President Donald Trump is more likely to lose the election than to win it. That does not mean he has no chance, however, and 2016 haunts most political prognosticators. It was a genuine upset when Trump's quixotic campaign won the GOP nomination and a moderate upset when it eked out an electoral college victory while losing the popular vote. This causes many of Trump's supporters to ignore completely  he extant data and many of Trump's opponents to game out worst-case scenarios.

The result of this dynamic is that anytime a new shock hits the American political system, Trump's supporters and opponents can agree on a narrative where Trump can spin dross into gold and coast to reelection. Take, for example, how the protests, looting and police overreactions of the past week might affect the November election.

The New York Times's Peter Baker and Maggie Haberman reported that "some in the president's circle see the escalations as a political boon, much in the way Richard M. Nixon won the presidency on a law-and-order platform after the 1968 riots. One adviser to Mr. Trump...said images of widespread destruction could be helpful to the law-and-order message that Mr. Trump has projected since his 2016 campaign."

Trump's advisers are not alone in this sentiment. CNN's John Avlon, the Atlantic's James Fallows and USA Today's Tom Nichols have articulated variations of this argument. Political science offers some backup as well. Omar Wasow, an assistant professor of political science at Princeton, just published the world's most timely paper in the American Political Science Review. Wasow uses the '60s protests to show that if protesters engage in violent resistance, the mainstream media will frame the conflict in terms of law and order, benefiting conservatives.

This is certainly what Trump wants to do, and he's not being subtle about it.

Will this work? It could.

But there are many reasons to believe that it will not play out the way it did in 1968.

First, as previously noted, the initial spark of violence this time around did not come from protesters, but from the police. Some protesters have engaged in violence, but many cops have engaged in disproportionate violence as well. In his paper, Wasow notes that "when peaceful protesters are the object of state or vigilante violence, the mainstream media are expected to issue frames that are especially effective for activists." In a contest as to which images will predominate, I'll put five bucks on the policy abuse ones.

Second, to run a law-and-order campaign, the candidate needs to offer a patina of reconciliation to make it sell in the suburbs. The toddler-in-chief lacks both the finesse and the self-control to be able to play this game.

"You have to dominate. If you don't dominate, you're wasting your time,” Trump told governors. “They're going to run over you. You're going to look like a bunch of jerks."

Maybe a softer version of this would have some positive effect; this kind of message is a pure base play.

Third, Nixon was running as the challenger from the party that had been out of power for eight years. Trump is the incumbent. Fair or not, what's happening now is on him.

Fourth, while everyone is paying attention to the social unrest right now, it's not like the country's other problems are going away. COVID-19 is still a pandemic. Forty million Americans are still unemployed. Absent therapeutics and vaccines, the United States is not getting back to normal. Those facts are likely to predominate as the campaign progresses, and they will not help Trump at all.

In the first few months of 2020, I heard that impeachment would benefit Trump, and then that his pandemic news conferences would help him. Neither worked. This could be due to deep polarization or to the fact that Trump is a bad president.

I have no doubt that Trump will try to run a law-and-order campaign. I have severe doubts that it will work.

Daniel W. Drezner is an American professor of international politics at The Fletcher School of Law and Diplomacy at Tufts University.