German Elections & Politics

[Filuwaúrdjan]

They aren't exactly unincorporated, though, as they are covered by a principle local authority; unincorporated areas in the United States have no municipal government. The English term unparished might be more apt.

[Beezer]

[mvd10]

Clickable new map of the 2017 federal election results by town (there are 11.000):

https://interaktiv.morgenpost.de/gemeindekarte-bundestagswahl-2017

I think this is the first map of its kind.

White = no inhabitants.

Also do not forget the +/- zoom sign at the bottom right, to show results for small towns.

What the  is wrong with the people living in Norderfriedrichskoog (39.3% FDP) and Fredeburg (51.8% Greens). I live in Schleswig-Holstein, but I have no clue why it is such a strong Greens and FDP stronghold.

For the FDP - protestants preferring not to vote for the traditionally catholic CDU?

The Greeny place is tiny, so it's probably statistics noise - but the Greens seem to do quite well between Lübeck and Hamburg, which is presumably quite wealthy commuter land etc, etc...

Maybe the strong FDP score in Norderfriedrichskoog has something to do with it's historical status as a tax haven? Mailboxes can't vote, but it's rather suspicious that a small village which used to be a corporate tax haven votes for the most pro-business party in a landslide .

[Anzeigenhauptmeister]

Soon, we're gonna have a grand coalition between AfD and Greens...

[jaichind]

http://www.spiegel.de/wirtschaft/unternehmen/manipulation-in-der-marktforschung-wie-umfragen-gefaelscht-werden-a-1190711.html

Seems to say that several German polling companies have compromised scientific standards and used smaller samples than officially stated.  It mentions instances of respondents’ identities, IP addresses or time of survey being falsified.

[Filuwaúrdjan]

What entirely surprising news. I'm shocked. Is anyone else here shocked? Shocking.

[Angel of Death]

Projektion: Wenn am nächsten Sonntag wirklich Bundestagswahl wäre ...

What a weird way of phrasing that. Even in the (impossible) hypothetical, the pollster feels bound to respect election law.

[Anzeigenhauptmeister]

Projektion: Wenn am nächsten Sonntag wirklich Bundestagswahl wäre ...

What a weird way of phrasing that. Even in the (impossible) hypothetical, the pollster feels bound to respect election law.

I don't understand your criticism. That's what German pollsters have been asking their pollees since 800.

[Tender Branson]

The GRAND Coalition has no majority anymore ... (47.5% vs. 48.5%):

[Oryxslayer]

The GRAND Coalition has no majority anymore ... (47.5% vs. 48.5%):

Presently an outlier.

Last Forza poll was 33 18 13 9 9 13
Last Infratest was 33 18 14 10 11 11
Last Emnid Poll was 33 20 13 9 10 11
Last Forschungsgruppe was 31 19 14 7 11 14
Last YouGov (a few weeks) was 34 19 14 8 11 10   

All that said, its certainly possible we start seeing lower CDU numbers, in this poll they think the FDP has recovered from their spiral. Every other poll, even the Forshungsgruppe one that also has a slightly lower CDU count still has the FDP declining - can our German posters comment if anything has changed?

[jaichind]

The GRAND Coalition has no majority anymore ... (47.5% vs. 48.5%):

If this is the result of the next election there will be massive pressure on FDP and Greens to join CDU/CSU into a government.  I guess if the trends last few election cycles and government formation continues like this we will eventually get to a place where it is  (AfD+Linke) > (CDU/CSU+SPD+FDP+Greene) 

[parochial boy]

The GRAND Coalition has no majority anymore ... (47.5% vs. 48.5%):

If this is the result of the next election there will be massive pressure on FDP and Greens to join CDU/CSU into a government.  I guess if the trends last few election cycles and government formation continues like this we will eventually get to a place where it is  (AfD+Linke) > (CDU/CSU+SPD+FDP+Greene) 

That wont happen, most European countries have a pretty solid ceiling for far right support, and Germany's will be lower than most

[palandio]

The GRAND Coalition has no majority anymore ... (47.5% vs. 48.5%):

Presently an outlier.

