[bandyaboot]

Personally, I don’t care what the “media” was saying. I care what the polling data and the election models based on the polling data were saying. They were saying pretty consistently that this could go either way, but that at the same time the result may not turn out to be actually that close. Those two aren’t incompatible.

[itsoktocry]

It could go either way? It might be close, or maybe not?

Imagine taking any of that seriously. Didn't they give Hillary an 80% chance?

[darkarmani]

How could Hillary lose if she had an 80% chance? That seems impossible (if i'm bad at math)!

[hughesjj]

For real, 80% is effectively coin flip territory for me. I wouldn't put any important bets on 80%, for the same reason I don't play Russian roulette.

[jkubicek]

Why wouldn't you take it seriously? "Could go either way" is a perfectly reasonable prediction based on the polls that we all were seeing.

[bandyaboot]

Imagine being such a condescending douche…and an ignorant one to boot.

[lazyeye]

If any other profession was as consistently wrong as pollsters are, would they be taken seriously? I think the main job of pollsters is to provide content for corporate media (the closer the polls the better for attracting eyeballs for advertisers). And they do this job admirably. It just has nothing to do with the election.

[bandyaboot]

You don’t value polling, ok. No use continuing to go back and forth about it. Instead, maybe you’ll feel like responding to one of the other commenters that replied to you about prediction markets…

[eastbound]

Polls are twisted to return falsehoods from gray information. It’s hard to fathom that you don’t notice neither the methods nor the results. It’s a bit like living in Beijin and saying that Tiannenmen is conspirationist storytelling, or a coin flip on whether it happened or not. It did. 100% chance.

“It’s 50% probability. Either it’s true or it isn’t.” — what meddlers pretend when they’re not happy admitting the high probability of their enemy candidate being elected. It wasn’t a coin flip.

[thomaslord]

Assuming that random factors like "it rained" or "voters got in car accidents and couldn't make it to the polls" aren't a significant factor, there's always a 100% probability of one specific candidate winning since everyone has made up their minds before the day of the election. What polls do is not telling you the real-world probability, it's telling you the likelihood of a given outcome given known data.

Polls always need to be skewed in some way to be accurate, since not everybody will vote. You can't just get a random distribution of the population's preference and assume the more-preferred candidate will win. Polls can never be truly accurate because people will lie about which candidate they're voting for and whether they're planning to vote, and sometimes people who genuinely intended to vote never make it to the polls. There are a huge number of variables to consider when trying to predict the outcome of the election, but it's important enough that it's still worth trying.

[hughesjj]

The polling was pretty darn close though, overall. Same as in 2016. The thing is, there's enough polls out there that people can pick the outliers and decide themselves into a narrative that makes them feel good.

It's an incredibly small number of voters in the key swing states that actually decide the election. It's under 1% of the voters to swing the election. Winner take all + electoral college will give you that.