HallsofIvy's post did not introduce the combination operation. His post would directly have led to:
p(n) = 1 - \frac{(365)(364)...(365-n+1)}{365^n}
which clearly holds for n \leq 365. It also holds for n=366, because one can say the 366th person has zero "choice" of birthday to avoid a collision. However, when n > 366, the last terms of (365-n+1) stop making sense, because, for example, the 367th person does not have "-1" choices of birthday, etc., he simply has no (zero) choice. Even though the expression will evaluate to the correct answer (one) for n > 366 because exactly one of the terms in the product is zero, the expression is still not correct mathematically.
The next commonly taken step, which involves simplifying the expression and recasting it in terms of combinatorial notation, is what you're referring to. Of course, based on the definition nCr = 0 for r > n, this expression would hold for all n, but the fact remains that this is not a mathematically sound way of deriving that formula since the initial expression does not make sense for n > 366, as I explained.
Hence the usual way of working the problem, which is to break the problem up into two cases, n \leq 365 and n \geq 366. Mathematical rigour matters.
You can call it whatever you like, and it is common sense, but it IS called the pigeonhole principle.