Subplot Kudzu

All perfectly true. But note that your original remarks said emphasis on combat, problem solving as an aspect of resource management (i.e., I have these spells, items, hyper tech toys or powers, how can I avoid or work through this non-combat encounter), low character personality development, with the PCs being free agents (in both an in-game sense and also that the GM has extremely limited control of of a PCs actions) operating in a semi-adversarial fashion against a GM were features characteristic of nonradical games. Well, I run games that usually have low emphasis on combat, that give moderate rather than constant attention to resource management, that have massive character personality development, that often use personality mechanics under which the GM can require a roll to avoid certain behavior, and that distinctly do not treat GM and PCs as adversaries. By most of your measures of radicalism, my games are radical. But they are not radical in the Forgeite style or, usually, in the diceless style. And indeed I find the Forgeite style so alien that my usual reaction to a game of that school is, "I can't imagine ever wanting to play that."

So I think "radical" is probably not a proper taxonomic category defined by positive traits that unify all its members, but a residual category defined by the absence of certain positive traits, in the way that "invertebrates" are defined not by any anatomical feature but by the absence of a feature (the backbone). Different "radical" games may be as different from each other as an ameba, a sea urchin, and an earthworm.