What ZAP advocates studiously ignore is that there is more than one Zero Aggression Principle, depending on what set of aggressors you have in mind, what constitutes aggression (http://libertarianmajority.net/free-variables-in-libertarian-theory), whether you think of zero as an implementable rule or an ideal asymptote, and (if the latter) whether you think that asymptote must only be approached monotonically.
Defining "libertarian" as fealty to one's own interpretation of ZAP is silly and tendentious. Words have meanings. The dictionary tells us that, to competent speakers of English, "libertarian" means "one who advocates maximizing individual rights and minimizing the role of the state". I accept that definition of libertarianism, even though I think the particular school of libertarianism that I favor is arguably superior to the other schools in terms of maximizing individual rights. However, I'm not so doubt-stricken about that superiority that I need the LP to officially reassure me of it, or that I need to define other schools as un-libertarian in a desperate attempt to avoid clear-eyed comparisons of them.
"Direction" versus "principle" is a false dichotomy. The BTP platform is entirely directional, but even it is not unprincipled. The current LP platform is "directional" only in the sense that it leaves some ambiguity about how much of the state would remain in Libertopia. I heartily agree that directionally defining "more libertarian" as "less government" is not optimally principled. It's indeed always more libertarian to minimize the role of the state, but not every function has a minimum of zero.
I've become steadily more radical during my involvement with the LP. I began as a Friedmanite consequentialist minarchist, but now I'm a radical geolibertarian (http://ecolibertarian.org/manifesto). I reject the quaint idea that radicalness can be measured as the inverse of the size of the bumper sticker that your worldview fits on. In particular, I think that the anarcholibertarian interpretation of ZAP is insufficiently radical in three distinct ways. First, it myopically focuses on how much aggression gets committed in your name, rather than how much aggression gets committed. Second, it institutionalizes the non-policing of certain kinds of aggression -- such as aggression involving non-excludable common resources, where negative externalities are too distributed and cumulative to be policed privately. Third, it completely denies the existence of an entire class of aggression -- monopolistic appropriation of ground rent. Any one of these considerations would lead me to regard Rothbardian ZAPstentionists as insufficiently radical -- i.e. as insufficiently opposed to aggression. However, I'm happy to work with suboptimal libertarians like them, and I don't need the LP to give me an intellectual security blanket by endorsing my school of libertarianism as better than theirs. I've got a nanny state to destroy, and I'll take all the help I can get.