Hey, I’ll grant yours is the most commonly held position, though it wasn’t that long ago when math was considered metaphysical.
I think we’ll see more and more people arguing that empiricism is not the sole determinate of what constitutes a science. In the case of mathamatics, one can argue collective knowledge (Kuhnian paradigm), links to other scientific disciplines, and creation of new knowledge.
People do seem currently more happy labeling it as a language or tool.