Last Forza poll was 33 18 13 9 9 13
Last Infratest was 33 18 14 10 11 11
Last Emnid Poll was 33 20 13 9 10 11
Last Forschungsgruppe was 31 19 14 7 11 14
Last YouGov (a few weeks) was 34 19 14 8 11 10   

All that said, its certainly possible we start seeing lower CDU numbers, in this poll they think the FDP has recovered from their spiral. Every other poll, even the Forshungsgruppe one that also has a slightly lower CDU count still has the FDP declining - can our German posters comment if anything has changed?

There is a lot to say about polling in general and how this can help to understand the current situation. When looking on polls there are at least two points to take into account:

1. House effects: Some pollsters in a certain period of time tend to show results that are systematically different from those of others. For example INSA (whose boss is regularly accused of being led by a pro-AfD bias) tends to show higher AfD results than the other pollsters and also lower CDU results. But when you look at the actual election outcomes, INSA has a good track record. They were right (for whatever reason), the others were wrong.

2. Random error: German pollster are polling the same scenario (federal election) again and again, they have experience. They also add a lot of secret sauce to their data and do not put their raw data in the headlines. You can immagine their raw data to swing much more. Because of this there still is random in the 1-2% range, but everything above that is certainly telling us something and cannot be dismissed as an outlier. I even would dare to say that the established German pollsters do never produce federal election polls that are outliers.

For what regards the FDP: They have been esentially stable for the last 6-8 weeks at ca. 2 points below their last result, with INSA showing relatively strong numbers and FGW lower numbers than everyone else.

I can certainly immagine that the results of the coalition negotiations will have an effect on the polling numbers. What is my impression so far is that the discussion about family reunions for refugees has damaged the SPD both with voters who lean right and who lean left on this issue.

[Anzeigenhauptmeister]

The GRAND Coalition has no majority anymore ... (47.5% vs. 48.5%):

I wonder what the first-vote map would look like, especially regarding the number of direct mandates for the SPD in the Ruhrpott and for the AfD in the East. i even think the AfD could win some constituencies in the Ruhrpott, especially in Essen (former Social Democrat Guido Reil).

Furthermore, I'm a bit apprehensive of the largeness of the Bundestag under such a result; our parliament could become bigger than China's...

[EPG]

The GRAND Coalition has no majority anymore ... (47.5% vs. 48.5%):

If this is the result of the next election there will be massive pressure on FDP and Greens to join CDU/CSU into a government.  I guess if the trends last few election cycles and government formation continues like this we will eventually get to a place where it is  (AfD+Linke) > (CDU/CSU+SPD+FDP+Greene) 

Whatever about Weimar, Germany's certainly heading toward a situation where AfD could be the largest party.

[Anzeigenhauptmeister]

I think this graphic says it all. Thank you, St. Martin!

[EPG]

Wow, I hadn't realised that SPD + Linke combined was less than the Union.

[Filuwaúrdjan]

1. House effects: Some pollsters in a certain period of time tend to show results that are systematically different from those of others. For example INSA (whose boss is regularly accused of being led by a pro-AfD bias) tends to show higher AfD results than the other pollsters and also lower CDU results. But when you look at the actual election outcomes, INSA has a good track record. They were right (for whatever reason), the others were wrong.

Though there isn't necessarily even a contradiction here - it's quite common for fairly new polling firms to consistently produce results that are especially striking most of the time and then carefully move towards the 'consensus' position as a General Election looms. Polling, we should not forget, is a business - firms need clients; the best way to get them is attention (firms often use political polling as a way of getting business clients, in fact that's one of the main reasons to do it in the first place). And often which firm gets to claim to be the most accurate based off the final poll is a matter of a handful of points well within the MoE and is generally no indicator of future performance (the British firm ICM - which for over fifteen years flaunted a dubious reputation for accuracy based off the 2001 election - is a case in point).

