As of August 2016, not many people believe that Trump will win. Since spring, Clinton has been significantly ahead in the polls. Present odds with bookmaker sites and political prediction sites place Clinton’s chances of winning in November around 66% and Trump’s around 33%.
However, basing one’s political forecasts on the polls tends to come with a lot of pitfalls. They don’t really tell you this in the media, but opinion polls tend to be pretty unreliable. And there is evidence to suggest that they’ve been getting more and more unreliable in recent years.
To give but a few examples, in August of 1988, Michael Dukakis was ahead in the polls by 17 percentage points. However, the winner of that race wasn’t Mike Dukakis, it was George Herbert Walker Bush. In August 2012, Romney had a lead over Obama as well.
So let’s look at some other measures. One is simply to count the number of primaries one. Electoral history suggests that the candidate who does best in the primaries will also win the general election. Trump did better in the primaries than Clinton.
Another way to approach the matter is through public choice research. Public choice predicts that after two terms or more, the voters will start craving change and be more inclined to vote for the other party. This phenomenon is referred to as “the cost of ruling,” and empirical studies, by the professor of economics Martin Paldam among others, have shown that the average cost of ruling amounts to a loss of 2,25 percentage points of popularity over the course of a term. Clinton will thus have a five percentage point popularity penalty come November.
Finally, a model that has received a lot of attention recently, and which factors in the cost of ruling as well, is the Primary Model, as proposed by professor Helmut Norpoth. This model has a better track record than most other models in political science, and predicts that there is a 87% chance that Trump will become the next president.
So all in all, there is serious cause for doubting the media’s narrative and to take seriously the possibility that Trump might win in November.