This is a rather careful study of death rates in the US, Italy, and a few other leading countries.

We don’t have accurate data on who has been infected, but as my friend Robert Tercek points out, we have reasonable data on how many people have died of Corona

We specifically know the following at this stage

  1. The first death in the USA was 14 days ago
  2. There are now 57 deaths

From these two facts we deduce that the daily exponential death growth rate is 33%

proof: 57^(1/14)=1.33 ; or 1.33¹⁴ = 57.

Another way to look at this is in terms of the doubling rate, which works out to 2.5 days

proof: 1.33^(2.5)=2.0. I won’t bore you with the math, which involves logarithms, but trial and error is fine too.

So now we have a first model of this.

without changes, the total number of deaths will be 1.33^d, where d is the number of days since Feb 29,

This yields the following

April 1: 1.33³³ = 12,200 deaths

May 1: 1.33⁶³ = 63,000,000 deaths

As you can see, the May 1 number seems absolutely preposterous. So let’s get back to this in a bit.

Let’s now look at Italy

In Italy, the first death occurred on Feb 21, just 8 days before the first death in the US. We are now at 1,441 deaths!

Plugging in the numbers, we get (using the longer 22 day history)

1141^(1/22)= 1.37

Italy’s daily death rate growth pre-shutdown March 10 was 37%!

Amazingly, our number in the US is actually low compared to Italy’s 37% growth factor.

Again, if we had assumed no quarantine etc… this is what we would expect

April 1: 1.37⁵³=17 million dead.

Do I think there will be 17 million dead people in Italy in 15 days? No. Transmission rates should dramatically slow down, as the measures were enacted on March 10.

However, bear in mind there is an up to 2 week incubation period. So even if there was zero contagion post March 10, the deaths would manifest themselves on that cohort somewhere around March 20 on average. We then get zero growth for the days from March 20 to April 1. The total growth days is therefor 40.

So we can assume a minimum amount dead by April 1 in Italy of

“Exp model” Italy Deaths April: 1.37⁴⁰=294,000 people

Now Italy’s population is only 60 million. And if we believe the 1% death rate estimates (as measured by south korea and others and estimated by WHO). this would correspond to 1/2 the country being infected as of 15 days from now.

Back to the US and those millions of deaths.

There are three caveats to these models

  • behavior changes
  • seasonality
  • mutation

Lockdown / Behavior Change

It’s possible that the US suddenly wakes up in the next two weeks and slams on an Italy / France / Spain lockdown. I hope so, and as a betting man I probably would think its in the cards. If that is so, we can probably seriously cut the growth factor to zero (decreasing new deaths). China and South Korea have proven this.

Seasonality

Let’s pray that like the Flu or even 1918 Spanish Flu, this thing does not like warm weather. This could buy us a lot of time

Mutation

These things Mutate. The Spanish Flu ended by a quick mutation. This probably will as well. But it could mutate into something even more deadly. We just don’t know.

Other Countries Death Rates

A look at the other highly growing countries shows growth rates in the 17%–60% range. Again, our own rate of 33% is very “average”

France. Started Feb 15. Now at 91 deaths. 91^(1/28) -1 = 17%
Spain. Started Mar 3. Now at 196 deaths. 196^(1/11)-1 = 61%

Iran. Started Feb 19. Now at 611 deaths. 611^(1/24)-1 = 30%
Germany. Started Mar 9. Now at 9 deaths. 9^(1/5)-1= 55%

Conclusion

Barring massive testing we simply don’t know the mortality rate per person infected. But we do know that the Virus tends to kill at a 17% to 60% daily growth rate, left unchecked.