Watching today's death toll, I wonder if I was too quick to concede the IHME model—which somehow registers a tripling of infections as a fatality drop 2 weeks later—deserves trust. My lay sense—I'd love to be educated out of it—is that a tripling shows up eventually. Maybe today. 
UPDATE/ It's only 4:15 PM EST. Today, I fear, is going to be tragically historic—and should cause a revisiting of some of our assumptions about where we're at with COVID-19. The *revised* IHME model projected 1,953 deaths by midnight tonight—i.e. nearly a third of a day from now.
UPDATE2/ To be clear: for those who don't know, the IHME model is being regularly *updated* to *conform to reality*. A week ago, it projected 1,863 fatalities for today; now, it's projecting 1,953; we're currently at 2,031; and there are 7+ hours left in the day. See the concern?
BREAKING NEWS/ IHME model increases projected COVID-19 death toll by more than 7,000 deaths.
I've *truly* never more wanted to be wrong than when I doubted the IHME model. But the ways it seemed wrong were so at the surface it was impossible for me—and many others—not to say so.
I've *truly* never more wanted to be wrong than when I doubted the IHME model. But the ways it seemed wrong were so at the surface it was impossible for me—and many others—not to say so.
BREAKING NEWS/ April 14, 2020 has just become—at only 4:44PM EST—the deadliest day in the U.S. novel coronavirus outbreak. The IHME model had predicted the peak in deaths would come last week. The model has since revised total COVID-19 deaths upward by more than 7,000 fatalities.
