NY Times prediction now on 95% likely. So we now have statistical significance.
Not entirely sure about their methods for making this prediction, so brenden is certainly correct to express some doubt as to its veracity. My major concern is that the prediction has come from 83% Clinton at the start of the night to >95% Trump now. The momentum has been with Trump all night and, as each result comes in, it becomes increasingly clear that the polls have underestimated the support for Trump.
Incidentally, foreign media outlets have more leeway with what results they can report and what predictions they can make. Even with the Western states factored in for Clinton, the ABC (Australia) still has Trump winning, although there's still a fair degree of uncertainty surrounding that.
At the moment, Trump is only 4 ahead on those states declared, so there is hope still for Clinton. She would have to have a fairly significant turnaround in a fuck tonne of states atm though, which seems increasingly unlikely.
As in the UK, today's results highlight the difficulty of obtaining accurate polling data in counties where voting isn't compulsory. Recall that in the UK, Labour was supposed to romp home in the 2015 general election (they didn't; they actually went backwards) and Brexit was also meant to be summarily defeated, it wasn't. Sad to think that this could all have been avoided if the US (and the UK for that matter) had compulsory voting.