If people have trouble separating what's true from what isn't, then the idea that people just assume it's something "normal" doesn't work nearly as well, because people won't have a solid boundary between what they know to be true and what they know to be false, and so they won't know when to assume something is real and when to assume it isn't.
For instance, if I see a zebra walking down the street, I'll assume it's real and be like "check it out, there's totally a zebra walking down the street!" If I see a unicorn walking down the street, I'll assume it's fake and be like "wow, they totally made that horse look just like it has a horn!" Zebras are real, so it's possible there really is one walking down the street; unicorns aren't, so it isn't possible. But if I'm not positive whether unicorns are real or not, I won't assume it's fake, because I don't know.
Anyway it only takes one scientist investigating one bona fide good-luck charm to create replicable proof that they do, in fact, work.
I don't buy the "they'd all be hunted down and killed" logic, either. Unusual groups do get persecuted IRL, but to the best of my knowledge*
none has ever gone so far as to attempt to completely hide their existence from the world at large, so the idea of magic users or whatever doing it is utterly without precedent, both in terms of whether people would try it and whether it would work.
(Not to seem like I'm coming down too hard on your ideas here—The Masquerade
is just a big, complex, problematic idea that gets hand-waved as "of course that's what happens" way
too often. The fact that you're considering all this is a very good sign.)