STUMP » Articles » U.S. COVID Mortality: Winter is Coming » 6 October 2021, 18:49

Where Stu & MP spout off about everything.

U.S. COVID Mortality: Winter is Coming  


6 October 2021, 18:49

I did a few videos a few weeks ago, capturing U.S. mortality trends from 2020 into August 2021.

Here is the first video:

Here are two more, in case you don’t want to wait for the blog posts:

In all three videos, I look at not only the trend for the whole U.S., but also the largest states (plus NYC):

  • California
  • Texas
  • Florida
  • New York (minus NYC)
  • New York City
  • Pennsylvania
  • Illinois

With these states, I capture a few widely separated geographic areas, though yes, there are huge swaths where almost nobody lives that I didn’t capture in this.

But the main point I had wanted to make was that there were different patterns of the “waves of mortality” coming from COVID, that is connected to the climate of the area.

In the videos I use the excess mortality dashboard from the CDC. I have liked its interface for all sorts of purposes, but to begin with, I want to look solely at official COVID deaths.

Four waves of COVID mortality so far in the U.S.

For the following, I’m using this data: Provisional COVID-19 Deaths by Sex and Age, last updated today, October 6, 2021. Instead of the weekly deaths, we have deaths here counted by month, running from January 2020 through September 2021. The total in the data set I’m looking at is at just over 700K. [I’m omitting the few October 2021 deaths that have already been reported]

Here is the graph we’re going to be seeing over and over again for each of the geographies I will look at:

I have labeled four “waves”:

Wave 1: Spring 2020

The first wave was centered around NYC, where the pandemic hit very hard. Neighboring areas, particularly with people who commuted by train and bus (such as New Jersey more than Connecticut), saw the worst spread of the disease and the worst mortality.

Wave 2: Summer 2020

This did not hit everywhere in the country, but was mainly along the southern border of the U.S. I understand it was a hot summer, and likely people inside in the nice A/C blowing things around fueled this wave.

Wave 3: Winter 2020-2021

Everybody got hit in this one, though the hardest-hit areas in the first wave seem to have not been hit as hard. This really built up in fall 2020, but peaked in deaths at the beginning of January 2021.

This is the same sort of pattern we see with seasonal flu deaths, by the way, at a much higher magnitude of effect.

Wave 4: Summer 2021

There wasn’t much extra mortality in spring 2021, but in late summer we saw southern states hit again, and Florida particularly had bad numbers. This has been attributed to new virus variants, being worse in their risk profile in the original strains that started spreading in early 2020.

Wave 5?

Whether we get another winter wave remains to be seen, and many people are hoping that higher vaccination rates in the northern climes may lead to lower mortality effects. I will come back to that.


Being a geographically large state with cities that are major international travel hubs, I suppose technically they’ve seen all the waves.

But really the only substantial peak for California was that monster third wave that hit the entire United States. Their summer 2021 deaths so far aren’t that different from the summer 2020 deaths.


Texas missed out on the “fun” that was the first wave, but the summer waves hit both years (about similar magnitude), and the winter wave was worst.


I don’t want to crack too many jokes here — but it seems there’s no difference between summer and winter for Florida.

Interesting that their summer and winter 2020 waves were similar, but the summer 2021 wave was the worst they’ve seen thus far.

Don’t assume that “low vaccination rates” are sufficient to explain this. Or, rather, don’t assume supposedly “high vaccination rate states” will fare better when their turn comes.

Here is the Mayo Clinic’s vaccine tracker, and you can check for yourself if, say, Florida’s vaccination rates differ greatly from New York’s, as an example.

New York

Now we come to the “fun” geographies. This first one is New York state deaths, minus those from New York City. We will see why in a moment.

New York had a nasty spring 2020 and a nasty winter 2020-2021. It has had essentially no COVID effects in the summer.

Unlike the other states, not only is there no summer effect, but the winter 2020-2021 wasn’t as bad as spring 2020 in terms of height of peak, but it’s more that about the same amount of COVID deaths were spread out over a longer period.

New York City

This one is very interesting.

Hideous spring 2020, as we all know.

But the summer waves are nothing, and the winter wave is very small compared to that spring one. If you add up all those fall/winter deaths, it’s less than half of just the April 2020 COVID deaths for New York City. So the acuteness of the pandemic impact went way down.


Pennsylvania is somewhat similar to New York….

…. but with a much worse winter than spring.


Illinois is similar to Pennsylvania.

But it has a more substantial impact in spring 2020, probably due to Chicago.

Winter is coming: Place Your Bets

The reason I’m doing this other than I just love playing with numbers and looking at graphs is that I think people need to be prepared for another nasty winter, potentially.

I’m hoping it won’t be as nasty as last winter’s, but looking at Florida’s summer results, I’m not sanguine.

People think: “Oh, but Florida has lower vaccination rates than other states!” (You can check out your assumptions about vaccination rates here: Mayo clinic’s tracker and CDC’s map.)

Just firm up your prediction right now in your mind, before the results actually start coming in.

Being an actuary, I generally consider a range of “reasonable” outcomes, given what I’ve already seen.

It may be that pandemic mortality will seriously subside this winter, but I’m not necessarily betting that way.

In future posts, I will be looking at differential COVID impacts by age and by race/ethnicity.

Related posts:

August 2021: COVID and Simpson’s Paradox: Why So Many Vaccinated People are Among the Current Wave of Hospitalizations

April 2021: Mortality with Meep: Ranking the States (and NYC and DC) by Excess Mortality

October 2020: Mortality with Meep: Comparing COVID-19 with Historical Mortality and Prior Pandemics

Related Posts
Mortality with Meep: Student Deaths at a Large University
Mortality with Meep: The Sex Gap in COVID and Total Mortality
Memory Monday: The Spanish Flu Peak in October 1918 and More