Your tables are not suggesting that the 11 sample languages are descended from Atlantean, but that words transformed as tribes or peoples splintered apart, with Atlanteans being one of these splinter groups. Is this somewhat correct?
Mostly correct. My hypothesis is that Atlantean could have been a "linking group" between various other groups by way of trade contact. This may have been a lingua franca in oceanic coastal cities as primitive coastal tribes prepared themselves for communication with Atlantean merchants bringing wondrous trade goods from far off lands. Over time the two-way influences of words would have transmitted in varying degrees over and across Atlantis' entire trade network. This isn't to say the various languages "come from" Atlantean, but that individual words here and there would have networked their way around the world, transmitted via Atlantean traders.
Do you foresee an effective way to track the path of splintering among the peoples associated with these languages, perhaps by somehow graphing out this linguistic splintering alongside genetic migrations?
I think it can be done, with a much more monumental amount of effort than I have put into it so far. What really needs to be done is a comprehensive etymological analysis among all ancient languages, similar to what has been done toward mapping the genome and tracking human migrations on that basis. Just as genetics has called certain archaeological dogmas into question with their discovery of haplogroup X, linguistics too might make similar discoveries.
Obviously, as you yourself admit, the resultant Atlantean words are highly speculative. Are they intended to be used as a rough guide in pinpointing the descendants of Atlantis, similar in the way computer morphing technology might approximate the appearance of a child in adulthood? How would this be done? Would this again involve genetic migrations?
The method used is an interpolation similar to what has been done to discover "Indo-European Roots". The commonalities among words are studied for what the original might have been, which could have morphed in one way in one location, and another way in another location. Key to that is to find cognate roots, and to also know when those roots are not present (what I call "morph offs"). The original is taken as simply a possibility, and all that it establishes is also a possibility: a possible relationship between the connected words. The counter-argument to that possibility has to rely on "mere coincidence", such as an assertion that the Basque and Polynesian and Hebrew words for "Law" are near-identical. Whoopsie, humans made a mistake and forgot they were supposed to use different words for that. The hypothesis I'm developing is that the word may have been transported by oceanic mercantile explorers and adventurers, most likely related to the "Phoenicians". In fact the Phoenician language is one that I want to study more thoroughly to see if the traces of those cognate roots are clearly visible there.
One of the things I noticed in your lists is how all the Atlantean words are lengthy and polysyllabic. Most of the sampled languages have a few monosyllabic words in the list and Polynesian, which does not, has many words composed of three or four letters and two syllables. Shouldn't common words like fire and blood have been weighted in your system toward shortness in line with the rest of the words in a column?
Length in terms of letters doesn't necessarily mean length of pronunciation. "Tzuarhei" could really be pronounced with the same speed as, if not faster than, the modern Tahitian "Auahi" (from the more archaic Polynesian "Ahi"), because the latter is drawled, and the former could be more slurred. Also, even within the same language, words can lengthen or shorten over time, so there is no reason to assume they could not in transmission via globe-trotting sailors.
