“Like magic” and “NO WAY!”

These are the latest phrases President Trump is using to express continuing disbelief that Hurricane Maria could have caused nearly 3,000 excess deaths in Puerto Rico. In his mind — or in his tweets, at least — the death toll remains in the double digits.

Multiple times last week, Trump lashed out at a study from George Washington University’s Milken Institute School of Public Health that estimated 2,975 “excess deaths” occurred on Puerto Rico in the six-month period after Hurricane Maria struck the U.S. territory.

First, on Thursday, Trump falsely accused Democrats of inflating the death toll as a political ploy to make his administration look bad. Late Friday night, the president tweeted again, offering no evidence but alleging that “they” had hired “GWU Research” to arrive at a death toll that was “FIFTY TIMES LAST ORIGINAL NUMBER — NO WAY!”

There are several claims in or related to Trump’s latest pair of tweets that have already been explained or debunked. But because the president this time appears to be questioning the GWU study’s methodology, as well as expressing a strong (read: ALL-CAPS) doubt that a death toll can jump from the tens to the thousands, let’s break those ideas down.

How did GWU researchers arrive at the 2,975 figure?

To reiterate, the GWU study estimated 2,975 “excess deaths” occurred on Puerto Rico in the six-month period after Hurricane Maria struck the U.S. territory. The emphasis here is on the word “excess”: Using projections from past census data, researchers first estimated what the death rate in Puerto Rico would have been between September 2017 and February 2018 had the hurricane not happened at all. They also took into consideration the population change caused by people moving away from the island for good in the months after Hurricane Maria.

Evaluating all those factors, the study concluded that anywhere from 2,658 to 3,290 excess deaths took place between September and February. (The 2,975 number represents the midpoint of that range.) Every social stratum and age group was affected by excess mortality, the study noted, although populations in lower-income areas were hit harder.

As The Washington Post’s Arelis R. Hernández, Samantha Schmidt and Joel Achenbach reported when the study was released, GWU researchers intentionally evaluated a long period of time after the hurricane to see whether the death rate would taper back to normal after a while. It did not:

People continued to die at anomalous rates long after the storm, as the territory struggled with infrastructure failures and political infighting. Nearly 900 excess deaths were reported in January and February of this year. The mortality rates remained high in the poorest areas, the study found. 

The GWU report has a limitation: It does not specify how people died. It is a statistical study based on death records and expected mortality rates. The researchers said they hope to conduct a more detailed investigation in the future. 

... the leaders of the research effort said that in low-income areas the mortality rate remained somewhat elevated even after six months. They said further investigation of mortality rates after February could push the estimate even higher.

Notably, the study was commissioned by Puerto Rican Gov. Ricardo Rosselló, whose administration had been criticized for its response to the hurricane and for its unwillingness to accept earlier reports of a death toll higher than 64. When the GWU report was released in August, Rosselló, a onetime Trump ally, publicly accepted the findings and promised that his administration would do better.

On Saturday, Lynn R. Goldman, the dean of the Milken Institute School of Public Health, defended the GWU study and denied that politics had played a role in how it was conducted.

“Make no mistake: The death toll did continue to rise in the months after Maria,” Goldman wrote in a guest column for The Post. “In September 2017, when Puerto Rico recorded a total of 2,906 deaths, we found there was an excess of 574 deaths above what would have been expected in a year without the storm. The death toll continued to mount every day, with an excess of 697 deaths in October, 347 in November, 479 in December, 558 in January, and 320 in February, for a total of 2,975.”

What about all these other numbers we’ve seen floating around? 16? 34? 64? 4,645? 16,608?

There have been quite a few numbers that have appeared in reports about the death toll in Puerto Rico. No wonder it’s confusing. Here’s when and how each of those numbers entered the conversation — starting with the two figures included in Trump’s latest tweets.

On Friday, the president selectively quoted from a report by The Post’s Philip Rucker, Robert Costa and Josh Dawsey that asserted officials had told him the official death toll was 16 when he visited Puerto Rico last October. This was about two weeks after Maria made landfall — during the same trip, Trump tossed paper towel rolls into a crowd in San Juan.

"Sixteen people versus in the thousands,” Trump told the crowd then, comparing the presumed death toll to that of Hurricane Katrina. “You can be very proud of all of your people, all of our people working together. Sixteen versus literally thousands of people. You can be very proud.”

That number was still a moving target, however. Later that day, the official death toll jumped to 34. By then, Trump had already left the island.

In December, Puerto Rican officials once again changed the death toll, this time to 64, after taking into account deaths that were ruled as being connected to the storm.

