Donald Trump holds a campaign rally at the Giant Center in Hershey, Pennsylvania. (Getty)
Just a few weeks ago, conventional wisdom suggested that the 2016 election was essentially over and that Donald Trump had absolutely no chance of becoming president. Since then, a dramatic shift has taken place, with the Republican nominee rebounding and starting to catch up in many key battleground states like North Carolina and Florida. As a result, we’ve seen Trump’s chances of winning the election increase in the major election forecasts, and although all of them still think a Clinton victory is extremely likely, they disagree on exactly how likely it is. So what are Trump’s odds of winning as of November 6th?
One of the most popular election forecasts is FiveThirtyEight, the data analysis website run by Nate Silver. In 2012, Silver accurately predicted the winner of all 50 states, although he severely underestimated Donald Trump’s chances of securing the Republican nomination earlier this year. FiveThirtyEight currently gives Trump a 34.5 percent chance of winning the election. This is quite a shift from October 18th, when FiveThirtyEight’s model had Trump at a 12.6 percent chance of winning. In their model, Hillary Clinton’s odds of winning are now at 65.5 percent.
FiveThirtyEight also calculates the odds of certain very specific scenarios taking place. According to their model, there is an 11.5 percent chance that Hillary Clinton wins the popular vote but does not win in the Electoral College and therefore does not become president. On the other hand, there’s only a 0.6 percent chance that Trump wins the popular vote but not the Electoral College. They also calculate that the odds of no candidate reaching 270 Electoral Votes is 1.2 percent, the odds of at least one state holding a mandatory recount is 9.7 percent, and the odds of a Hillary Clinton majority, i.e. Clinton winning at least 50 percent of the vote, is 26.8 percent.
Another prominent election forecaster is The New York Times’ The Upshot, and they’re much more certain of a Hillary Clinton victory than FiveThirtyEight. According to their model, Donald Trump only has a 16 percent chance of becoming the next president. Just a few weeks ago, Trump was at his lowest point ever in The Upshot’s forecast, having just a seven percent chance of victory on October 22nd. The reason for Trump rising nine points is largely due to a new batch of polls from battleground states. The Upshot previously forecast that Trump had a 20 percent chance of winning Florida, but that has now jumped to 30 percent. He had a five percent chance of winning New Hampshire before, but now he has a 20 percent chance. Even in Pennsylvania, Trump’s odds of victory have risen from five percent to 11 percent.
The Upshot’s model is also unique in that it outlines every single possible way that Hillary Clinton or Donald Trump could be elected president, and this helps illustrate how narrow Trump’s path is compared to that of his opponent. According to The Upshot’s model, Hillary Clinton has 693 different ways to win the election, but Trump only has 315 ways to win. And if Hillary Clinton wins Florida, she has 469 ways to win from there, but Donald Trump only has 39.
Next up is PredictWise, which using a combination of prediction markets and polling data to calculate the two candidates’ odds of winning. They currently forecast that Donald Trump has a 13 percent chance of victory. Trump’s best odds in this forecast came at the end of September, when PredictWise gave him a 32 percent chance of winning. But after the three debates and after over a dozen women came forward to accuse Trump of sexual assault, his odds started to plummet, going down to just nine percent on October 22nd. In terms of the swing states, PredictWise forecasts that Donald Trump has an 18 percent chance of winning Florida, a 35 percent chance of winning North Carolina, a 68 percent chance of winning Ohio and an eight percent chance of winning Nevada.
Then there’s DailyKos’ election forecast, which currently gives Trump a 13 percent chance of winning the election. In their model, Hillary Clinton has actually not dropped quite as much as she has in some of the other forecasts. On October 30th, DailyKos had Hillary Clinton’s chances of victory at 96 percent, whereas they’re now at 87 percent. That’s still a significant decline, but in FiveThirtyEight’s model, Clinton dropped nearly 20 points in that same time period. It’s clear why DailyKos’ model is leaning so heavily towards Clinton when you look at their forecasts for the swing states. DailyKos predicts that Donald Trump has just a 26 percent chance of winning Florida, a 44 percent chance of winning Nevada, and a 40 percent chance of winning North Carolina.
Finally, there’s The Huffington Post’s election forecast, which is by far the most confident in a Hillary Clinton victory. They give Donald Trump just a 1.3 percent chance of becoming president, the lowest of any major forecast. They also predict that Hillary Clinton will win 317 Electoral College votes, whereas Trump will win 164. In the battlegrounds, they give Trump an 8.4 percent chance of winning Florida, a 9.6 percent chance of winning North Carolina, and a nine percent chance of winning New Hampshire. Some other forecasters have taken issue with The Huffington Post’s model for having Trump’s odds so low; on Twitter, Nate Silver recently railed against this forecast, saying that “a model showing Clinton at 98% or 99% is not defensible based on the empirical evidence.”
You might be wondering how all of these models can be so certain of a Hillary Clinton victory when, on a national level, Clinton and Trump are now only separated by 1.8 percentage points in the polls. Simply put, that’s because this is not a national election. It comes down to the battleground states, and it’s all about reaching 270 Electoral College points rather than actually winning the most votes. So even though Donald Trump is close to Clinton in the polls, the fact is that winning the election is much more cumbersome for him than it is for Clinton when you look at the Electoral College map. He has a realistic chance of winning Florida and North Carolina, for example. But if he loses either of them, he’s essentially out of the race. On the other hand, Clinton has such an advantage in states with a large amount of electors like California and New York that she can afford to lose a few battlegrounds and still be fine. Trump has no such wiggle room. This is all taken into account in the models, and as they suggest, even after the tightening of the polls, it’s clear that Hillary Clinton is still the clear favorite heading into Election Day.