Yes, the quality of the dissertation research matters a lot more than the name of the university. But on the other hand, many people's research turns out to not be exceptional, but mediocre. Even people that end up getting good postdoctoral positions often have somewhat mediocre research and it's the influence/connections of their advisor or committee members that got them the job. And guess what, if you're at a prestigious university, the latter will be more likely to be maximized.
So unless you really, really think you are going to end up as one of the elite (hard to predict, I admit -- perseverance and courage to try new things, and your psychological makeup can be even more important than how sharp you are) I'd go for the prestige.
Lastly, you seem to have the idea that because you'll have a PhD that a nice, cushy quant job is waiting for you. Well, guess what. The field is super competitive now. Long gone are the days you could just walk off the street knowing a simple Black-Scholes derivation and get a Derman-type quant job. All a PhD in science, engineering, cs or math signifies to people doing the hiring is: 1) You managed to accomplish something that was difficult and took time 2) You have an above average level of intelligence in quantitative work. You'll still be competing with guys with only a Masters or Bachelors, and they still will win out a significant portion of the time because the selection criteria are how sharp you seem in interviews and your fit/personality. And after years grinding away at some esoteric field, you're not going to be that sharp anymore. Your soft skills might even deteriorate and you won't be able to explain something simply to some MBA without coming off as arrogant. The latter happens more than you might think. "PhD arrogance" is very common; you won't notice that in yourself until you interact more with regular people and then you'll realize, damn when did I become a jerk?