Since I have no inside into the code, it is hard to tell, but I observed a statistical deviation that is beyond any reason when it comes to replacing starting cards.
So I made a deck with only 1 copy per card, and played it extensively. The amount of times I received exactly the same card I selected to replace was insane. It should be 1/28 but it was nowhere near it.
Maybe there is a bug or just subpar implementation of the random seed or something like that?
Or maybe I just got a really unlucky streak for a Long period of time? Just wanted to bring this up for discussion.