The survey comes at just about a perfect moment for Clinton. It's after Trump narrowed the gap thanks to his having consolidated the Republican vote, post-primary. It's after Trump sailed his campaign into two brutal news cycles, one focused on the allegations of fraud against Trump University and the other Trump's disparagement of the judge overseeing that case as "Mexican." These are not small things; the poll finds that 55 percent of voters were "bothered a lot" by the latter comments. Seventy-one percent of those asked were bothered a lot or a little by the Trump U issue. (Sixty-two percent were bothered a lot by Trump's past comments about women.)
The timing was also good because half of the poll didn't overlap with the aftermath of the attack in Orlando. Asked who they'd trust to do a better job should a similar event occur at this time next year, Trump holds a slight advantage over Clinton, in keeping with other recent polls on the subject.
You can see how good the timing was for Clinton on this chart of the RealClearPolitics polling average. After Trump clinched the nomination and while Clinton was still battling Bernie Sanders, Trump caught up. Clinton's transition into presumptive nominee status likely hasn't kicked in fully yet -- but you can see how support for Trump has been shrinking for a few weeks.
Clinton is likely also aided by President Obama's strong numbers. His approval rating is at 53 percent in the survey, his highest number since right after the 2012 election.
One of the more remarkable bits of data in the Bloomberg poll is that a full 55 percent of respondents say they would never back Trump. That number is boosted by women, 63 percent of whom draw that line in the sand (and who probably make up a big chunk of the 62 percent upset with his comments that we mentioned earlier). Only 43 percent say that about Clinton -- slightly less than those who say it about Johnson.
The poll found that more than half of Sanders backers are planning to support Clinton, with 22 percent leaning toward Trump and 11 percent to Johnson -- numbers that aren't great for Clinton, if they hold. Again, the party is just starting the consolidation process, so we'll see. It's clear, though, that Johnson benefits to a large extent simply from not being Clinton or Trump, and it's not clear what effect adding other third-party candidates, like the Green Party's Jill Stein, might have.
The standard caveats apply: It's still fairly early in the cycle (at this point in 2004, John Kerry was winning) and this is a national poll which means a lot less over the long term than do polls in competitive states. But there's another caveat worth pointing out: In June 2012, a Bloomberg/Selzer poll showed Barack Obama leading Mitt Romney by about the same margin, 13 points.
That led to some head-scratching and prompted Bloomberg to try and explain why its results were so far from other polls conducted in the same period of time. Mark Blumenthal of Pollster.com put it succinctly: "The most likely possibility is that this poll simply represents a statistical outlier."
As may this one. But in this contest and at this moment, that seems somehow less likely. The poll came at a very good moment for Clinton, in a race that she has led by a decent margin for quite some time. The recent narrowing of the gap between the two was an abnormality in the pattern more than a Clinton lead has been. This poll may be an outlier in the way it was in 2012. It may also be an outlier in that it is capturing a shift in the race back to what we'd seen for months.