One might reasonably ask how it is that the two major parties produced presumptive nominees for president who both have more people who dislike them than like them.

A Washington Post-ABC News poll released June 15 found that Donald Trump has an unfavorable rating of 70 percent and Hillary Clinton has an unfavorable rating of 55 percent.  A Rasmussen Reports poll released June 17 likewise found that only 38 percent of likely voters had a favorable opinion of Clinton (19 percent very favorable) and nearly identical results for Trump, with just 37 percent favorable (16 percent very favorable).  Conversely, Clinton was viewed unfavorably by 59 percent (49 percent very unfavorably), while Trump's numbers were 61 percent and 48 percent, respectively.

The answer has to do with the how primary elections are held.

What is known as public choice theory has long shown that when there are more than two candidates, the winner using the plurality method can be the candidate least preferred by a majority of voters.

For example, suppose there are three candidates and that the winner is the candidate who has the most votes.

Candidate A gets 34 percent of the vote, and Candidates B and C both get 33 percent of the vote.  Suppose that Candidate A is the least preferred of the three of those who voted for Candidate B and is the least preferred of those who voted for Candidate C.  The result is that Candidate A wins when he is the least preferred candidate by two thirds of the voters.

That is essentially what happened in the Republic primaries.  Because there were so many candidates, Trump was able to win a number of the early primaries when the circumstances were much like the example given above.  But because the plurality winner took the total number of delegates in a number of the early contests, he amassed more delegates than many thought possible.

The media portrayed the race as an outsider unexpectedly beating traditional candidates, giving him a good deal of free publicity, and the rest is history.

There are a number of ways to solve this problem.  One would be to allocate delegates based on the percentage of the vote the candidate received.  This is what is known as proportional representation.

There are a number of methods of addressing what happens with fractions when the number of delegates is not evenly divisible by the percentage of the vote.

However, the important point is that no candidate that is least favored by a majority can get a majority of delegates.

A second option is for voters to list their preferences in order.  The candidates are then awarded points based on where they are in this order.  In the example above, if someone ranked his preferences as Candidate B first, Candidate C second, and Candidate A third, then B would get 3 points, C would get 2 points and A would get 1 point.  The candidate with the most points is the winner.

There are numerous other ways to choose candidates.  The point is that if the major political parties want to avoid a situation where the nominee is the candidate who a majority of voters think is the worst choice, then they will need to choose a different method of primary voting from the plurality winner taking all the delegates.

Gary Wolfram is the William Simon professor of economics and public policy at Hillsdale College.