Jimmy, if I remember right, you have a fairly good background in Greek and have studied textual criticism before? If so, then I will clarify for the other readers.
The idea is not that "eclecticism" in the raw sense can be altogether avoided. At some level, we have to combine two or more different manuscripts to arrive at a whole. But how we choose to arrive there is very different, and in my mind critical.
Westcott-Hort says we have all these different texts (and in the Alexandrian family, there is no clear transmissional history...it's just a grab bag), and none match, so let's combine them all to arrive at something inclusive of all likely variations (how they arrive at what they call "likely" is a mystery, considering their starting assumptions). The result (by design) is a hodge-podge, mish-mash that really can have no resemblance to the autographs. By their reasoning, they have GUARANTEED that what we have in the NA, or UBS, or whichever, IS NOT "autograph-like."
Robinson, and those who follow a Byzantine priority, say what started with the autographs was copied exactly to the next generation, to the next, to the next, etc. Over time and geography, some variations crept in due to copying errors, etc., but we must trace back, as best we can, the transmissional history. We know there IS a history, so let's try and find it. That is the only way we can arrive at something "autograph-like". It may not be exact (of course, there is no way of knowing, even if it were), but at least there is a hisory involved, and the possibility is there. The text produced ACTUALLY has ancient counterparts (with an s, as opposed to W-H and the Alexandrian folks, where they have guaranteed the opposite.