Someone challenged me in an e-mail discussion to explain how Donald Trump can possibly lose the Republican nomination. I've said over and over why he isn't going to be the party's nominee, but, true, I haven't explained the how of it.

After all, he's leading in the polls in Iowa, New Hampshire and nationally. It's one thing to say early polls don't predict nominations, but how do we get from here to there? I see three different paths to his candidacy's end point:

1. Trump sees he isn't going to win and finds an excuse to drop out. This also covers the possibility that he doesn't have any interest in being president -- that he did this as part of some clever business strategy well beyond my understanding of branding and that he always intended to drop out.

More coverageOpinion and analysis about the 2016 presidential campaignCartoonsCartoons: The race to the presidency in 2016 CartoonCartoon: Early for 2016?

2. Trump fades, winding up closer to 10 percent of the vote or even a lot less, as people begin focusing more seriously on the election and shift to candidates supported by Republican opinion leaders. In other words, the process of discovery, scrutiny, decline that worked in the case Herman Cain in 2012, for example, applies to Trump in 2016. It could happen quickly, if he somehow manages to find the magic words to alienate his supporters. Or it could happen slowly, as Republicans who only have seen Trump so far discover several more qualified (and more reliably conservative) candidates in the race. Either way, in this chain of events, he isn't a factor by the time voters get involved in Iowa, New Hampshire and South Carolina.

3. Trump manages to hold on to a 25 percent or so share of the vote, perhaps even winning in a few places while the field stays large, but eventually gets clobbered once it comes down to him against a single opponent. Or he is defeated once the field is small enough that a non-Trump candidate receives a third or more of the vote. If this happens, it means that a celebrity candidate can win and retain a fair amount of support no matter what he does but that his support is capped well below the total he would need to win a presidential nomination.

I'd risk a guess that the third of these outcomes is the least likely. Trump's surge, to this point at least, is only a reminder that short-term public opinion isn't very predictable.

advertisement | advertise on newsday

True, plenty of candidates who have had unconventional qualifications and who displayed unconventional political behavior have won statewide primaries for senator or governor, especially in multicandidate fields. And in presidential races, plenty of candidates who were unlikely occupants of the White House have won single primaries -- sometimes more than one.

But the long, sequential process in presidential nominations -- with week after week of scrutiny and time for party actors to react -- prevents those fluky winners from coming close to being nominated. The politicians, campaign and governing professionals, formal party officials and staff, donors and activists, and party-aligned interest groups and the partisan press -- the people who care the most about nominations -- have plenty of opportunities to repel someone if they reach a consensus against him or her.

Think Trump really has a chance? Remember: Endorsements by high-profile party leaders such as members of Congress and governors are much better predictors of eventual results than are early polls. And Trump is still taking a goose egg on those.

Jonathan Bernstein is a Bloomberg View columnist covering U.S. politics.