And what would have happened if Deoxys-N was broken, but was not banned as quickly as it was? People would resort to using it in order to win, and it would rise in usage. This is a poor example you've given, particularly when something is banned quickly within a meta still in its infancy.
I don't presume to know what UU is like, or what the reasons were for banning Gothitelle, and how that relates to its usage. That being said, it could suffer from the same issue I presented with Deoxys-N. If a meta has relatively stabilized and has had a sufficient amount of time, any problematic suspects will inevitably make their way up the standings as people resort to using them to win unless they're preemptively taken care of.
Personally I don't agree with the idea that statistics should be ignored, especially when we base our tiers around the very idea of usage. The premise is that OU Pokemon will be OU because if they are better than average, players will naturally gravitate towards them as they are effective and competitive options, and they will rise in usage. Broken Pokemon are just an exaggeration of this very idea, so it makes sense that with enough given time, this will happen with them too. Obviously this shouldn't be the only consideration, or even the most important, but I think it is valid information to take note of.
If anything, something like Deoxys-S being #69 in usage implies it has yet to definitively prove whether or not its worthy of a suspect ban. It doesn't mean it doesn't deserve a ban, but what I think it does mean is it needs more time on the ladder for people to make a proper decision.