NBC/Survey Monkey: National Clinton +8 (2-way), +5 (4-way)

[Speed of Sound]

Among registered voters.

2-way:

Clinton 50
Trump 42

4 way:

Clinton 43
Trump 38
Johnson 11
Stein 5


http://www.nbcnews.com/politics/2016-election/trump-faces-hurdle-minority-voters-clinton-maintains-lead-poll-shows-n636061

[windjammer]

Hillary's margin has a bit tightened, like from +9 to +7, but I'm confident it will grow again in the future

[ElectionsGuy]

I think taking the average of the two is accurate, since Johnson/Stein numbers will eventually fade and probably half of it will go to Trump/Clinton or stay home. So Clinton up 6 to 7 is what I would say.

Trump is getting 8% of blacks, 22% of Hispanics, and 23% of Asians. Also interesting split on perception of gender equality, nearly a 50 point gender gap on that perception.

[Fmr President & Senator Polnut]

I think taking the average of the two is accurate, since Johnson/Stein numbers will eventually fade and probably half of it will go to Trump/Clinton or stay home. So Clinton up 6 to 7 is what I would say.

I think that's fair.

[dspNY]

I still maintain that the D v. R head to head is more accurate since third party support usually collapses on Election Day. If we're still seeing these numbers after all the debates then it is time to at least include Johnson and run 3 way heats

[Erich Maria Remarque]

I still maintain that the D v. R head to head is more accurate since third party support usually collapses on Election Day. If we're still seeing these numbers after all the debates then it is time to at least include Johnson and run 3 way heats

Third party support is usually stable after conventions. But this election we had early conventions, so I'd wait polls after Labour Day.

Debates? Honestly, it wouldn't shock me if third parties support would INCREASE after debates

[Classic Conservative]

I can't get the crosstabs to work.

[Mr. Morden]

I still maintain that the D v. R head to head is more accurate since third party support usually collapses on Election Day.

In '96 and '00, poll #s for Perot and Nader respectively stayed pretty flat from late August until November, and were pretty close to their actual election day performance.

[Maxwell]

rofl lmao the great tightening of... 1 point.

[Dr. Arch]

rofl lmao the great tightening of... 1 point.

Which could just as easily be noise.

[Xing]

The race may have tightened slightly since Trump's "Khantroversy", but he's still far behind.

[Panda Express]

I can't get the crosstabs to work.

I'll just go ahead and tell you - you're losing everyone and everything except old white uneducated men

[Erich Maria Remarque]

I can't get the crosstabs to work.

I'll just go ahead and tell you - you're losing everyone and everything except old white uneducated men

Lol

[Dr. Arch]

I can't get the crosstabs to work.

I'll just go ahead and tell you - you're losing everyone and everything except old white uneducated men

[NOVA Green]

Again, another data point that would appear to indicate that the race has not really tightened over the past week, although the Pew Poll definitely got some attention.

Reality is that we'll need to wait for a few more decent national polls to see if the race is basically static or if there has been a tightening around the margins.

At this point, state polls don't appear to indicate any tightening of the race, so is there a lagging/leading indicator that could change that?

Appears to still be in a holding pattern.... we'll see if there is a breakout either way after labor day when people aren't out in vacation, enjoying the end of the Summer, etc, and start paying more attention to the election.