I've seen great work coming from just about every software package that's out there. I think it comes down to a personal preference and the amount of time spent on learning the software and then learning how to exploit the strengths and mitigate the weaknesses of the software (always an ongoing process, for me).
It also depends on the type of work you intend to do. I'm 90% conceptual only and enjoy the freedom of not being tied to a parametric structure tree. Are there times where it would come in handy? Sure. Would I want to deal with a parametric tree every time I wanted to build a quick concept? Nope.
I have a parametric modeling tool at my disposal, SolidThinking Evolve, which is pretty cool, but I don't use it too often. I prefer the tools and the way Rhino functions. I've used Evolve, and while I'm no expert with it (that's part of the problem, right there), it's nice at first having the parametrics, but when something breaks the tree, I find I'm no better off than having done the same model in Rhino.
We have a CAD group that uses SolidWorks. There are things that are easier for them to do and there are things that are easier for me to do in Rhino. It all depends. The one thing that I've never experienced in Rhino and I hear it often from those using SolidWorks, is the lag of updating the structure tree on large or complex models. On some of the dense and complex surfaces we sometimes deal with, it's literally a "go have your lunch and come back" situation. Granted, that's an extreme example, but it shows that there are pro and cons to having a parametric structure or not.
For me, the History in Rhino is just about right. When I also figure out the best way to model in Rhino, to emphasize its strengths, I find it great for what I need to do. If you throw in T-Splines (also using history) and the potential of Grasshopper for helping to create commonly modeled components, there is a huge potential. It isn't necessarily right for everyone in every circumstance, but it's pretty powerful.