Reading this thread, I found myself quite surprised that Freud seems to be so disliked. I had no idea that him being a nutjob was the consensus, though this niche forum might not be the greatest sample size.
He's a nutjob to evidence based researchers when it comes to theories of the mind. Psychoanalytic theories, especially Freud's, are often unfalsifiable, meaning they aren't effective science because no observation can disprove them. Successful scientific theories must predict better than alternatives, and must make testable predictions. There's also simplicity, but that's a more informal rule.
However, this doesn't mean everything he did was worthless. His theory of the mind is similar to Plato’s theory of the soul, and Socratic philosophers have had a significant impact on modern clinical psychology separate from Freud. Psychoanalytic and psychodynamic talk therapy are still measurably effective. Large issues like generating false memories exist in Freud's methods, but many therapists now use an eclectic approach. They take what works best about different approaches to best meet their patients' needs. Certain methods work best for some people with some conditions, but aren't effective on other people or with other conditions.
Understanding unconscious trauma, motivations, and emotions still have their place in clinical psychology, even if his conception of the mind is outdated.
On a certain level it's physicists complaining that musicians talk about F# and Cb instead of waves at a certain frequency. Or physicists complaining that artists use Goethe's colour wheel -- don't you know Goethe was wrong and Newton was right? Figures that Goethe is more right about perception, though. As the map is not the territory so it follows that there's multiple sensible maps for the same territory.
Artists communicate their subjective experience, while science tries to be as objective as possible. This is why psychoanalysis has therapeutic value. Objectively, it is very effective at treating some disorders and maximizing happiness, as those results can be empirically measured. At the same time, therapists can adjust their practice to minimize known dangers with the theories. Some of Freud's ideas can give people the language to express themselves and heal, even if it's not totally accurate. Improving mental health is the important metric.
That's interesting. If accurate, it might be because manual based methods are designed to work as a whole, while fragments of certain methods only work in the context of the rest of the manual. I'd want more independent confirmation to trust the result myself though.