Just like Brexit, I estimate that there are 2-5% of undecideds who really are for Trump but afraid to admit it to a pollster.
"The poll also measured if third party candidates are added to the survey, Clinton gets 39 percent with Trump at 37 percent. Libertarian Party candidate Gary Johnson gets 8 percent and Green Party candidate Jill Stein gets 4 percent."
If a pollster called me ... I wouldn't answer the phone. But if I did answer the phone, I'd say I support Johnson. (I'd consider him the most conservative candidate running.) But come November, if Trump is the Republican candidate, and Clinton is the Democrat, of course I'll vote for Trump.
I truly doubt Johnson will win 8% of the vote and Stein will win 4%. Johnson won 0.99% last time. If (conservatively) 1/2 of the Johnson voters can be expected to split for Trump and 1/2 of the Stein voters split for Clinton, that's 37 + 4 = 41 Trump, 39 + 2 = 41 Clinton.
Additionally, 1% "someone else", 3% "won't vote" and 7% "don't know" -- my guess is that these will break 2-to-1 for Trump. (Who wants to be called a racist-sexist-xenophobe?)
After 2 horrible weeks for Trump with the 'rats, MSM and eGOP all pulling out ALL the stops against him, these are amazing numbers.
QED = Landslide for Trump!
Bingo! I'm not whistling past the graveyard here but I do think there is a fudge factor of "The Wilder" effect voters. I think the closing days of the Brexit votes, the Globalists smugly believed their own bulls*t, aka pollsters telling them what they wanted to hear. I can say one thing you WILL NOT see any of the MSM touting any poll showing Trump in a lead in state polls getting him the required number EVs for the presidency.
You will see useless national polls trying to dispirit you with headlines like "Hillary appears to be closing the deal with voters, victory appears likely according to Nate Silverman". Nate the left's oracle or happy thoughts wiffed on the Brexit vote. Google how badly he missed it and his excuses.
This is what puzzles me in the question we all struggle with - Are these polls accurate?
We know the media is lying to us and part of the propaganda disinformation arm of the elite.
But companies publishing fake polls can't be a good business practice - I would think future customers would look at past results its not a psychic you are hiring.
However I've got my self into a state were I believe lots of the conservative talkers are in on the scam with the ruling gentry so bogus polls from companies getting real kick backs makes sense in my mind.
It also could be its impossible to get a accurate poll of voters with the methodologies employed. As more voters communicate differently and filter unwanted interruptions .
I think trending information is the only valuable real thing one can get from the aggregate of lots of polls. i.e realclearpolitics
The rest I think is a result of the initial conditions of the environment, the context in which the poll is taken contribute to a meaningless noisy residue when it comes down to predicting a voting action.
I think number nuts go back retroacatively look at the internals and then cherrypick what they want to explain away what happened.
Seem like its all part of some mass hypnotic collective delusion