Clinton widens her edge in the electoral college over beleaguered Trump

The electoral map in the United States may be in flux as the presumptive presidential nominees feel their way forward in a volatile election. But if the map is changing, it could be getting worse for Donald Trump.

The problems surrounding the Republican's presidential bid have been mounting. Trump's team is being outspent, out-staffed, and out-fundraisedby the presumptive Democratic nominee, Hillary Clinton. He continues to be outpolled by Clinton as well.

• Wherry:Trudeau'sfight with the forces of Brexit andTrump
• Sanders says he will vote for Clinton to defeat Donald Trump

But the national popular vote only tellsa small piece of the story. The real challenges for Trump are at the statelevel, where the electoral college decides who becomes president.

Based on a weighted average of polls, Clinton is projected to win 358 electoral college votes in an election held today, leaving only 180 for Trump. Even taking the swing states into account, Trump is not projected to be in a position to win more than 257 electoral college votes.

He needs 270 to take the White House.

This is a worsening portrait for Trump compared to last week'sprojection, asArizona and North Carolina haveswungfrom Trump to Clinton.

A recent poll by OH Predictive Insights in Arizona gave Clinton a five-point edge over Trump. Though that is a surprise in a state won by Mitt Romney by nine points in 2012 and which was last won by the Democrats 20 years ago it follows on a survey conducted in mid-May by PPP that gave Trump a lead of just two points in Arizona.

Two surveys conducted over the last week in North Carolina(by PPP and CBS/YouGov)put Clinton in a tie with Trump or gave her a two-point lead. Last month, two polls had given Trump an edge of only two or three points in North Carolina.

Romneywon the stateby two points in 2012. The state narrowly voted for Barack Obama in 2008.

The model still considers these two states to be extremely close, awarding Clinton the advantage by 0.5 points in North Carolina and just 0.1 point in Arizona. That means theyremain toss-ups. But while it might have been expected that North Carolina would be close in November, that Arizona is on the bubble suggests just how much the Clinton/Trump contestshakes up the electoral map.

Swing states getting swingier

But this is not only bad news for Donald Trump.Recent polls suggest a few Democratic-leaningstates are closer than they should be, too.

Based on national levels of support Clinton is currently averaging 47.8 per cent among decided voters, compared to 42.1 per cent for Trump states like Colorado and Pennsylvania should be relatively easy Democratic wins by a margin of about seven points. Ohio, too, should be in Clinton's column by about five points.

But new polls by CBS/YouGov and Quinnipiac University suggest that Clinton holdsleadsof just one point in Colorado and Pennsylvania. The polling is thinner in Colorado, so it is more difficult to judgethe latest set of numbers. But in five of the last six polls in the Keystone State, Clinton's leadhas been estimated at three points or less.

Pennsylvania has the potential to move towardTrump as other swing states follow the national trend line and slide further to the Democratic side.

Quinnipiacput Clinton and Trump in a tie in Ohio, though previous polls have had Clinton ahead by a more comfortable margin. Nevertheless, along with the Pennsylvania resultsit signals that the Republicans' strategy to focus on the Rust Belt makes some sense.

But Ohio, Pennsylvania, and Colorado combine for only 47 electoral college votes. Trump is currently projected to need 90 to close the gap on Clinton.

The destabilizing Others

While the usual swing states move in different directions due to the unusual candidacies of Clinton and Trump America's least popular presidential candidates in decades some states that have normally been written off as immovable are suddenly looking looser. And this is in part thanks to the Others.

Who are the Others? They are Gary Johnson, the presidential nominee of the Libertarian Party, and Jill Stein, the presumptive nominee of the Green Party.

At this stage of the campaign, these two third-party candidates (along withsupport for unnamed "others")are averaging 10 per cent of decided voters. Johnson's candidacy gets the bulk of these when his name is included.

• Libertarian could swing votes from Trump, Clinton

A number of states could be unbalanced by thesecandidacies. Maine, which would be expected to favour Clinton by about 17 points based on her national support, was instead given to her by just seven points in a University of New Hampshire poll. About oneinfive respondents opted for the "other" option.

In Texas, where Trump's lead should be about 14 points, the Republican nominee was given the edge by just eight points in a YouGov poll again with 19 per cent of respondents supporting an "other" candidate.

And in Utah, the reddest state in 2012, Trump was up just nine points in a poll by Dan Jones and Associates. In that survey,18 per cent supported an "other" candidate, 10 per centsupported Johnson, and twoper cent went for Stein.

Two other polls conducted earlier in June also showed significant support for third-party candidates in Utah.

So the election campaign is setting up to be an unpredictable one, withstates swinging in different directions based on how locals view these two unpopular candidates and their options. Clinton could win some states that Obama never picked up, while narrowly winning or even losing states that Obama had triumphed ineasily.

Butunpredictability has been the only constant of this U.S. presidential campaign.

• The Pollcast: Subscribe to ric Grenier's politics podcast
• Leader Meter: Track federal party leaders' approval ratings

This average of U.S. presidential polls includes all published mainstream surveys, a list of which can be foundhere. The polls are weighted by sample size and date, as well as the reliability of eachpollster as rated byFiveThirtyEight.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 to the 2012 presidential election. Surveys included in the model vary in terms of sample size and methodologyand have not been individually verified by the CBC.A full methodological explanation can be found here.