Maybe predators are wary of rattles and don't step on the snake. Or maybe the rattle diverts attention from the snake's head.
The point of a rattle, as I understand it, is that it's metabolically expensive, and time consuming, to produce poison. A snake that can chase off a dozen threats a day by wagging its tail is much better off probability-of-producing-offspring-wise than one that can only bite and poison three threats before being left defenseless for a few days.
It does leave me wondering what benefits the intermediate mutations provide though, since going from a normal snake tail to a rattle seems like it would take more than one step.
Interesting. That corresponds remarkably well with the best advice I've ever gotten for estimating software schedules: take your best guess as to when you'll be done, then double it.