I've just gone to vote, at roughly the same time of day I've chosen at previous elections, and...holy dooley. I've never seen such queues. It was quite a contrast to a certain European election a few years ago (turnout 25%) when they looked completely shocked to see me! So as much as I hate reliance on anecdotal evidence (naming no names at this point) I think it's a fair guess we're in for bumper turnout figures tonight, perhaps exceeding the low 70s range we saw in 1997.
For months now I've been resistant to the idea - repeatedly put forward by the likes of Michael Portillo - that the perception of a close contest would be enough in itself to fuel a significant increase in turnout. I don't necessarily think I was wrong about that, though - it's been the (rigged) leaders' debates that have made most of the difference. That point in itself might seem like cause for concern for the SNP, but more broadly, how can we expect a high turnout to affect their prospects? Logically, it ought to be a bad thing, because support for the party is highest among older age groups, who are usually more likely to vote regardless of circumstance. In theory, a higher turnout among young people might be expected to dilute that advantage. But there's one big problem with that theory - if anything, the SNP have underperformed in the last four general elections, with the assumption being that nationalist voters are less likely to take Westminster elections seriously. A higher level of voter interest this time might counteract that problem.
So you can look at it either way - not long to wait until we find out which way it will go.