Quote:
Originally Posted by Kukraprout
French pollsters probably all know one another and apply very similar recipes. A systematic error is definitely possible imo.
I'm really not confident in Macron anymore. I had spent the last few months telling Melenchon enthusiasts that their candidate had no chance and that Macron was the only way to make their vote count, that's not true anymore. On the other hand Melenchon could very well collapse when people become aware of his less moderate positions. I have no idea which way it will go and I wish the election was yesterday.
You still actually believe that telling people Melenchon is an anti-establishment candidate will put them off him? Did it never occur to you that people who are thinking about voting for a communist don't want a moderate?
It will be interesting to see what happens with Melenchon's numbers.
He is currently 10.5 at betfair? That's a value bet IMHO. If the polls are accurate that could be justified on the basis of standard error.
There is a possibility, and it is
just a possibility, that he will benefit from voters who aren't showing up in samples the way far right voters did in the UK and in the US. Melenchon voters, like the far right, are very mistrustful of the establishment almost by definition and this may extend to not wanting to participate in offifcial surveys.
The interesting thing with probabiility theory is that you only need a small probability of something like this happening to have a value bet. I'd put Melenchon's chances at around 12.5% making a fair bet 8.0.
That said, the value here seems to be on fading Le Pen and Fillion. I can't see any justification for 5.2 giving the enormous hurdle she faces in getting to the second round. The polls would have to be further out than any election in human memory. It could still happen due to effect I mentioned above but it would have to be a much stronger bias than in any previous case.
Fillon looks like a dead-in-the water establishment dinosaur now, I can't see any justification for his price of 4.2.