I can't find reliable stats overall, but main types of therapies studied are quite effective.

But there are a few problems with some of the findings, in my opinion.

1. Just because the research says a therapy is effective it does not mean it is truly effective in a way that is meaningful to a particular person. They might use specific criteria, like reduction of certain symptom or being able to work or whatever, as signs of success. But the patient might see those as improvement but not feel truly helped by therapy, as you say.

2. Some studies don't use good methodology. Sometimes people with worse symptoms actually drop out of the study. The people who complete it are not the same as people who drop out. They might be more determined, bigger believers in the modality used, etc. In addition, some mental health issues are time limited. Like some episodes of major depression lift on their own after a number of months (even without therapy). And so on.

3. Many different kinds of therapy have similar results. Which is why some research now indicates (not a consensus though) that success is more about the therapist and client match and connection, client engagement, etc, than about a particular modality.

In my view therapy is not helpful for all people or with all problems. Sometimes you can't know that till you've gone for therapy. Then again, sometimes people go for therapy when they give up hope of finding any other way to better their situation.

Interesting observation, and a not so uncommon criticism of psychotherapy, as enforcer of societal status quo.