If you have real numbers, use them.
If you don't, this all sounds a lot like the sort of "numerical" support for anti-piracy measures made by record companies. The interconnection between the marketing role of a free option and its numerical effect on sales is going to be highly nuanced.
Further, there are clear-cut reasons that "Big customer databases" are valuable even if "they’re [not all] ponying up cash."
The first, discussed elsewhere here, are network effects. Dropbox gets more and more valuable for paying users every time a free user signs up because it makes it all the more likely that the next person they need to share a file with will be able to easily handle a DB share.
Second, building a big user-base of devoted users, even if a large percentage are free users, potential builds a barrier to entry for any possible competitors – thus protecting the revenue from the existing paid users and encouraging future paid users.
Still, I believe that underlying your whole post is the important point that you must have a solid business plan and understand just how you are going to drive profits – be it freemium or otherwise and that's a very important point, so thank you for calling it out.
In terms of making people pay to enter, I think it really depends. As we explained, if you're not in a position to truly move folks from a value limited free taste of a product to a premium tier, then I'd say you're better off growing slowly before opening up the floodgates. Take a look at Wistia or MailChimps approach. Of course, there are great counterexamples, as well.
Thanks for the feedback!