The hopeful case for Republicans concerned about Donald Trump’s persistent lead goes something like this: 

At this point in the last Republican primary, Rick Perry was pulling away from Mitt Romney. He briefly opened up a 10-point lead, and then just as quickly gave way to pizza magnate Herman Cain, who in turn gave way to Newt Gingrich, and so it went until Romney cleaned up in the end. Yes, Romney polled better than Jeb Bush is currently polling, but Romney was pretty much alone among establishmentarian candidates, while Bush is splitting that share of the primary electorate with two or three other candidates. Likewise, in 2012, the reactionary share of the vote was about the same as it is now—larger than the establishment share—but it wasn’t enough to win then, and it won’t be enough to win this time around.

The differences between the 2012 and 2016 fields makes this rosy scenario hard to envision. Which is why Republicans are having a harder and harder time articulating scenarios in which Trump becomes a non-issue before the primaries, for reasons that don’t involve nativist GOP voters undergoing a sudden, collective epiphany.

For this race to play out like the last one did would require a series of increasingly unlikely assumptions to come to pass:

1) That the establishmentarian field winnows sooner than later.

In past cycles, the eventual nominee has benefited from consolidating the establishmentarian vote early. Today, Bush is splitting that vote with Marco Rubio and John Kasich and probably to some extent with Scott Walker, none of whom is likely to exit the race anytime soon.

2) Trump fades.

Trump does not appear to be an extinguishable threat, the way Perry, Cain, and Gingrich were. Trump might have hit his ceiling, but there’s nothing in the offing right now that promises to drag him back down. If Ben Carson surges, that’s not going to be at Trump’s expense. It will be at the expense of the rest of the field.

3) As conservative also-rans drop out, their supporters break for establishment candidates, rather than for Trump.

This seemed plausible earlier in the race, but Trump is now not only leading the field, but he’s also polling well as a second-choice candidate, along with Carson and Rubio. As the field narrows, these candidates will be likeliest to benefit. If Trump never collapses, Republicans will have to count on primary voters to coalesce around someone preferable. But it isn’t clear that Carson’s actually preferable, and it also isn’t clear that Carson’s and Walker’s and Ted Cruz’s and Mike Huckabee’s supporters would break for Bush or even Rubio over Trump, after traveling so far with candidates who promised to meaningfully challenge their own party’s establishment.

Short of sabotaging Trump by changing the rules in the middle of the race, which would risk driving him to mount an independent candidacy, the race itself will have to take on a completely new character for Trump to lose steam. Otherwise, he will win.