I have an observation to make about the polls, or a handful of them -- all to suggest that there will be a bigger landslide in the election than the polls suggest.
Reputable polls are performed with unbiased sampling techniques to assure a certain level of accuracy and random selection to avoid bias. In the case of the polls of interest here, the pollsters may choose as their population to be sampled a list of "likely voters." Likely voters are people in voter registration lists who at least had voted in the previous election -- therefore, they are "likely."
Trump has little in the way of a ground organization. Only now that the GOP has provided lackluster support for his candidacy, and in fear of losing down-ballot elections, would their ground organization be useful to Trump.
Clinton, on the other hand, has a ground organization in all the states, and it is tempered and experienced in get-out-the-vote efforts. They set up pop-up tents in malls to register new voters. They arrange for rides to the polls of people with transportation problems.
Finally, there is the ongoing wave of millennials, who are more likely to favor Clinton -- according to polls. They, too, will contain new registrants.
So the polls and the spread or lead they show for Clinton over Trump may be biased downwards. The spread in the sample results may be smaller than what materializes in November. And this isn't a matter of intentional bias in polls, but only a matter of "likely voters" already registered, and voters that haven't established a record of being "likely."
In this regard, I am hopeful, only because I think this election should be a sharp, resounding repudiation of Trump, his supporters, and their toxic ideas. If Clinton only makes it by a squeaker, it is less than an adequate victory.