I’m aware of four polls of the presidential race conducted since the McClatchy/Marist survey that I wrote about here. Two of them have Donald Trump ahead: Rasmussen by 7 points and USC/LA Times by 3. One poll, CBS News/New York Times, has the race dead even. The other, Economist/YouGov, has Clinton ahead by 2 points.
If one throws in the McClatchy survey (Clinton +3) and the two conflicting battleground state polls, all of which were taken after James Comey’s statement, it looks like the race is basically even. On this view, if Trump gets a meaningful bounce from his convention, he will find himself ahead. (Trump may not get a bounce, though; it probably depends on how he handles himself in Cleveland. In any case, Clinton may well get a bounce from the Democratic convention a little further down the road).
The poll analysis model used at FiveThirtyEight does not find the race to be even. Nate Silver states:
Despite a relatively poor run of polls, Clinton is very probably still ahead of Trump right now. . . .[Our] polls. . .model has [Clinton] ahead of Trump by 3.4 percentage points nationally, similar to the margin by which Barack Obama beat Mitt Romney in 2012.
Using this analysis, Silver says that Trump has a 36 percent chance of winning the election. That’s about twice as high as the chance he gave Trump less than a month ago, but not high enough for Silver to rate the race a toss-up.
The 3.4 point margin in Clinton’s favor that Silver finds is pretty consistent with the RCP poll average, which has Clinton up by 2.7 points. The question in assessing recent polls appears to be whether polls that predate Comey’s description of Clinton’s handling of sensitive emails should largely be discounted.
Whatever one makes of this race, Silver’s piece is worth reading. So is Sean Trende’s discussion.
For what it’s worth, and my recent track record isn’t good, I still think Clinton will defeat Trump. If anything, it seems to me that Silver overestimates the odds of a Trump victory.
But if Trump puts on a sane but passionate convention and the polls show a bounce, I may need to reassess.