I have never seen anyone with a toupée that I couldn't spot. Therefore, I can always tell when someone has a toupée.It's a fairly obvious fallacy, wouldn't you agree? Just because you haven't seen anyone with an unspottable toupée doesn't mean there isn't anyone. In fact, it would be impossible to spot an unspottable toupée, pretty much by definition.
If you limit yourself to people with toupées that you could spot, then of course it will seem as if you can spot every toupée. If you really want an accurate assessment of your toupée-spotting abilities, you need a way to also include people with toupées that you didn't spot. It's impossible to do so under everyday conditions. Obviously, you're not going to go around asking every person you meet whether they are wearing a toupée. They're not all going to answer sincerely, especially the people who are embarrassed to admit wearing a toupée.
As an example of a real-world use of the toupée fallacy, consider stereotypes, specifically those of gay people. They're stereotypically flamboyant and effeminate in a really obvious way. Of course, some gay people really are that way (and there is nothing inherently wrong with that), but as an overarching generalization, it's quite inaccurate. Allow me to claim (without any empirical evidence) that this stereotype is at least in part due to a toupée fallacy. Most people rarely spot a gay person that isn't really obviously gay. They often assume that the gay people they can spot are representative of the entire group.
Atheists, too, are really hard to spot, unless they make it really obvious in some way. For example, they will usually only be spotted if they choose to argue about religion. Is it any surprise, then, that atheists are stereotyped as being angry and argumentative?
This is one reason why it is important for people to "come out", both in the case of sexual orientation and in the case of atheism. Only when everyone in the minority is visible does the toupée fallacy disappear. (Of course, negative stereotypes would exist even without the toupée fallacy, as is the case for Black people and women. But I hypothesize that the toupée fallacy influences the magnitude and direction of stereotypes.)
If there's a larger lesson here, it's that our everyday experiences can lead us astray. Our everyday experiences are not a representative sample of reality. Particularly in the case of invisible minorities, what you see is a small, unusual subset. It is useless as evidence. If you think surveys and statistics have problems (and they do), just imagine the problems you encounter when relying on anecdotes and other ordinary life experiences.