I take this to mean therapy for depression (particularly for men) is barely effective at all and exercise is not quite as barely effective at all.

If therapy for depression were a pill, I'm not sure it'd demonstrate enough efficacy to get approved.

Anecdotal obviously, but from my experiences and those in my friend-group, therapy seems actively harmful in many circumstances. Therapy can be an enabler of bad behavior, helping to build a new fancy framework of excuses just as easily as it can help you land somewhere better. How long has it been since you've seen something bad (maybe it was just slightly selfish, maybe it was something worse and chronic) waved away as "self-care"? Or if you've lost faith in people or communication generally, then you're not going to feel better by going through many rounds of therapist mismatch and at some point you've aggravated problems rather than fixing them. Fortunately for the industry, measuring the harm is just as difficult as measuring the benefits. The advice to "just exercise more" may not help, but at least it really can't hurt.

Psychotherapy is similar to pharmacotherapy in effect size but longer lasting after discontinuation. Given that the latter has been approved, it's safe to assume psychotherapy would be approved if it were a pill.

I don’t think that conclusion follows. These results don’t mean “barely effective.” They mean that, on average, people who get therapy or exercise end up noticeably better off than people who don’t. When researchers translate the results into intuitive terms, therapy typically moves the average person from the middle of the pack to around the top 30 percent, and exercise to roughly the top 35 percent. That’s the same general magnitude of benefit seen with many approved medications, including antidepressants themselves. Depression trials have large placebo effects, which makes the numbers look smaller, but that doesn’t mean the treatments are doing almost nothing.

I think the major problem we have is we are lumping depression and anxiety into the same category.

Some people probably do have issues quickly resolved by SSRIs. Others are magically fixed by bupropion while it spikes anxiety in others. Others have major trauma that they have to work through and many therapies (like internal family systems therapy, as one example) are great at handling that. Others are stuck in cognitive distortions and merely learning about them and handling them (through cognitive behavioral therapy) can be life changing.

But if you have major trauma in your past, exercise is probably not going to do much. If you are heavily overwhelmed and your body is stuck in perpetual flight or fight) exercise and meditation might be a giant help.

But right now the practitioners are aware of this, but it's really hard to double blind test these divisions until people can do the analysis first. And at that point, you've basically already started therapy.

The problem is that the same is true for many medications and other treatments, but if you're in the "lucky" minority (a pretty large one) for whom there's some relief... it's worth the try. It's not like choosing bleeding and emetics instead of chemotherapy for cancer, it's just the best we have right now to treat a complex system that we barely have the earliest understanding of.