COVID-19 USA Cases and Death total: What a difference a day makes!

Occam’s Razor

When you put it to the test
Occam’s Razor we suggest
Is the one stressed you  knew
You may truly pursue
The simplest answer’s the best

Alan Grace
12 May 2020


We look at yesterday’s USA deaths:


It now looks like there is a midpoint around 21 – 23 April.

We use the cumulative number of deaths on these dates to estimate the total number of deaths:

Date Cumulative 21+22 22+22 22+23
21-Apr 45,536 93,430
22-Apr 47,894 95,788
23-Apr 50,234 98,128

As usual we add on 10% to allow for a long low tail:

Deaths Add 10%
Date Cumulative 21+22 22+22 22+23
21-Apr 45,536 102,773
22-Apr 47,894 105,366.8
23-Apr 50,234 107,940.8

This gives us an estimated range for the number of USA deaths of

102,773 to 107,940 deaths

We round this to give us a range from 103,000 to 108,000 deaths with an estimated total number of 105,500 deaths.

Below are estimates for deaths in the USA for 6 June and 13 June:


Similarly, more tentatively, we estimate the total number of cases.

It looks like the midpoint for cases may be around 19-21 April.


As above we get these tables:

Date Cumulative 19+20 20+20 20+21
19-Apr 770,014 1,568,159
20-Apr 798,145 1,596,290
21-Apr 824,229 1,622,374

Once again we add on 10%:

Cases Add 10%
Date Cumulative 19+20 20+20 21+22
19-Apr 770,014 1,724,975
20-Apr 798,145 1,755,919
21-Apr 824,229 1,784,611

We estimate the total number of COVID-19 cases in the USA has a range from 1.7 to 1.8 million cases with an estimated total of 1.75 million cases.

We note that 105,366.8 / 1,755,919 = 0.06000664

i.e. A death rate of almost exactly 6%.

Currently in the USA there have been just over 1.38 million cases to date.


However we see that the Case bar graph has six peaks but the Death Bar Graph only has five.

We expect the Death Bar Graph to lag behind since deaths for new cases will occur days later.

We therefore expect one more peak in the Death Bar Graph and need to accommodate this.

We expect the midpoint now to be in the trough on 26-27 April.

We choose as the midpoint value the peak on 28 April.

For the cumulative graph, there is a value of 59,265 deaths on 28 April.

We double this to get 118,530 deaths and add on 10% to get 130,383 deaths.

This gives us a range for the number of deaths from 105,500 to 130,500.

We anticipate the actual number to be near the top end of this range and adopt 125,500 deaths.

Using the 6% (0.06) death ratio we get 2,173,050 cases for 130,383 deaths.
( 130,383/.06=2,173,050 cases).

We get 2,091,667 cases for 125,500 deaths.

This gives us a range for the number of cases from 1.75 to 2.2 million cases, with the expected number of cases likely to be around 2.1 million cases.

As above, the range for the number of deaths is from 105,500 to 130,500.

We adopt 125,500 deaths.

These are our final totals.

We note that MIT forecasts 126,000 deaths for 13 June.


The midpoint is almost a week later than before.

We therefore expect the end of the outbreak close to two weeks later than previously.

Instead of the end of September, we now expect in the middle of October.

Since 20 October is the date projected by linear regression in a previous post, we adopt this date for the end of the outbreak.

The number of deaths could still be double the minimum value (over 200,000) if

  • There are more peaks and troughs in the number of cases
  • The death rate increases significantly
  • The opening up of the US economy leads to a large number of additional cases


Therefore it may be best to assume that the range from 105,500 to 130,500 deaths is a range for the minimum number of deaths.

My other COVID-19 posts can be found here:

Data for my posts can be found at:

I share my posts at:

2 Comments Add yours

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s