Well, I would like to think observability, testability, and predicability are the greater part of it. You know. All those things that make creation theories so unfit in this regard. Theories of star development and continental drift theory are little more than sophisticated story telling. Sure, they have some basis in the evidence to allow rational assertions, but that's as far as it goes.
As someone who is well aware of the current body of astronomical work, I suspect if you examined the depth of the data involved, you wouldn't call it "sophisticated story telling". Far from it. You would have to reject three quarters of what we know about physics if it were simply "sophisticated story telling". After all, all most astrophysics is taking the physics we have well established on earth and applying it to other areas. Unless you'd like call most of established physics "sophisticated story telling"?