We've written about the problems political opinion polling has demonstrated in recent elections, most pointedly in 2016 when most polls predicted a Clinton win that never materialized. For another look at polling's problems and approaches to circumvent them, check out this article at American Greatness.
Asking variously who your social circle will vote for, or alternatively your neighbors, is one approach to getting beyond the social desirability bias issue. Another with less track record is asking which major party will carry your state. That wording puts even more distance between respondent and the expressed opinion.
----------
Let me note that polls which report percentages of Americans favoring each candidate are effectively useless. It is entirely possible to win the popular vote and lose the election, Clinton did so in 2016. The U.S. is a republic, not a plebiscite democracy.
Whether a candidate wins a state by one vote or a landslide makes no difference, the number of electors voting for him or her is the same. If most of a candidate's voters live in a few states where they run up huge super-majorities, he or she is at a disadvantage. This fact is normally a disadvantage for Democrats as their voters are more concentrated.
----------
Also, if you live in CA, NY, MA, AL, WY or UT, the campaigns mostly ignore your state. Everybody believes they know a priori who will win those states. Dems will win the first three, Republicans will win the second three.
It is the states where both sides have a decent shot at winning where the campaigns spend their time and money. These are usually labeled "battleground states" and if you live in one of them you'll be hectored by both sides till you are sick of it.
