I would disagree w/ "Most academics are pretty clueless about statistics", but agree that there are opportunities there. Big Data is opening more possibilities for computer intensive statistics and the GPUs are a current tool to make these possible in a "reasonable" amount of time. For example, check out the book "Bootstrap Methods and their Application" by Davison and Hinkley - it is old at this point, but very worth reading. (Inter-library loan can come in VERY handy here. :-)
Working in Statistics generally can be interesting. I have gotten paid fairly well to have experts teach ME about their field and specialty so that I can then advise them on how to analyze their data. (If you don't know where it came from, you can't really interpret the data.) For me, it was like having a hobby I got paid for ... I even hung around long enough to get a PhD.
Finally, formal training and credentials are helpful (but not ALWAYS necessary) in Academia. Results are critical in or out of Academia. You can freelance or look for someplace that has an in-house consulting group. Getting a "client base" takes time but could be a path for you to start a freelance practice.