While I agree that a proper, unbiased study with a good statistical sample is the ideal basis for any assertion, I believe that professional experience such as CPA's is a close second as long as that professional is addressing an area where they have deep experience. Personal opinion without deep experience is a distant third, and should be qualified as just that: opinion.
Here's an example: Among many other things in my career, I spent 2 years setting up data privacy standards for a company that handled several billion data records with private information each year across many industries and organizations. So, I saw a LOT of data points. If a question of data privacy implications were to come up, I could render a very qualified opinion based intuitively on my experience. And I think that's what
@cpa is doing. I'm guessing he has seen into the finances and personal histories of hundreds (maybe thousands) of clients, so his opinion in that area - though not as reliable as a comprehensive study conducted by an unbiased researcher - might be pretty darned reliable compared to my opinion in that area... though I would be more reliable in areas I've been immersed in.
Then, to take it a step further, we have the issue of proper use of statistics and the issue of bias or incomplete information in studies. For example, I recently read a Forbes article quoting an unattributed study saying the majority of Model 3 reservation holders are coming from Toyotas. The journalist" then went on to characterize Toyota owners, and assert that Tesla would have trouble living up to their quality and service expectations. Now, this was from a dodgy author that I do not trust. But putting that aside, he may have a point if these are predominately corolla or Camry owners. However, isn't it a different story if they are Prius owners? So, here's a supposedly data-based article with incomplete information and a biased author.
So I'm always happy to hear educated opinions if credentials are explained. Even ok with non-educated opinions if qualified as such. And, always suspicious about studies because they are often biased or misused.