Most polls are inherently dishonest just by the fact that they control the answer they want by asking the question to get the answer they want. If someone asks the question, "Is bacon good for you?" in Omaha, NE you're probably going to get 83% say yes but ask it in San Fransisco and 95% will say no. Depends on who is paying for the poll. They will always get the answer they want eventually. That's why I have a very dubious view of polls. I learned my lesson after believing the polls in 2012 election.
Polls should ask the following questions of representative proportions of men/women and among the races in a given state among registered voters:
-Do you plan on voting?
-If yes, who do you plan to vote for at this time?
First Q is yes/no. Second one is open ended- they should be able to say Mickey Mouse.
That gets you polls at the state level, with representative proportions in the state of the sexes and races among registered voters planning on voting. The questions are about as un-leading as you can get. Then post the results on an electoral college map.
