It's useless to the degree that it generally devolves into a ____ measuring contest. Fill in the blank on your own...
Splitting hair is useful in one regard: It shows you how easily the apex of your edge will cut into the very narrow diameter of your hair. Hair is also a good test material because it's about as hard as copper at that diameter--or so I've heard. In any case, there's just not a lot of material that is as thin and pliable as hair is, so an edge has to be very keen to actually bisect it, rather than simply push the hair aside or skate off of the surface.
Past that it doesn't do much to tell you how sharp an edge is, because like you said hair varies so much. On top of that though, your hair is different from one place on your body to the next, and the humidity and moisture content of it changes every day, etc. There's just a ton of variables in that realm to where I think that once people get into the semantics of "tree toping" vs "hair whittling" vs "hair hanging tests" and all this, that they're really just getting far too concerning.
In my experience, an edge that is hair whittling... That is, will carve little splinters of a hair held outstretched and unsupported like a stick. This type of edge is going to degrade from that level of sharpness immediately. There's no point in trying to reach this level of sharpness, because once you've shaved your arm a couple of times, cut up a few pieces of paper, it's not going to be as sharp. That's just my experience working with a lot of mild steels, or steels with large carbides like S30V though.
Anyway, I think I said it somewhere once before... But I think that all this talk of "hair splitting" and "hair whittling" and "tree topping" and "hair popping"... It's all up in like the 99th percentile of how sharp one can get an edge, and there's not even that big of difference in practical use between any of them. Once you get an edge up to this level of sharpness, unless you're actually shaving with it, then the blade geometry behind the edge is going to be the bottle-neck. You can take an edge from sharp enough to shave, to so sharp it will make the air bleed, but it doesn't mean it's going to cut through anything you actually need to cut through any better. Unless you're talking about razors and actually shaving, there's no real discernible performance increase when going from one level of hair-splitting sharpness to the next.