With just two days to go, this is my last look at AfD support in the polls before the upcoming EP elections. Pinky promise.
The graph shows a model-based estimate of the development of AfD support over the last 11 months. The latest survey that’s currently included in the dataset is the Forsa poll that came out on Tuesday. If the knife attack on a so-called ‘Islam critic’ that led to the death of a police officer and the subsequent debate had any effects on AfD support, they will not be fully reflected in these data.
Disclaimer: I only include information from surveys that ask about voting in a federal election. While some pollsters specifically ask about the EP elections and while I fully believe that respondents are not stupid and can differentiate between the two, presenting them with two different types of hypotheticals and expecting meaningful differences is a bit too ambitious in my book. Also, second-order-election theory and stuff.
Taking this caveat on board, you see … not much.
The confidence interval is quite wide at the moment, but that is partly an artefact. My puny little model works with synthetic months, and we’re right in the middle of the current one, plus national vote intentions are currently not as often polled as usual because of the EP election. Which means that this month’s estimate is based on fewer polls and hence less precise.
Having said that, it is not significantly different from the mid-May estimate of about 16 per cent. All the excitement of the last weeks has done absolutely nothing to shift the mood. If the election on Sunday plays out along these lines, that would give the AfD 15-17 seats, a cool ~45% increase on their 2019 result. More as we get it on Sunday night.
Reposts
Likes