Personally while I appreciate what PyPy are trying to do, for a lot of real world use cases you can get really impressive performance out of python using the right libraries (see: the entire Python scientific computing stack).
You get impressive performance out of python the same way you get impressive performance out of any dynamic language (with few exceptions): make sure your hot paths are executed by code written in C. The nice thing about PyPy is that it makes your pure Python code a lot faster. For instance, I wrote some log-parsing code in pure Python that got a 2.5x speedup when ran from PyPy.
Personally while I appreciate what PyPy are trying to do, for a lot of real world use cases you can get really impressive performance out of python using the right libraries (see: the entire Python scientific computing stack).