That is a good question, and one where I had to do some self examination. I think it is because of the closeness of these claims to just-so stories, despite their posing as science (
https://en.wikipedia.org/wiki/Just-so_story ). That is, stories that fit empirical data to an explanation of it, without ever considering at any point if there is any reason to discard it, or if your hypothesis can even be disproved -- like the famous "not even wrong" quip by Wolfgang Pauli.
> I mean, to be honest I have no idea what's actually true, and I don't think anyone (including you) does either. So why be annoyed when people bring up a possibility, not disproven, just merely unlikely...
I'd like to point out that I'm not at all annoyed at people talking about the possibility in potentia. I disagree that it's a good possibility, but I am only annoyed when this is presented by professionals, most probably scientific journalists*, as something we should expect to be probable. This also extends (a bit unfairly, I guess) to amateurs repeating that claim as truth -- this is just bad pop science.
I do think, however, that the two quoted sentences above do at least imply a logical fallacy: of course we don't know the truth in this case. This does not make any kind of speculation an equally valid hypothesis to any other, there are distinctions, as per Russel's Teapot. And a claim as maximalist as "all the worlds flooding myths come from the Black Sea" better have some actual evidence to back it up, or be preceded with "In further publications, it might be interesting to look for evidence that ..." in the paper.
> As for how long oral traditions can survive without writing -- it's up to anyone's guess isn't it?
Similarly, the logical conclusion of that is not "there are stories that have been passed down from our common time in Africa 200k years ago, because who knows how far oral history can go" but rather "until there is actual evidence to the contrary, oral tradition remains suspect as a carrier of stories". We do know that correct details, at least, can be transported over centuries: the Iliad contains descriptions of soldiering that are, in part, clearly Late Bronze Age Mycenaean, though by the time it was written down the authors were unable to disentangle these tidbits from descriptions more fitting to the Archaic Age. That is the upper bound that I'm aware of: a few hundred years, information strongly mangled, but still recognizable if you know what to look for. How do you propose to go from there to an order of magnitude more? Of course, you might reply "you can't disprove it", to which my reply stays the same: then there is absolutely no reason to believe it true.
* I checked the original paper about the deluge, the authors didn't mention all of the worlds flooding myths at all in there. The wild speculation wisely appears in a book they published, which I do not have and which seems to propose a connection of all Indo-European and Egyptian myths. This is a far cry from "the Chinese are unable to think of a flooding myth themselves".
Update: a bit of searching around has mainly offered up thinly-disguised young earth creationist sites or sites clearly pointing out the problems as well as the allure of the "all flood myths are from the Black Sea Deluge" speculation. Maybe I should go softer on science journalism here...