Trump closes in on Clinton as U.S. election enters final stretch, polls suggest

Labour Day is traditionally when the U.S. presidential election enters its final stretch. The next nine weeks will feature a handful of debates and a lot of frenzied campaigning. They will also be decisive, as new polls suggest Donald Trump is closing the gap on Hillary Clinton.

The latest projections of CBC's Presidential Poll Tracker show that Clinton's lead over Trump has been cut in half over the last month. In the wake of her party's national convention, the Democratic nominee had opened up a margin of more than six points over her Republican rival in early August.

• The Pollcast: Politico's Steven Shepard on polls, 3rd parties and debates
• Libertarian Gary Johnson's presidential fate to be decided in coming days
• Presidential Poll Tracker: Follow the latest numbers

Now, Clinton is estimated to have the support of 45.4 per cent of decided voters, followed closely by Trump at 42.2 per cent. That 3.2-point margin is the smallest it has been in the national polling average since the end of the Republican convention in July.

The polls have tightened quickly. In polls conducted over the past week, Clinton has averaged a lead of just under two points among decided voters. Her lead was twice as wide in polling done the previous week — and Trump led in just one of 13 surveys done at the time.

On average, 12.4 per cent of decided voters say they will vote for third party candidates, including Libertarian Gary Johnson. With 15 per cent being the threshold for inclusion in the debates, it appears unlikely he will be on the stage with the two major candidates.

Swinging states

Clinton's lead in the electoral college is much more comfortable. If the election were held today, she would likely win 341 electoral college votes, compared to 197 for Trump. In order to win the White House, Clinton needs 270 votes.

But the certainty of her electoral college edge has decreased dramatically. As recently as Aug. 29, the Poll Tracker estimated Trump's upper electoral college vote potential (which assumes he wins all of the "lean" states) to be just 218. It is now 274, putting it just over what Trump would need to become the next American president.

Swing your own states

Our Presidential Poll Tracker has been updated to show how the state-by-state electoral college projections break down — and to allow you to swing states from one party to the other to put together a winning electoral map for either candidate. Try it out here.

Considering the projected margins in the swing states, the mark of 274 electoral college votes is well within Trump's grasp. In addition to the states he is projected to win more comfortably, he would need to take Colorado, Florida, Nevada, North Carolina, and Ohio, while also holding on to his lead in Iowa.

New polling done for the Washington Post, conducted in all 50 states, put Trump in a tie with Clinton in Colorado, ahead by one point in North Carolina, and up by four in Ohio. He was down by just two points in Florida and three points in Nevada.

This is a plausible map for Trump, though by no means an easy one. While the Washington Post survey had Trump in a good position in Colorado and Florida, polls done within the last month have shown much wider margins for Clinton — as high as the double digits. Polling in Ohio has been more consistently positive for Clinton in the past as well.

In Trump they trust (more)

But an enduring problem for Clinton is that a majority of Americans do not trust her. A recent IBD/TIPP poll showed that just 37 per cent of Americans considered Clinton honest and trustworthy, compared to 62 per cent who did not. They were more divided on Trump, with 46 per cent thinking him honest and trustworthy and 52 per cent thinking otherwise.

A CNN/ORC poll found when asked who they thought was more trustworthy, 50 per cent of Americans selected Trump and just 35 per cent selected Clinton.

But Clinton still has some advantages over Trump. The CNN poll put her ahead on who Americans trusted on foreign policy by 16 points, though Trump was ahead on the economy and the two candidates were closer on terrorism and immigration. On who has the "temperament to serve effectively as president," Clinton was favoured by 56 per cent compared to 36 per cent preferring Trump.

Trump's score on this question has increased by just four points since he became the presumptive Republican nominee in June, suggesting he has done little since then to make himself appear more presidential.

That makes the question of Trump's suitability for the Oval Office Clinton's most powerful rhetorical tool in this election campaign. That's why it remains the focus of many of her advertisements and public statements.

But while Trump played into Clinton's hands on this issue during the primary season and for much of the summer, his improving poll numbers and the past (relatively) controversy-free few weeks suggest he may be learning the lessons of the campaign trail. Can he keep it up for 62 more days?

The Pollcast: The state of the race

On the latest episode of the Pollcast, host Éric Grenier is joined by Steven Shepard, Politico's chief polling analyst, to talk about the polls in the United States, the role of third party candidates, and what impact the upcoming debates might have.

The CBC Election Pollcast18:569 weeks left before election day in the U.S.

We talk to the chief polling analyst for POLITICO, Steven Shepard, about the U.S. presidential race.

You can listen to the podcast above or subscribe to the podcast here.

The Presidential Poll Tracker includes all published mainstream surveys, a list of which can be found here. The polls are weighted by sample size and date, as well as the reliability of each pollster as rated by FiveThirtyEight.com. The electoral college is projected by applying the same weighting standards to state-level polls and combining this with a uniform swing model, based on how the current national polling average compares with the 2012 presidential election. Surveys included in the model vary in terms of sample size and methodology and have not been individually verified by the CBC. A full methodological explanation can be found here.