Why are predictive models still a thing?

Why are predictive models still a thing?

To make the political opining of analysts and "experts" to seem more legitimate than it can actually possibly be.

what Im genuinely wondering is: if the polls were rigged to favour hillary, how would that help her cause?

if people thought hillary was going to win anyway, they might not be as motivated to vote

I've thought about this a lot. I knew the predictive polls were biased to Clinton, but was always curious about what kind of impact that would have on voters.

I guess, at a most basic essence, someone in the Clinton campaign determined the consensus effect would outweigh confident supporters that lose motivation to go to the polls.

>what Im genuinely wondering is: if the polls were rigged to favour hillary, how would that help her cause?
They were rigged to make the rust belt appear to be in hillarys favor.

People thought that they DIDN'T live in a swing state so why would they bother voting?

Turns out every single rust belt state was a swing state along with florida and others.

they weren't rigged, just included likely voters only

>if people thought hillary was going to win anyway, they might not be as motivated to vote
That's the point. Trump winning was the plan all along. It's just that /pol/ doesn't realise it's just another Jewish snare yet.

>what is a push poll

The only plausible explanation I can think of is that they were intended to demoralise trump supporters into thinking he had no hope of winning so voting for him was pointless

Which would be a monumental miscalculation stupid if true

It must be fucking nice being a geoscientist or a petroleum engineer right about now

People are uncomfortable with ambiguity and would rather be told what to believe by a fancy (((((mathematical))))) model that they don't understand than just admit that no one knows what's going to happen.

The point is to make you believe it's an accident, a lucky shot, something that shouldn't have happend, and is not representative to the will of the people.
It's to entertain the idea that the people still want more immigration, crime and terrorism, because diversity is worth it. So in 4 years they can come back with the exact same plans.

Breaking news: when a predictive model tells you that something has 80% chance of happening, it means that it has 20% chance of not happening, i.e. 1 chance out of 5, which is nothing near negligible.

Sure, if we are only dealing with a single model, you have a very valid point. The reality is, however, that over a dozen models with 80-95% estimated chance of happening all turned out to be skewed to the same party. Apparently the mdels all suffer from the same bias rendering the organziations behind them as incapable. An alternative explanation would involve some vast conspiracy.

don't assume the models were independent. They suffered from the same mistakes

You assume the goal of the rigging is to get Hillary elected.

The actual goal is to move eyeballs, mouseclicks, screentaps, and ultimately the markets.

US elections are always a lucrative venture.

Good points and an even better full house get. Very impressive. When multiple studies all commit the same error or fall in similar biases, they should still not be excused though. It is incompetent modelling after all.

Statistic models, like all models, depend on some soft assumptions about reality. E.g. that your events are independent and that your data is representative.

Journalists, and other idealistic idiots who need a college degree but no brain, don't understand this, they don't understand what the model predicts under which circumstances. The whole poll thing is an example that garbage-in-garbage-out is still a thing

Not just that, but there were definitely a significant amount of people who voted for Trump not expecting him to actually win. A lot like what happened with Brexit.

Hilarious.

There's nothing incompetent about it. If I had a dozen models for predicting the number of times I don't roll a six on a die, I would hope they all work out at about 84% and I wouldn't complain that they were all wrong and incompetently made if I did happen to roll a six.

This.
There were some models that gave Hillary a 7% advantage over trump but reporting about a model like that will give you in no way the same amount of attention as a model that has a clear victor.
You can see the same thing happening in the """"science"""" part of many newspapers, where correlation and causation are happily mixed and statistical confidence is thrown out the window for sensationalism.

Personally I'm also speculating that predictive models fail to factor in the levels of enthusiasm present on either side.
The decision to vote for a candidate is very different if you have to actually leave your house and stand in line instead of just being asked which candidate you favor with no further personal inconvenience.
I think Trump voters were on average more invested in their candidate therefore more people actually showed up on election day.

I imagine it's kind of like with caring about the environment.
If you ask people randomly if they care about the environment many will say they do.
However, in situations where people are actually making decisions you'll find that even small personal inconveniences outweigh the abstract damage damage done to the environment.