I agree.
I also want to add, and I know I am going to get flack for this, but it is how I feel on the matter. I've found it hard to know whose research to trust. There are artifacts, ruins and maybe some text to go with, but the truth is researchers are speculating. They were not there at that time, records were usually not kept, so a lot of what you are reading is assumption from clues they have put together.
You can generally tell a lot, frankly, by doing a little research. For instance, look at this book:
The Mysteries of Mithras: The Pagan Belief That Shaped the Christian World by Payam Nabarz and Caitlin Matthews. If you look on Amazon
http://www.amazon.com/dp/1594770271/?tag=absolutewritedm-20
You will note it's by a publisher who only publishes neo-Pagan and new age books. You will further note, with a little research, that one of the authors is a "revivalist" -- neither author has academic credentials, and, moreover, while Nabarz is Persian, neither Narbarz or Matthews are fluent with Latin.
This is essentially a book for people who want to re-invent Mithras worship, and the description is filled with sweeping assertions about the connection of Mithras and Christianity. Scholars don't do that.
If you look at Matthews, she writes for Neo-Pagan publishers on a variety of subjects, mostly though, about things Celtic. She is equally lacking in the language skills you need to deal with primary sources--such as they are--about Mithras in Persia, and the Mithraic cults of the Romans.
Language is a big clue; if authors are writing about something that has ties to a particular language, then they need to know that language in the form used in the primary sources.
Moreover, a publisher like Inner Traditions or Llewellyn, ideal if you're interested in Neo Pagan practices, are not at all reliable scholarly publishers like Oxford or Columbia, or Chicago or any number of other university presses, or a reputable mainstream publisher like DK or Norton, or Harper Collins, etc.
Also: If the author is long dead, like Frazer, and if the author does not cite sources specifically, and if the author is not cited by other reputable authors--all of which you can tell via Internet searches, they're likely not the best.