I think if we agree that there are individual differences in predisposition towards juggling aptitude, and that the predisposition is mediated genetically somehow, and if juggling (in this hypothetical) is biologically advantageous for survival/reproduction on one of the islands (really stretching the analogy here) - then I don’t see a way how my 1000 years experiment doesn’t produce actual, population level genetic drift in juggling predisposition between the population on island A and island B (unless we could somehow prove that juggling predisposition is not heritable)

This is the crux of the issue. You assume, in the absence of evidence, that juggling aptitude predisposition is mediated genetically, in such a way that it is variable among humans; and you place the burden of proof on "somehow prove[ing] that [it] is not heritable" (emphasis mine). But it's far easier to prove the positive (in a world where the positive is true) than prove the negative (in a world where the positive is false), so the appropriate perspective to adopt if you want to investigate via experiment is "try to prove it true, and if we honestly fail, that's an argument that it's false".

If you don't intend to run the experiment, then different considerations dominate: you should justify this bias. Maybe you're drawing an analogy to something similar, which you know to work this way? Bias isn't the same thing as wrong, after all: but unjustified bias often is.