to take away
If you’re well, and you want to stay that way, this is one of the strongest arguments for regular movement that exists — stronger, honestly, than the arguments about how you look. Build the habit while it’s easy to build, because it’s a much better shield than it is a rescue.
If you’re not well: it’s worth a try, in the smallest possible version, alongside — not instead of — actual treatment. A ten-minute walk outdoors is a genuinely reasonable place to start and requires no equipment, no plan and no membership. If it does nothing, that’s information, not failure.
And if someone tells you to go for a run in a tone that suggests you simply hadn’t thought of it, you have my full permission to be annoyed about it. They’re right about the mechanism and wrong about everything else.
