Michael cohen says he tried to rig online polls ‘at the direction’ of donald trump – slashdot

With all due respect, it’s time for you to stop talking like you’ve got the pulse of "real Americans." The FBI does a fine job. The FBI is staffed with many, many people who love and care about the country and have made it their life’s work to help the country. I believe that anyone who goes out of their way to make it seem like the majority of government workers have some hidden, secret agenda is probably full to the tipping point with bullshit. Sorry pal, that argument doesn’t pass the smell test. You may believe the turds that fall out of the president’s mouth, but let me assure you that you are in a very small majority with regard to that.

The majority of the country still believes that the people working in the government – REGARDLESS OF THEIR POLITICAL BELIEFS – are there because they love the country and want to help it, not damage it. Trying to act like the literal federal law enforcement agency would go against the Constitution and break the law is laughably absurd, and it says very bad things about your though process and where you get your information from.

You know what? Don’t take my word for it. How strong are your convictions? Are you just a big talker or do you back it up? Because I would encourage you, whole-heartedly, to find the bars around your local FBI office and go in and start telling everyone how all FBI agents are corrupt and hate the country. Go ahead, let me know how that works out for you.

1) Understand what polling is and how it works before saying "polls mean nothing". The actual election itself was a poll. The main difference between the election and the pre-election polls was that the election included the entire sample set of people whose votes were actually counted, whereas a pre-election poll, by necessity, cannot know whether any given voter will actually show up on election day, and even if they could predict that perfectly, they have a margin of error regarding whether or not their MUCH smaller sample size is adequately representative of the population of voters on election day.

3) It is possible that a) late-breaking events or b) the polls themselves altered the behavior of voters when it came time to vote on election day. People can change their minds about who they will vote for or their likelihood of voting. Events that happen after a pre-election poll cannot be retroactively fitted into existing poll results. People deciding that the election is a foregone conclusion and staying home is also difficult to incorporate. People deciding that they simply cannot stand the projected result is the other side of that coin. Constant exposure to polling information is demotivating to the projected winner’s supporters and motivating to the supporters of opponents.

6) Different polls have different reliability levels. Online polls of the sort that the President apparently sought to cheat on have some of the worst reliability levels you can get. Online polls have the abysmal selection bias, among other problems. And since his target audience is filled with people who discount science when it comes to things like evolution and climate change, whose education in mathematics, social science, and statistics is almost certainly lacking… it might be useful for him to have polls he cheated on to point to as a counter-point to those produced by the "fake news" folks. Then the narrative becomes "fake news trying to prevent Trump win with fake polls, don’t let that happen! The real polls show Trump can win!" Competing polls results can heighten the effects mentioned in #3, to emphasize an "us vs. them" narrative.