BTW, generally speaking, any given poll can be either good or bad, depending on methodology; what I trust most are poll averages of several polls as they smooth out whatever methodological mistakes each poll makes.
RCP generally gives that data. I also like to check polls against prediction markets and see if those match. Supposedly prediction markets are even more accurate. I wonder about that, but studies show it. "Wisdom of crowds" and all that and especially if they have to put real money on the line.
At the end of the day, though, though I like to watch the data trends, on election day, I assume all polls, personally, are voodoo, and go out and vote even if the candidate is supposedly 30 points ahead of their opponent.
A lot of people sit home on election day if they see polls like that, and then bad things happen. Especially because polls may be closing in the last days of the election and they stopped paying attention.
Yes, I think this was a factor in 2016. And maybe other times.
-- Tis an ill wind that blows no minds.
Malaclypse the Younger