Your conclusion sounds very plausible. That could be happening now. I don't know. But a couple of things: if you're talking about phase 2 candidates that are getting close to moving on to phase 3, you're talking about substances discovered a long time ago.
From time of discovery to first clinical trials is probably something like 5 years on average, but certainly not less than 3 years. Then the phase 1 trial, which itself might only run for a few months, still takes around 2 years when you factor in setting the trial up, recruiting, running and analyzing results. So you're 5-7 years from discovery when you start PLANNING your phase 2 trial. By the time it is planned, recruited and had a chance to run for a while, that's another 2-4 years, so anywhere from 7-11 years from discovery and you're still in phase 2. That 7 year best-case scenario actually seems too short to me too...but anyway my point is that the promising stuff you're looking for is stuff that isn't based on the new knowledge we've acquired over the last 10 years.
Another question I have is: how does anybody not working for a research lab somewhere even know what researchers are focused on right now? It takes a long time from when they do the work in the lab until it is published and we all hear about it. There seems to me to be some evidence that neurodegeneration and grey matter involvement are being researched quite extensively today. And when it comes to actual drug development research, the researchers generally need a target to go after, so if a theory sounds promising, but they haven't characterized a promising target for a drug, then they can't develop a drug.
I think we tend to over-simplify the MS research world when we say that people just believe the autoimmune model and that doesn't work. I think the most common model for MS these days is something like: an embryo is conceived and it has the genetic makeup that makes it possible to get MS. In the womb certain factors might influence development towards MS (mothers vitamin D status? other?). In childhood, environmental factors such as vitamin D status, diet, pathogen exposure or other hygeine factors influence whether a genetically pre-disposed person develops MS. Most researchers would probably say that you need some kind of herpesvirus exposure (EBV, HHV6a etc), possibly multiple viruses, and possibly there's a retroviral element.
I think all of these things are considered part of the main-stream theory. Is there any information out their to conclusively prove that this general picture is right or wrong? I don't think there is.
I think there is a missing link in the process though. All of these elements may be involved in triggering the disease, then we have a lot of information about how the disease progresses once it manifests itself, but I don't think we don't really know for sure how the disease process starts (the good old Inflammation vs. Neurodegeneration thread).
Anyway, what all this points to for me is a much messier, more incremental research world than your characterization and so I don't think things just shift en mass. Different people and organizations focus on different things at different times. I agree that companies especially get caught up in trying to make money, so they go after a target that has already been addressed by another drug, but I guess that's a separate issue. I'm not saying that I think MS research has been conducted perfectly and that there aren't people out there just trying to make money or advance their careers, I'm sure there are.
Anyway, having said all that, you may be right Frank, and we'll know in a year or two when we look at the phase 3 trial in progress. If there are fewer than there are today, then you were right...and I'll buy you a beer...