It seems to me that there are two questions at the beginning of a suspect. One, "does the pokemon overcentralize the metagame?". This is the question that is slavishly adhered to in a lot of threads like this. It is my opinion that once that question is posed, there is a sizable contingent that tends to search for ways to answer "yes" at the expense of any alternative. However, I think that this is a bit of a misleading question to base a suspect test on in this generation and metagame. From my experience, there are simply too many strong threats in any given tier, all of which become exceedingly popular through time, that you will quite literally never have a tier where the top mons are not, in some sense, centralizing. For example in the current UU climate, if you do not have an answer to Suicune, you will lose the game. It is as simple as that. Do I think Suicune should be banned? No, I do not. Similarly, if you do not have an answer to Serperior, you will lose (again, from my experience, far less so than other S rank threats in this tier).
It is easy, and a bit misleading, from my perspective to throw around a lot of calcs and jargon. Simply put, when a suspect mon is discussed, it seems to me that the theoretical situation involving that mon always assumes the worst case scenario when facing it. In this particular test, Serperior was always at 100%, always hit every move, always predicted correctly on switch ins with Glare, always just so happened to be running the HP that countered whatever was being talked about. I don't buy it, really. I think this is why suspect tests are useful, because in my playing I did not have that experience - when facing Serperiors, I could switch, predict, play around them much the same any other mon. No, I didn't always have a binary counter, but I do not think that is a bad thing. I enjoy the metagame where prediction, deception and ruse are prized over copying an RMT and using your designated switchins turn after turn. I found it both enjoyable, and not terribly difficult, to predict whether the Serp was going to use Glare, or Leaf Storm, or Dragon Pulse. And in some situations I had to sacrifice a mon to regain momentum against Serp, that is simply part of the game and is a constant in any given tier. Knowing when and where to do so is a skill that should be promoted, not an option that is derided because you didn't have a 1 to 1 counter to switch in.
So with all of this in mind, I return to that original question. Frankly, answering re: the centralization of a mon is, to me, both boring and misleading. The discussion around it in these million post long threads is obfuscating and often disingenuous (whether that be purposeful or not). So instead on the ladder I thought it'd be better to see the answer to a different questions, "is the tier more fun with this pokemon in it?". There are many mons that I think make tiers much less fun. But UU with Serperior is more offensive, faster paced, encourages prediction and creativity. All of those factors mean a more interesting, and less boring metagame to me. Simply put, you're unlikely to have a 100+ turn bore fest with a Serp on either team.
With all of this in mind, I vote no ban on Serperior in UU.