Is the performance difference actually still that big? I'm genuinely curious, I know that virtualization software will naturally perform a little worse than installing a system natively like via Bootcamp, but the thing is that Apple's Bootcamp drivers haven't been so great in recent times either. Parallels' virtualization on the other hand is pretty tightly optimized (obviously, since that software is their main horse that earns them money, whereas keeping Bootcamp up and running is more of a byproduct for Apple's software engineers and not something they focus their attention on). Parallels usually advertises some sort of performance boost in almost every single big upgrade, whereas the Bootcamp drivers oftentimes didn't really allow the machines to run to their full potential under Bootcamp, so much that users even come up with their own drivers to improve that (e.g. for the iMac Pro).
So the question is, is Bootcamp really that much faster nowadays than Parallels that it's worth using even though Parallels has a ton of convenience and quality-of-life features in comparison? I recently tried out the newest version of Parallels on my new MBP and was blown away by how seamless, polished and well-integrated everything is. I was originally intending on only trying it out and then installing Windows via Bootcamp, but now I'm strongly leaning towards chugging down the cost of a Parallels license and just going with that instead. Outside of performance which I myself can't really judge, Parallels seems like the vastly superior option to me at the moment (assuming of course that the cost for it is no issue).