The paradox of media coverage of the 2016 GOP presidential race is that the longer Donald Trump dominates the polls, the more insistent pundits are that the maverick candidate is headed for a fall. “Donald Trump isn’t going to be the Republican nominee,” Ross Douthat bluntly stated in a column for The New York Times last week, although he admitted that this flat prediction was becoming more difficult to argue with conviction. As Douthat noted, the conventional wisdom that Trump is doomed to fail is an assertion that increasingly “inspires sympathetic glances, the kind you get when you tell friends that you think your new personal-investment strategy is sure to beat the market.”

Yet Douthat is not alone in thinking that The Donald is going to go bust (politically, that is). A broad spectrum of pundits—ranging from Ezra Klein at Vox to Nate Silver at FiveThirtyEight to John Fund in National Review—share this bearish outlook on Trump’s prospects. The pundit class has coalesced around the theory that Trump’s seemingly high level of support is a balloon ready to be punctured and that the Iowa caucus—now less than three weeks away—will be the occasion when the Trump campaign meets the pin that will prick its hopes. But these pundits might be underestimating how robust and intense the loyalty of Trump’s fan base is.

In late December, The Weekly Standard’s Bill Kristol predicted that when “Trump loses Iowa, the mystique disappears, [and] he’s just another candidate.” Writing in National Review, John Fund voiced a similar thought in an article titled, “Losing Iowa Could Be Trump’s Kryptonite.” 

Yet attempts to explain how a Trump loss in Iowa could lead to the demise of his campaign tend to be cloudy and hand-wave-y. Consider Ezra Klein’s projection at Vox: “But there’s another model of failure. Trump could just ... not win. He could lose the Iowa caucuses. He could fall short in New Hampshire. A loss in any early state might lead to a loss in every state. Losing a presidential primary is often like going bankrupt: It happens slowly, then all at once.” As Klein himself admits, “A lot of reporters and politicos believe something like this is going to happen to him. But the prediction is hard to talk through explicitly because it’s so maddeningly vague.”

In an attempt to flesh out this “maddeningly vague” sense that Trump will lose steam, Klein’s colleague David Roberts offers a theory of Trump’s supporters. Trump sells himself as a winner, Roberts contends, which makes his popularity brittle because it is subject to disenchantment if he ever loses. Like the proverbial rodents fleeing a sinking ship, Trump supporters will flee him once the stench of failure can be sniffed.

If your value proposition is that you’re a winner, your value evaporates the minute you’re no longer winning. Losing refutes a winner, and no one wins forever,” Roberts argues. “Trump’s vulnerability (like his strength!) is that his appeal is entirely personal, entirely based on the expectation that he’s a winner who will win.”  

There’s a smidgen of truth to this argument: Trump does ceaselessly talk about how he’s a winner. But he does so in the manner of a military leader like Patton or an athlete like Muhammad Ali, as a way of rallying his supporters and his own psyche for combat. Losing individual battles doesn’t refute such boastfulness; it only reinforces a sense that victory must be won.

To say that Trump’s appeal is “entirely personal” ignores the fact that Trump has won an enthusiastic fan base by taking hardline stances on immigration and terrorism. Pat Buchanan was on surer grounds when he told The New York Times that, under Trump’s influence, the Republican Party is likely to become “more nationalist and tribal and more about protecting the border.” Buchanan’s sentiments were backed up by Leo Martin, a 62-year-old machinist who told the Times, “The Republican Party has never done anything for the working man like me, even though we’ve voted Republican for years. ... This election is the first in my life where we can change what it means to be a Republican.”

As these remarks make clear, Trump’s support comes not just from who he is, but what he stands for and what he promises to do. As Fund acknowledges, focus-group research shows that Trump’s supporters display “remarkable loyalty to the real-estate mogul and scant interest in other candidates.” This loyalty is best understood as devotion to the nationalist and tribalist policies Trump is putting forward, rather than simple enthusiasm for Trump as a man. And losing a few primaries isn’t likely to make such devotion melt away.

Trump, for his part, has some experience bouncing back from losses outside the political arena. He knows how to craft a comeback story for himself. In the field where he claims to have mastery (business), he’s declared bankruptcy four times, but has turned that into a narrative of his cunning in exploiting existing law. The need to overcome adversity doesn’t necessarily tarnish a winner, but can instead reinforce the idea that he or she is a fighter and a hero. If Trump were to lose in Iowa, there are any number of ways he could turn the narrative to his advantage, either by implying trickery on the part of his enemies or by selling himself as a “comeback kid” if he wins another primary. 

On top of all this, it’s uncertain that Trump will lose Iowa, or if he loses whether the loss will be a significant one. Fund, like many others, points out that Trump might be weaker in Iowa because the caucus system, which requires not just casting a ballot but devoting hours to meetings, tends to weed out poorer voters (who lack resources to spend a day caucusing) and those who haven’t participated in the caucus before—both groups that skew toward Trump. But Byron York, writing in the Washington Examiner, reports that Trump is building a get-out-the-vote machine in Iowa that could overcome such hurdles.  

As of right now, the polling we have doesn’t support the idea that Iowa will be the anti-Trump firewall that his opponents are hoping for. According to the aggregation of Real Clear Politics, Cruz has only a narrow lead over Trump: Cruz is at 28 percent and Trump at 26 percent. Marco Rubio stands third at 14 percent. The most likely scenario is a close three-way race as Rubio improves his position. But Trump could easily spin such a narrow race in a way to make himself the winner.

It’s easy to understand why both the Republican establishment and many liberals want to see Trump disappear fast. He’s a toxic presence in American public life. But scenarios of a quick solution to Trump—some silver bullet or Kryptonite to finish him off in Iowa—simply don’t have plausibility.