I don't understand why you post any polls at all. You have stated ad nauseam that the pollsters were correct in the last Presidential election predicting Hillary had a 92% chance of winning. According to you, they were correct because Hillary won the popular vote. Therefore, you are inferring that the statisticians employed by the pollsters did not incorporate the electoral college into their statistical models. If you are correct by stating the pollsters were correct, that makes statisticians employed by pollsters the dumbest statisticians in the history of the universe. That makes all of the polls toilet paper at best. Perhaps, that is why they work for pollsters. They were too stupid to pass the actuarial tests.
Hillary did have a 92% chance of winning.What many of you don't seem to understand is 92% is not a 100 %. While the 2016 polls got the EC vote winner wrong and popular vote winner right the 2004,2008 and 2012 polls got the election and popular vote winner right and polls got The House and Senate vote right in 2018.I would certainly rather be on the same side as the polls as they are right far more than they are wrong.
Do the statistical models only calculate the popular vote probability? That is what you have inferred ad nauseam. That is like building a statistical model for chess and not including that checkmate wins. I do understand that 92% is not 100%. However, I don't think you understand the significance on 92% in statistics. If I could build a trading system with a 92% probability of winning, I would be a gazillionaire.
Our foreign trolls should be very happy! Like when they said on the night of the November 8, 2016 election that Hillary Clinton had a 98% chance of being the next President of the United States. Exactly, how did that turn out?
Oh, geeh! One more problem for President Donald Trump being re-elected. https://www.yahoo.com/finance/news/us-employers-hired-record-number-143349518.html
Prior to the 2016 election the popular vote winner won the election 92% of the time.I understand the significance of 92% in statistics quite well.Since presidential elections began in this country the popular vote winner has won 90% of the time. Some other nice odds.Only 1 President who lost the popular vote got a second term.Since polls started in the 1940s no President with an average 1st term approval rating below 49 won a second term.Trumps 1st term approval rating is 39.
But you would still be wrong 8% of the time. 92% is significant but if it comes up in the 8% that is well within the realm of statistical probability over a large sample (i.e. 225 years of elections).
Okay, that is not my point. Tony has stated countless times that the pollsters probabilities were correct even though Hillary lost the election, therefore, Tony is inferring that the statistical models do not incorporate the electoral college and only calculate the popular vote probability. Hence, my analogy that building a model like that would be like building a stat model for chess that doesn't incorporate that checkmate is winning. Tony has posted that so many times that Tony has selfishly taken the parrot crown away from Fraudcurrents. I have to assume that neither one of you have any experience with respect to machine learning. I have to further assume that both of you are pretending to understand stats. Either one of you post the sigma or stdev of Trump winning. A person that has a basic elementary understanding of stats could give me a nearly accurate stdev instantaneously off the top of their head. If you can't, I really don't know why you are on a trading board.
I don't give that much of a shit to be honest to start delving into z-score and standard deviations. I was just highlighting that a normal distribution has tails for a reason that no matter how slight the probabilities, they do exist. I also have repeatedly said election polls are useless because of the sampling methods used given the geographic diversity of the country and variety of political biases. So I never pay attention to them. Comes from my graduate studies of econometrics and regression analysis. Just formal guessing is all it is.
What does President Trump do when approval ratings are down ???? hold more rallies and sell more merchandise. ‘Why am I not doing more rallies?’: Trump desperate to get out of working in Washington President Donald Trump believes he has the answer to his poor poll numbers: more rallies. Trump has bizarre priorities for which aspect of his re-election campaign require his personal attention, The New York Times reported Monday. The newspaper noted, “Trump has griped about traveling too much, but then lashed out at aides, demanding to know, ‘Why am I not doing more rallies?’” “He insists on having final approval over the songs on his campaign playlist, as well as the campaign merchandise, but he has never asked to see a budget for 2019,” the newspaper reported.