When you consider t
he results per constituency in 2015, there was only one seat (Berwickshire) where the SNP won with a small margin (less than 1,000 votes), while the Tory seat (Dumfriesshire) was also marginal and the LibDem seat (Orkneys&Shetlands) as well. And in the 2016 Scottish Parliament elections, the SNP only solidified its support. Seeing that go away seems wishful thinking.
It's probably more about narrative. In 2015 the SNP hit a sweet spot it's unlikely to repeat quite so spectacularly. That election happened after the huge rise in support for the SNP but before the unionist element of the Labour vote had collapsed towards the Conservatives in the way it has since. The party got 50% of the vote in that election and is currently polling around 47%.
So it's quite likely that it will slip back slightly from the 2015 mark. My friend Calum who is MP for Berwickshire may well lose his seat (which won't faze him too much personally as he was finding the London commitments a bit wearing as he has a young family). There could be one or two more. We don't know if the seats won by the two SNP MPs who subsequently lost the party whip can be retained by the new candidates.
Ian Murray won Edinburgh South for Labour on the back of the Tory vote going to him tactically. They may do that again, thought it's bucking the trend elsewhere in Scotland which is for the section of the Labour vote that hasn't already gone SNP to go Tory.
Alistair Carmichael is a proven liar who has disgraced his office, but the Northern Isles seem to vote LibDem on some sort of spinal reflex so he may hold on again.
In my own constituency of Dumfriesshire, Clydesdale and Tweeddale we have the solitary Tory. Lots of people gloating that he'll go, as he won by a smaller margin than the Green vote last time and the Greens have sensibly decided not to stand this time, but it'll be tough. We need a seriously heavyweight Big Beast candidate and I don't know who we can scrape up. (Well, I do know actually and he says he's considering it, but he didn't sound too enthusiastic so maybe that idea is a bust.)
So yes, the SNP could end up on maybe 53 or 54 seats instead of 56. And on 47% of the vote instead of 50%. And May intends to narrate that as a fall in support and "no mandate" and the SNP doing really badly. Even though she can only dream of polling 47% and getting (proportionately) that number of seats. This happened last year too, when the SNP actually increased its share of the Holyrood vote and the number of constituencies won over 2011 but because of the vagaries of the top-up system ended up with slightly fewer MSPs overall than before. This was then reported as Sturgeon having "lost her mandate".
Because the SNP got an absolute fluke of an unbelievable victory, any fall-back from this point, no matter how slight, will be spun as a loss. They're now saying Sturgeon has no mandate in Holyrood because she doesn't have an overall majority in an electoral system
specifically designed to prevent overall majorities, just because the SNP fluked it once.
The SNP will do eyepoppingly well in June by any normal metric, it will do better in Scotland than the Conservatives do in England and it will knock the Conservatives in Scotland out of the park. However it will be spun as a loss and a rejection of another independence referendum and so on. And if the Conservatives do a bit better in Scotland than they did in 2015 that will be spun as an enormous victory and Scotland embracing the Tories and so on, even though they're still beaten into a cocked hat by the SNP.
We'll just get on with things anyway.