You know, the question about wombats is interesting and valid. But the answer is not as cut and dried as you seem to think. Let's back up a bit.
Let's assume we have a prehistoric wombat with a pouch that is right-side-up You could imagine that there are a few options for a new mother. She could deal with the problem of storing her young behaviorally. In other words, baked in instinct could be compensate for the poorly oriented pouch.
But it is also possible that she has specially developed muscles to hold her brood in the pouch while she is upside down. Now maybe this holding isn't perfect, maybe infants have a 50% chance of falling out..In this scenario, it's very possible that over evolutionary time, the orientation of the pouch changes. and the opening slowly rotates until it is completely upside down.
Now although I 've studied evolution in general, I don't know the specific answer to this. But I would guess that somebody is looking at the problem and trying to figure out exactly how the opening evolved that way. One way this could be done is through comparative anatomy.
But back to ID and why it is discounted in the world of grown up science. The difference here is that real science takes a problem like this and tries to figure out what the answer is through various means. What ID does is find something odd like this decide that the only way it could happen is for God to have done it, and then go out for pizza.
Basically you are arguing that the wombat pouch is irreducibly complex. But how do you demonstrate that it is. How do you prove that it could not have evolved. Just off the top of my head I have proposed a mechanism for such evolution. And I'm guessing that somebody with more expertise with wombats has a better answer than I do.
Irreducible complexity has yet to be demonstrated. And until it is demonstrated, ID is not a science.