In reply to The New NickB:
> Do you think? In the recent Heywood & Middleton by-election. The Labour vote held up and the Tory and Liberal vote collapsed. With UKIP benefiting from a tactical switch from the Tories and the BNP not fielding a candidate.
I think you're looking at the % not the votes. The Lab % held up on a massive decrease in turnout. What you have to decide is who didn't turn out? LD+Lab voters in 2014 were only 45% of LD+Lab in 2010.
For UKIP to have only gained the BNP and Con losses, that would mean all 16000 missing voters were LD and Labour. That would be 85% of LDs not voting and 50% of Labour. Is it really credible that in a Labour stronghold, with a threat of UKIP only half their supporters turned out. That would be on a par with some of the worst turnout records for the total voters.
The two extremes are:
1. Labour voters are not voting, other are
2. Labour voters are voting UKIP, turnout is generally low
I think the assumption is (1) but the reality nearer (2). The mistake that some are making is that assuming UKIP will only gain from Conservatives. If that were true, it would split the vote and Labour would win a lot of marginal seats. With (2) the effect is lessened, and UKIP will win seats. I think instead of a Labour victory, we'll get a Labour minority, with the danger than a Conservative + UKIP coalition might be larger. A labour+SNP coalition might be the only alternative.
Either way, neither sound comforting.