According to FiveThirtyEight: "... Polls of the November 2016 presidential election were about as accurate as polls of presidential elections have been on average since 1972." When the margin of victory may be relatively small as it was in 2016, why don't polling agencies simply gather significantly larger samples to bring this error close to zero

The littlest differences or changes in poll results are always opinionated with deep meaning from different perspectives, though they are at most times, overstated.

Now that we have seen that the predictions are driven by the sample size and the size of the bias. A larger sample size has reduced the sampling error.

Margin of errors are disclosed so that people will have a deep knowledge and understanding of how much precision they can reasonably expect.

Unfortunately, if there is bias, then the predictions are close to the biased estimate. If the bias pushes the prediction from one candidate(Trump) to another (Clinton), then we have a ”surprise” upset.