I wanted to give you a quick polling update, with a scattering of new polls out this morning and in recent days.
The big news is a group of Quinnipiac polls of Florida (Trump +3), Ohio (tie) and Pennsylvania (Trump +2). Those are daunting numbers if you’re operating on the assumption that Clinton is the strong or perhaps prohibitive favorite to be elected president in November. There’s an important caveat about these polls however. In recent cycles, Quinnipiac has had a decidedly Republican tilt, mainly because their models seem based on a distinctly whiter electorate than most see. To be clear, ’tilt’ or ‘bias’ in this context doesn’t mean bias in the sense we sometimes mean when we talk about a news organization. They’re not necessarily; they’re just seeing a distinctly more GOP electorate this cycle. Also notable, Ohio and Pennsylvania show little or no change from the Q poll last month. It’s only in Florida where there’s a big move.
Meanwhile there’s also a new McClatchy/Marist national poll out this morning showing a three point lead for Clinton (42% to 39%). That’s in line with a NBC/Survey Monkey poll from yesterday which also showed a three point spread.
All of which captures one of the notable dynamics of this race so far. Clinton is very consistently ahead. As you can see here from the PollTracker data set, Clinton has never been behind. Her current margin is 4.8%. Equities analysts would speak of benchmarks or fundamentals with a line at about 42% that Clinton can’t get below (in a composite of all polls) and Trump can’t get above. We also see lots of poll like this one from yesterday from Purple/Bloomberg which shows demographic subsets Trump is losing badly enough to seemingly put the race out of his graph. In the case of the Bloomberg poll, it shows Clinton is winning college educated voters by 11 points, a group exit polls suggest Romney won by 14 points.
Since Romney lost by a decent margin, you don’t need to be a statistician to realize that he can’t afford to underperform any of Romney’s number with major demographic groups without significantly excelling him at others. And at the moment, he’s not excelling him on any front.
So what’s the upshot? If you’re a Hillary supporter, the Q polls should be a source of concern but not worry. The Q poll has been consistently more friendly to the GOP. They appear to be modeling an electorate that is more white than recent elections suggest. But the move in Florida is substantial. (Also worth noting is that Monmouth yesterday released a poll with Trump ahead by two points in Iowa, another key swing state.) More to the point, the national numbers are just pretty close. They’re ranging from a 3 to 6 point margin in the PollTracker average. I would certainly prefer to see Clinton further ahead. But her margin also seems remarkably durable, despite the real and deep-seated doubts about her in the electorate.
(It is worth remembering that President Obama was never more than 3 or 4 points ahead of Mitt Romney in 2012 and actually fell slightly behind him for much of the last month of the campaign.) I suspect that is because key segments of the electorate are just close to off-limits to him. But it’s close. This seems to be the cost of our extreme polarization. Trump just can’t get below 40%.