There was a NYT poll of North Carolina earlier today that finds Clinton +7. Their methodology is a bit different (using the state voter file to target respondents) fitting their required strata, so this could be a bit outlierish (although this is generally considered a good but more labour-intensive method). But another NYT reporter tweeted that these numbers are very consistent with GOP internal polling.
Also shows a massive, +25% edge for Clinton in early voting...
http://www.nytimes.com/interactive/2...tionfront&_r=0