Can Trump come back? We should remember that it is only August, and the media will move on from the Khans. We should also recall that for all of Trump’s problems, Hillary Clinton has many of her own. But for her opponent’s missteps, the past week would have been dominated by an awful news cycle for Clinton, potentially stepping on her convention bounce (you can read Amy Walter’s excellent piece on this subject here). The two daily trackers this cycle – Reuters and the Los Angeles Times – have both shown Clinton’s lead either leveling off or declining, while the RCP polling average has been fairly steady for the past few days, so a regression to mean may be forthcoming.

But perhaps it is best to once again distinguish between the typical race and the Trump-Clinton one. Could a typical candidate overcome Trump’s current seven-point deficit in the polls? The answer is yes. In 2012, Nate Silver looked at the state of the race two weeks post-convention, compared it to the result on Election Day, and found that the median shift in the polls was three points toward the challenger. This makes some sense, as the incumbent party always holds its convention second, and we would expect the post-convention bounce to gradually fade. A similar shift this year would produce a result roughly similar to 2012, with Clinton winning by four points.