Kevin Drum writes at Mother Jones, Does Donald Trump Really Have a 30% Chance of Winning?. He points to two articles that are both worth reading in full.
In the first, Trump expands the battleground…to Utah and the Deep South, Sam Wang points out
Historically from 1952 to 2012, the likely range of movement in two-candidate margin from this time until Election Day has been 10 percentage points, which is the standard deviation from the 16 past elections. Therefore, even though Clinton currently leads by a median margin of 7 percent (12 national surveys) and would certainly win an election held today, she could still lose the lead, and from a purely poll-based standpoint, is only narrowly favored to be elected President in November (probability: 70%).
It is also the case that Clinton is the only candidate who is poised for a blowout. Her “plus-one-sigma” outcome (current polls plus one standard deviation) is a popular vote win of 58.5%-41.5%. Trump’s plus-one-sigma outcome is a narrower win, 51.5%-48.5%.
He goes on to do his electoral college math and compares current state polls against 2012 election results finding only UT to be an outlier and otherwise a slight Democratic shift from Obama-Romney. The closest thing to an anti-Hillary state of the 15 tossups is NY, which is Trump's home state and even there Hillary is +20% while being -8% off Obama's results.
Every prognosticator at this point says that the election is far off and lots of things can happen, so we shouldn't be looking too closely at polls at this point. James Wimberley in Lies, damned lies and election statistics, looks at ": what reasonably foreseeable factors are capable of changing voters’ preferences for Trump or Clinton between now and the election?" He makes up a numerical estimate of risk factors which I ignore, but I found his (lite) analysis of ten factors a pretty interesting read.