What Donald Trump is doing on the campaign trail

U.S. Republican presidential nominee Donald Trump speaks at a campaign event at Trump Doral golf course in Miami, Florida, U.S. July 27, 2016. REUTERS/Carlo Allegri (Carlo Allegri/Reuters)

By Saturday night, after polls close in South Carolina, Donald Trump is poised to have a gigantic lead in the Republican delegate race. By no means an insurmountable one, of course, but a big Trump victory there will start to raise questions about where, if anywhere, he can be stopped.

Unlike Iowa and New Hampshire, delegates in South Carolina are allocated by a modified winner-take-all system. If Trump wins South Carolina -- which a new CNN/ORC poll suggests he's still well-positioned to do -- he gets 29 delegates, without qualifications. That's only slightly fewer than all of his competitors have to date, combined.

On top of that, he gets three additional delegates for each congressional district he wins. He won't win the state without winning at least a few of the seven districts; he could certainly win them all. If he does, he'll walk away with 50 delegates -- just about the same number that were awarded in Iowa and New Hampshire combined. The scorecard, as tallied by FrontloadingHQ, would be Marco Rubio 10, Ted Cruz 11 -- and Donald Trump 67.

That's a 6-to-1 margin, but with relatively few delegates in total. Trump will need 1,237 to win the nomination, meaning that even a South Carolina sweep would put him at only 5.4 percent of what he'd need. Or, put another way, Trump will be in the sweet position in which Newt Gingrich found himself after winning South Carolina in 2012, on his way to coming in fourth in the overall nominating contest.

As the weeks pass, more and larger states will weigh in. South Carolina could be the high-water mark for Trump's lead.

Or, it could not. Sam Wang of the Princeton Election Consortium has looked at the allocation systems for the early Republican states and figures that, for Trump not to win the nomination, one of two things needs to happen quickly: Either he needs to stumble badly, or a bunch of his opponents need to drop out.

The problem for the non-Trumps is that the system is working against them. After South Carolina, the big contests on Super Tuesday (March 1) and March 15 will distribute delegates proportionally. The winner gets more delegates than the losers, but the losers get something.

As Wang notes, though, it's not a 1-to-1 ratio. The winner often gets proportionally more delegates than those further down in the rankings; according to Wang's models, winning 30 percent of the vote through March 1 would yield 50 percent of the delegates. That's in part because candidates receiving too low a percentage of the vote would get no delegates at all.

By the time March 15 rolls around -- less than a month from now, we would note -- nearly half of the total available delegates will have been allocated. After March 15, that number jumps to more than 60 percent. There are a lot of contests between now and then, but if the overall dynamic doesn't change -- a Trump lead with other candidates trailing -- Trump will have banked a lot of delegates.

The other question raised by a Trump victory in South Carolina, though, is why that dynamic would change. In CNN's poll, Trump beats Ted Cruz by a wider margin among evangelical voters than he does among voters overall -- a voting bloc that is supposed to be in Cruz's corner. (In Iowa, where the majority of Republican voters identified as evangelical, Cruz won evangelicals by 12 points. In South Carolina, Trump leads with them by 20.) Trump also leads with conservative voters in South Carolina, although by a smaller margin than he leads overall.

If that holds, it's not clear why Trump wouldn't then win most of the other Southern states that make up the March 1 primary, much less most of the other states between now and March 15. Sure, he probably won't win a majority, but he'll keep increasing his delegate lead for the next month. In other campaigns one might assume a stumble or gaffe that could shift perceptions, a risk to which Trump still seems immune.

Which leaves a consolidation of field as the main risk factor for a Trump nomination. There's not much incentive for the field to narrow much before March 15, since that's the day of the Florida and Ohio winner-take-all primaries. Jeb Bush, Marco Rubio and John Kasich will all want to stick around to get those delegate boosts, if possible. (Ben Carson also claims Florida as home, but the odds are very good he'll drop out of the race at some point sooner.)

Even if the field narrowed to only two or three candidates, that's still not a guarantee Trump would come up short. While Republicans have often been more likely to say they'd never vote for Trump than they are to say that about any other leading candidate, it's not the case that more than half of Republicans say it. Since the beginning of the campaign, the number of Republicans categorically ruling Trump out has fallen dramatically.

Trump could still lose, of course. You can see all of the cracks in the wall described above, and, frankly, I still think a non-Trump nominee is more likely than a Trump nominee. For example, I suspect that the turnout operation that benefited Cruz in Iowa will tighten South Carolina significantly -- if Trump even wins.

If Trump wins South Carolina, though, something would need to change dramatically and quickly for Trump not to be the prohibitive leader in the Republican contest by one month from today.