You are absolutely correct. Better to ignore the NYT, however, which is fairly radical now, and go right to the source the NYT writer cites, which is the CDC - here's the link . . . Excess Deaths Associated with COVID-19 (cdc.gov) . I recommend scrolling down to the blue histogram, which shows the data. The Hopkins author pointed out that the peak weekly death rate in '18, about 67,664, was only exceeded by 12,000/week at the peak of Covid-19 mortality in March, and was only exceeded for 4 weeks during that peak period. In other words, total deaths in U.S., per week, since Covid began, were less than those experienced week-ending Jan 13, 2018 for all but 4 weeks. Her point was that non-Covid-19 deaths have been swapped with Covid-19 deaths. Others have extended her point to - unscientifically - attempt to minimize what this is, by making the point, "well if fewer people died in pandemic week ending Dec. 14, 2020 than did pre-pandemic early January 2018, then how bad can this pandemic really be?" Terrible logic, of course.
This is a severe pandemic, no doubt. How severe? That's a question for history, I suppose.
Thanks, the pictogram is indeed helpful. Looks like the Hopkins author picked the one brief period during January 2018 when, for a short time, actual deaths exceeded predicted deaths, and used that as a basis of comparison to indicate that things aren't that bad. You concede this is terrible logic, and I would add totally misleading.
The pictograph shows that except for the January 2018 period, actual deaths (the blue bars) have been trending below predicted deaths (the yellow line) consistently up to March 2020, when actual deaths began to consistently exceed predicted deaths in a months long trend. Add up all those blue bars above the line and you get the meaningful picture.