As a general rule, any subsample, even a subsample comprising half of a poll's respondents, is not going to be anything like as reliable as the full poll. The full sample will be correctly weighted to demographic and political target numbers, but that isn't necessarily going to be the case for every individual subgroup, so you'd expect a degree of random fluctuation among those. A No lead among men in an individual poll shouldn't, therefore, be treated as gospel. Stuart would have more of a point if a succession of polls had repeatedly shown the same thing about male voters, but as he may or may not know (has he bothered to check?), that isn't the case at all. In fact, this should be pretty obvious even just from a common sense point of view. It would have been extremely improbable for Yes to have ever built up such a sustained lead if Nicola Sturgeon had been repelling male voters while she had been attracting female voters - the two trends would have cancelled each other out and No would have stayed in the lead. In the real world, the general picture across most polls is that men have remained pro-Yes while women have been converted to independence in significant numbers. The Survation poll doesn't disprove that or establish a new trend - the figures for men are probably just caused by meaningless sampling variation.
When I pointed out a couple of weeks ago that Stuart was actively campaigning to bring Nicola Sturgeon down, he angrily denied it, and yet here we have a post that paints the 19th Yes majority poll in a row as bad news, and then attempts to 'blame' Ms Sturgeon for it! And all on the basis of a very silly false premise.
It's getting almost comical now.