‘Amateur Hour’ White House Graph Shows Covid-19 Deaths Hitting 0 in 10 Days

On Tuesday, the White House Council of Economic Advisors, an agency tasked with giving Donald Trump advice on the economy, tweeted a graph that looks like it was edited in MS Paint and includes a line, marked in red, that suggests deaths from Covid-19 will drop to 0 in a little over a week from now.

This obvious nonsense, which does not jibe with the latest projections or present reality, is spectacularly dangerous as the rightwing in the U.S. pushes for people to return to work even as the virus continues to claim more than 1,000 lives every day. Its existence was reported over the weekend by the Washington Post, which noted that the so-called “cubic fit” model, represented in red, was concocted by Kevin Hassett. Hassett is the former chairman of the Trump CEA from 2017 to 2019 and is now back working with the committee. He has a reputation for being wrong a lot and no background in infectious diseases, the Post notes. Nonetheless, “White House officials have been relying on [the cubic fit and] other models to make decisions on reopening,” the Post reported.

The problem with the cubic fit model in the graph, Justin Wolfers, a professor of economics and public policy at the University of Michigan, explained over the phone, is that it uses a wildly inappropriate technique to “smooth” the up-and-down wiggly curve of Covid-19 death data and then projects forward from it in a massive deviation from field standards. The result is an extremely misleading representation of the data that doesn’t match up with reality.

“The problem with any of these curve-fitting exercises is that we know a lot about Covid,” Wolfers said. “We know hundreds of thousands of people have it. We know on average 20,000 new cases or more are being discovered each day. We know it can take about two weeks from someone catching it to dying. All of that information is ignored by the curve-fitting, and all of that information is how you know there will definitely be people dying of Covid in 10 days’ time.”

To better visualize observed data, we also continually update a curve-fitting exercise to summarize COVID-19's observed trajectory. Particularly with irregular data, curve fitting can improve data visualization. As shown, IHME's mortality curves have matched the data fairly well. pic.twitter.com/NtJcOdA98R

— CEA45 Archived (@WhiteHouseCEA45) May 5, 2020

Wolf said that projecting, or “smoothing forward” with the cubic fit model doesn’t make sense, because all it takes for a cubic function to shoot upwards is a small shift in x, in this case, time. Moreover, smoothing a curve that fluctuates up and down in this way produces all sorts of strange artifacts, he said.

“This is just amateur hour stuff,” Wolf said. “There is no statistician, epidemiologist, or economist who would have drawn that graph. It’s not in the range of things that one would think to do.”

There are other issues with the graph. For example, it includes outdated IHME projections that predicted lower death numbers over time than the latest projection, which is also included and predicts 45,000 more deaths by August 1 than earlier iterations. Sitting next to outdated, more optimistic models makes the cubic fit seem more reasonable, even though it’s out of touch with the latest IHME projection.

Jason Furman, head of the CEA under Obama, had harsh words for the chart, saying “it might be the lowest point in the 74 year history of the Council of Economic Advisers” on Twitter.

“The ‘cubic fit’ is based on an approach to epidemiology that has long been absent from any serious epidemiological discussions. It made terrible predictions back in March and April,” he tweeted. “The functional form was chosen to get the result they wanted.”

In a tweet, the CEA stated that the curve-fitting was done to “improve data visualization,” which implies it isn’t meant to forecast, per se. But putting the model next to actual forecasts without any disclaimer on the graph itself says volumes more than a tweet.

Spokespeople for the White House did not respond to our request for comment.

While many will focus on the model’s bad math, the truth is that whether a crappy chart is the result of incompetence or malice matters little when it is abundantly clear that his administration doesn’t care about public health to begin with. The mass human sacrifice being planned to grease the wheels of capitalism is murderous and atrociously coercive, and its architects will tell each other (and us) whatever they need to in order to get it done.