For the case of a black hole, the solution you get by reversing time describes a white hole., which is again an entirely valid solution to the Einstein field equations.
Also if you invert time, black hole behaves differently.
Suppose you have a video that follows the moon and the Earth from some distance but that no other celestial bodies are visible, for simplicity.
If you reversed the orbit of the moon but didn't change the rotational sense of the Earth (sun rises in the east) then you could distinguish a video of that system from the original true-to-life video played backwards. Only if you ALSO reverse the rotational sense of the Earth (sun rises in the west) do you have a system that you can't distinguish from a backward true-to-life video.
The same thing is happening in your scenario; if you want to apply time reversal, you'd better time reverse EVERYTHING relevant. Here, what you're missing is that you also need to time-reverse the black hole, which produces something called a white hole [0]. We have never observed any white-hole like object. But the point is, if you take your time-reversed object at the just-inside-the-bh-event-horizon radius away from the white hole's center, you'd get the correct reverse simulation.