Regular guest on Hugh Hewitt radio, Chuck Todd who is moderator of Meet the Press, said that the poles in 2016 were right but the analysis was wrong. I take that to men that like climate change, the published polls are adjusted data. They asked lots of questions to lots of people and someone who knows more than all of us put it together and says that Biden leads by four, for example.
In other words further, the pole is not what it claims to be. They did not ask 2000 people for example and these are the results. They asked even more people and they chose the results. They of course are biased by the results that their peers are publishing and by the results that they want to publish.
As long known, this makes a mockery of the published margin of error, which is the statistical, mathematical sampling error, and ignores a whole host of potential human errors including intentional error and outcome bissed errors.
Sampling error: imagine a huge jar of 200 million red and blue marbles all randomly mixed up throughout and randomly select a few hundred of them, count them and project the result for the entire population. The amount you can be wrong by is the sampling error. Evenly distributed is not what the demographics look like nationwide or in any state and that's not how the 'polling' process works. It's more like a secret sauce mixed up by some pretend, really smart people. And they are only judged by their final poll. All the rest is BS.
The final 2016 Wisconsin poll average that decided the election was wrong by 7 points. Think about that, how can all these polls be wrong by more than 2 or 3 times the 'margin of error', all in the same direction? Simple. They aren't polls, it isn't an evenly distributed electorate and it isn't random sampling. It's a media manufactured product, for better or worse.
