The drag is bad enough---somewhat chaotic, hard to compute due to atmospheric variation---but I bet they could give a reasonably good estimate as long as it rem

The drag is bad enough---somewhat chaotic, hard to compute due to atmospheric variation---but I bet they could give a reasonably good estimate as long as it remains in one piece. The problem is, the moment it starts to break up, all bets are off, because we don't know which pieces will break off first, how large the respective sub-pieces will be, and so on; and each break could send one or more of the pieces into a tumble and/or heat up some other part more than another, which affects chances on further breakage, etc, etc, etc. Basically, once it starts breaking up, the system goes from "sort of unpredictable" to "wildly chaotic".