But yes: most German firms have strong and observable house biases. None of these make a given firm any more or less accurate - that's not how this works. The correct thing to do when that's the case is to pay a little bit of attention to all of them, but not too much to any of them. Which is why I always get a bit annoyed in threads like this - cherry picking for the most SENSATIONAL finding is even more irritating than cherry picking for partisan reasons, and tends to lead to a pretty mindless and hysterical discourse.

[Filuwaúrdjan]

2. Random error: German pollster are polling the same scenario (federal election) again and again, they have experience. They also add a lot of secret sauce to their data and do not put their raw data in the headlines. You can immagine their raw data to swing much more. Because of this there still is random in the 1-2% range, but everything above that is certainly telling us something and cannot be dismissed as an outlier. I even would dare to say that the established German pollsters do never produce federal election polls that are outliers.

Yes - in a fragmented political landscape like Late Merkel Germany two points here or there matter a great deal, but are just as likely to be statistical noise as in a less fragmented landscape. What's striking is that though house effects have become much more noticeable over the past few months, each individual firm's polling results are very close to the last election. This is unlikely to be a complete coincidence.

You must be logged in to read this quote.

Exactly, yes. People in all countries are much more flexible with their votes these days: except for partisan die hards (and there are less of those) support is generally conditional by default. Which means that, and even more than used to be the case, polling reflects whatever is going on at the time - that might seem like a banal and obvious point, but it's easy to lose sight of the obvious sometimes. The present situation - with a tired CDU Chancellor and a very weak SPD leader negotiating in a frankly pretty weird way about a possible third Grand Coalition in four elections, following the collapse of talks with a different set of partners and with the actual election itself a very recent memory - is just about perfect for polling fragmentation. The situation that replaces it might be quite different (or not).

Of course there is one way in which present polling does matter - the impact it might have on government formation. Because, like it or not (and I'm not sure if I do), politics is now the most postmodern part of Western societies.

[Anzeigenhauptmeister]

But yes: most German firms have strong and observable house biases.

The founder and CEO of Forsa, for example, is Manfred Güllner, who is not only a member of the SPD, but also belongs to the neoliberal Seeheimer Kreis wing of his party.

[Filuwaúrdjan]

But yes: most German firms have strong and observable house biases.

The founder and CEO of Forsa, for example, is Manfred Güllner, who is not only a member of the SPD, but also belongs to the neoliberal Seeheimer Kreis wing of his party.

lmao yes that's the funniest one.

[Filuwaúrdjan]

Anyway - decent chance I might have a bit of fun with the municipality stuff posted a few days ago. Nice to have all that data in one place...

[Anzeigenhauptmeister]

Anyway - decent chance I might have a bit of fun with the municipality stuff posted a few days ago. Nice to have all that data in one place...

Yes, I love that map.
It's nice to see in how many Bavarian municipalities the AfD came second. This year's state election will become amusing...

[Anzeigenhauptmeister]

Maybe the strong FDP score in Norderfriedrichskoog has something to do with it's historical status as a tax haven? Mailboxes can't vote, but it's rather suspicious that a small village which used to be a corporate tax haven votes for the most pro-business party in a landslide .

Norderfriedrichskoog's election result's are astonishing, anyway...

Out of the 28 voters, most did not only vote for the FPD, they didn't even give the SPD one single vote.
If the federal trend continues, the SPD will be the first party to get negative votes.

FDP CDU AfD Linke Grüne Freie Wähler

39,3 % 32,1 % 10,7 % 7,1 % 7,1 % 3,6 %

11 Stimmen 9 Stimmen 3 Stimmen 2 Stimmen 2 Stimmen 1 Stimme

[Anzeigenhauptmeister]

In Fredeburg, however, where the Greens received a majority, the AfD ironically didn't get one single of the 27 votes. However, the PARTEI and the BGE (basic income party) received votes; hence it's a freedom municipality.

Grüne CDU SPD Die Partei FDP BGE

51,9 % 22,2 % 11,1 % 7,4 % 3,7 % 3,7 %

14 Stimmen 6 Stimmen 3 Stimmen 2 Stimmen 1 Stimme 1 Stimme