"[F]inance" or "think tanks" is kind of broad. You probably mean small- to medium-sized hedge funds in finance (or their equivalent prop desks, as far as they s

"[F]inance" or "think tanks" is kind of broad. You probably mean small- to medium-sized hedge funds in finance (or their equivalent prop desks, as far as they still exist, in investment banks) because otherwise I can't think of finance as a place for creative natures. Most bank divisions and asset managers have set rules and procedures for everything and force you into using their software (environments), like Excel or Matlab or C# or Windows. I'm not saying those are bad in general, it's just I myself hate them and there's absolutely no way I could work for an operation that uses them, which precludes me from practically 98 percent of firms in the industry (ok, I pulled that one out of a hat). Also, although I generally agree with Merton's point that "first you have to show you can do it like them, then you can show you can do better", I think more and more this applies only outside of "paradigm shifts" in Thomas Kuhn's sense. During such shifts, or to propel them, one may do better by totally disregarding the viewpoint of the old guard and "just do". EDIT: Of course, one would still need to have thought enough about the "old" way of doing or explaining things that one would know why the "old" way failed, in a fundamental way. For example, I think in economics and finance, the fundamental failure is not so much in the statistical assumptions (that is actually a pretty cheap shot to make, theory-wise, Herr Taleb!) but in aggregation.