Not at all. As I wrote long ago: I do not kid myself when the polls something I wish were not true. The quality polls are by far the best information we have on where public opinion is on any particular issue or person at any particular time. Below is a table showing how the Real Clear Politics margins for national races have compared to actual margins 2006 through 2018. "Bias" is the average of the differences (Democrats minus Republicans) indicated by the final Real Clear Politics Average of polls and the actual differences. Perfectly unbiased would be 0 and you can see that the calculated bias is very close to that at -0.01. Simple Average Error is the average of the absolute values of the difference so it gives the average of how far off each polling average was regardless of whether it was negative or positive. You can see that it is 2.01.css75 wrote:JohnStOnge wrote:
It's not week to week. This is a trend over a month since the Whistleblower report was released. Yes, one has to be cognizant of the fact that there is noise. But you can indeed see how different things are playing with the public. For example: One could tell from the polls that Trump was losing the PR battle over the late 2018/early 2019 government shutdown. And one can tell from the polls that Trump is losing the PR battle now.
Trump actually beat Clinton among independents in 2016. Right now the majority of independents think he should be impeached and removed from office. He's not doing well.
Of course, things can change. People in the middle swish back and forth all the time. Right now, though, his approach is not working.
You are right,it isn’t week to week, it is any poll out there at anytime that is anti Trump.
Sent from my iPad using Tapatalk
Note that the overall national average for the 2016 election was off by 1.2 percentage points, which is LESS than the overall average of 2.01 for the period. That alone tells you that the idea that there was some kind of historically bad poll performance in 2016 is complete nonsense.
Also, you can see that the maximum extent to which the polling average over estimated Democrat performance is 3.2 percentage points (2012) while the maximum extent to which the polling average over estimated Republican performance is 3.6 percentage points (2006). That gives you an idea as to the extremes with respect to how far off the Real Clear Politics average might be.
So when you see something like the situation we have now where Trump is 13.4 percentage points under water in the Real Clear Politics job approval polling average you can take it to the bank that he is clearly underwater to a substantial degree. At least something like 10 percentage points and possibly up to something like 16 percentage points.
What a good movie.