Isn't it supported by observation? Whatever happened to pursuing a line a reasoning even if you it takes out of your comfort zone?
You have observational data which specifies what race is and which races there are? That is the question. The reason I'm asking is that there exists no definition of any races, and for very good reasons. Attempts to define the damn things have been intermittently ongoing for a couple of centuries. I will go out on a limb here and propose that the best you're going to be able to come up with is some weak generalisations based on the kind of common-sense thinking which cannot be substantiated.
I'm quite happy to inhabit the discomfort zone in this matter. (I know of a few interesting areas of unquantifiable matters of skill in possibly discerning a few things that are very non-PC, which is something that can be discusse behind closed doors between osteologists, forensic anthropologists et al.)
That is, as long as it amounts to more than someone demanding the concept of race just be taken for granted as unproblematic, with nothing to substantiate the claim - except possibly insinuations of everyone asking for a bit of precision and intellectual honesty is just a scardy-pants.
Argue your position, if you have one.
What is a race? How does one define one? What does it mean?
The reason for asking this is that as soon as you claim "race does something", it is "that which explains". If asked "how do we know?", race suddenly becomes "that which needs explaining".
At least the better class of 19th c. skull-measurers knew quite well, and admitted as much, as that "race" first needed to be defined, before discussion could begin about what it actually did (if anything). That was at least logically coherent and inetellectually honest. They failed of course, but that's another matter.