>Czech elections: How I got it wrong (again) – and why
And, if I was head of the Czech Politics Pundits Party, I too would resign, because as you will gather, I was badly wrong in my forecast (again). However, mistakes can be instructive, so let’ go over how I got it wrong. I predicted
Social Democrats 27%
Civic Democrats 23%
Communists 13.5 – 14.0%
Christian Democrats 6%
My first main mistake was to assume – perhaps thinking of how British voters behaved earlier this month – was that new parties support would be less than that in the polls and that established parties somewhat greater. My assumption was that new parties new found popularity was fairly flakey and that some of their supporters either wouldn’t turn out to vote or would make a better-the-devil-you-know choice at the polls and opt in the end for an established parties. So Mistake No 1. was to underestimate the frustration of Czech voters and to overestimate the underlying appeal of established parties. A very West European error.
This led to two smaller errors: Mistake No. 2 was to overestimate the core electorate of historic parties such as the Communists – who lost votes in both relative and absolute terms – and the Christian Democrats (KDU-ČSL), who did indeed fall below the 5% threshold. Interestingly, the gambler in me got this one right – I staked 10 euros on them crashing our of parliament winning have a princely four – while the more cautious blogging political scientist didn’t. Mistake No. 3 was to discount the prospects of small left-wing parties, despite the fact that polls showed at least one (SPOZ) creeping up in the polls to 3-4 per cent.
Overall, I show (as ever) a lack of political imagination – or an engrained sense of disbelief – about likely changes. Borrowing from the trends picked Kevin Deegan-Krause’s poll analysis, I at least see that both main parties are not going to suffer a dip in support, but what I failed to see is that far from rowing back from these trends being , in the actual results conistent of these trends writ very, very large indeed. A bit of imagination and the Deegan Krause analysis and you could have been there.