No matter what the result tonight, someone will pop up and say “oooh, you can’t read too much into by-elections, voting patterns are different for by-elections”. Which spectacularly misses the point, because by-elections are important not because of what they show about public opinion, but because of the effect that they have on public opinion. The most celebrated by-election results, like Hamilton 1967 and Govan 1973, produced extraordinary nationwide snowball effects that fed into successive general election results.
However, to produce a snowball effect, a by-election result would need to have the “wow” factor, and because there’s such a strong expectation that Labour will win tonight, I’m not sure a narrow Labour victory would be enough to produce a game-changing form of momentum. To get a sense of where expectations lie going into the by-election, I ran an unscientific Twitter prediction poll the other day, which produced the following results –
SNP win: 30.2%
Labour win by 0-9 points: 43.0%
Labour win by 10-19 points: 18.6%
Labour win by 20+ points: 8.3%
I suspect the high percentage for an SNP win comes from people who aren’t aware of the basic arithmetic of the situation, ie. that Rutherglen is an unusually favourable seat for Labour and that the SNP would generally only win it if they’re in landslide territory nationally, which plainly they aren’t. Nevertheless there does seem to be a genuine perception that the SNP have reeled Labour in somewhat and a big Labour margin of victory is no longer on the cards.
That means the expectations game has produced dangers for the leaderships of both parties. If the SNP win (which I really can’t see happening), it could be the beginning of the end for Labour’s chances of a comeback in Scotland next year. But if Labour win by more than 20 points, which I’d suggest is a more plausible surprise than an SNP win, it could now be a sufficient shock to the system to raise big question marks over Humza Yousaf’s future as leader.
Related