For months afterward, 64 remained the official death toll — but only because Puerto Rico’s government refused to release further death reports while GWU researchers conducted their study, as The Post’s Glenn Kessler reported:

In the chaos of the natural disaster, officials had failed to properly document whether deaths on the island were related to the storm. Ordinarily there would be a better distinction between “direct deaths” — such as people who drowned in a storm surge, were hit by lightning or died when a building collapsed because of winds — and “indirect deaths," such as heart attacks, house fires, electrocutions from downed power lines, vehicle accidents on wet roads and so forth. So the government froze the death toll at 64 and asked George Washington University’s Milken Institute to examine the data. 

That didn’t stop other media and research groups from coming up with their own rough calculations of how many excess people had died in the wake of Hurricane Maria. As Kessler noted further in the same article:

In May, Harvard’s T.H. Chan School of Public Health published a study that estimated there had been 4,645 excess deaths in Puerto Rico in a roughly three-month period after Hurricane Maria made landfall there. That study was problematic in large part because it extrapolated death estimates from a survey of 3,299 households (rather than from actual death records like the GWU study that would come later).

“Given that this report is based on a survey, with potentially huge margins of error, it should be treated cautiously,” The Post’s Fact Checker determined shortly after the Harvard study was published. “Five other studies, based on preliminary death certificate data, have all come up with much lower numbers — about 1,000 in the two weeks after the storm.”

As for 16,608, that would have been the death toll if anybody from Puerto Rico who “died for any reason, like old age” had been added to the official Hurricane Maria death count — as Trump has accused officials of doing. But that is not what happened.

Is there an official “list” of those who died in Hurricane Maria and its aftermath?

Ah, the elusive list. Trump on Thursday tweeted that “if a person died for any reason, like old age, just add them onto the list.” In reality, there is no such list. At least, not yet.

A joint investigation by the Associated Press, Puerto Rico’s Center for Investigative Journalism and Quartz has, however, detailed names and stories behind 487 of the people in Puerto Rico who died in the aftermath of Hurricane Maria. It is so far “the most extensive record yet of who died and why.” (The emphasis is ours.)

More names and stories are likely to emerge publicly as time goes on, the investigation noted.

“This was a slow-motion, months-long disaster that kept Puerto Ricans from getting the care they needed for treatable ailments, even as President Donald Trump lauded his administration’s response,” the AP reported.

Why is the death toll politically important to Trump?

In short, Trump seems to consider the Puerto Rican death toll like an inverse to his inauguration crowd size of 2016: The lower he can get the number, the better he looks. Trump has bragged that the federal response to Maria was “an incredible unsung success,” contrasting it with “a real catastrophe like Katrina,” in which an estimated 1,833 died.

As The Post’s Philip Bump pointed out, Trump’s use of Katrina as a rubric against which his administration should be evaluated “hasn’t held up well” now that the GWU report estimates more than 1,000 more people died in Puerto Rico.

Trump’s refusal to accept that the Puerto Rican death toll from Hurricane Maria is (a) potentially in the thousands and (b) not politically motivated could cost him. Scores of Democratic lawmakers, as well as some Republicans, denounced Trump’s remarks this week.

Rep. Ileana Ros-Lehtinen (R-Fla.) accused Trump of “making it about himself” and called his tweets “so low.”

“It’s all about having people not believe anything they see or hear, so even life or death has a political spin,” Ros-Lehtinen told The Post.

Even Florida Gov. Rick Scott (R), a Trump ally who is campaigning for a Senate seat, uncharacteristically distanced himself from the president.

“I’ve been to Puerto Rico 7 times & saw devastation firsthand,” Scott tweeted. “The loss of any life is tragic; the extent of lives lost as a result of Maria is heart wrenching.”

House Speaker Paul D. Ryan (R-Wis.) said he had “no reason to dispute” the numbers in the GWU study but stopped short of admonishing Trump, instead seeming to characterize the high death toll as an inevitability of Hurricane Maria.

“You couldn’t get to people for a long time on the island because roads were washed out, power was gone and the casualties mounted for a long time,” Ryan told reporters Thursday. “So I have no reason to dispute those numbers. Those are just the facts of what happens when a horrible hurricane hits an isolated place like an island.”

Read more:

Study’s authors: We calculated the deaths from Hurricane Maria. Politics played no role.

Fact Checker: President Trump’s Four-Pinocchio complaint about the Maria death toll figures

Puerto Rico after Maria: Residents see a failure at all levels of government

Trump creates political storm with false claim on Puerto Rico hurricane death toll

Trump is rebuked after questioning number of deaths attributed to Hurricane Maria

‘It totally belittled the moment’: Many look back in dismay at Trump’s tossing of paper towels in Puerto Rico