I pulled some data and did an actual analysis.
Data is here:
Election Resources on the Internet: Parliamentary Elections in the U.K. - 2015 General Election Constituency Results
I'd like to upload my analysis but it is 391kb zipped and that seems to be too large for this site, unfortunately.
All I did was take each of the 330 Conservative victory seats in 2015 and considered what would happen if x% of the non-voters instead decided to vote Labour, with everything else remaining constant.
- At a 20% rate of non-voters becoming engaged with Labour, you start to get a swing. It starts to pick up rapidly from there.
- At a 25% rate of non-voters becoming engaged with Labour, you get enough of a swing for the Tories to drop to 316 seats, which would be enough to deny them victory.
- At 30%, Tory seats drop to 306. By 33% it drops to 298. By 35% it is 288. This is in the range of Labour largest party, I would have thought.
- At 40%, the Tories only have 265 seats, which (depending what happens in Scotland) is likely to see a Labour majority.
So there it is. Don't know where Polly Nonsense gets her information from but it is certainly possible for Corbyn to win purely on engaging non-voters. A 30% engagement rate would be an overall turnout of 76%, a 40% engagement takes it up to 80%. These are high turnouts but not impossible.
That doesn't even start to consider the effect of winning over LibDems, Greens and UKIP (and SNP of course, but that's another issue). If Corbyn can engage the 4 million UKIP and 1 million Green voters, that would also see him move a long way forward.