One graph predicts US deaths from covid19, for the next day and for the next 18 days

How many people will die in the US from covid19, tomorrow and for the next 18 days?

Red (deaths) and green (new case count). The dates are offset by 18 days, which is also one major division. Notice how Nov 15, a prediction date, is further right in green (new cases) than red (deaths). Note how the scales of the y-axis are consistent with 1.6% deaths per new case. The red curve catches up to the green curve. The deaths for the next 18 days are to all intents and purposes pre-ordained.

In my previous work, including predicting covid19 deaths for the past 11 days accurately, I have shown the 18 day lag between new case count and death.

In this one relatively unconventional graph you can see red (deaths) and green (new cases) represented in their correlated, caused form. We will see that deaths retroactively follow new cases which have already happened. The deaths are essentially pre-ordained.

The green is annualized new case rate per 100K of population. Similarly the red is annualized new case rate per 100K of population. For reference, the heart attack death rate is also shown, at approximately 200 heart attack deaths per 100K population per year. Also, the 2018 annual death rate in the US (most recent available)is approximately 869 deaths per 100K.

Against that background, here’s the key bit. Each major division of the graph is 18 days, and also the upper x-axis is time-shifted by exactly 18 days relative to the lower axis — see how Nov 15, for example, is near the right edge in the green and one major division closer in the red. This means we can directly compare with the time lag. See how if we superimpose the two curves, time lagged by 18 days, and with the two y-axis scales adjusted for the proportionality factor which has been stable near 1.6% since mid-July, the two curves superimpose. The green has already happened. The red will follow the green. And so we can see clearly where the death count is predicted to go — the red curve simply retroactively follows the breadcrumbs left by the green curve.

Let’s test drive this. On Nov 15 the third covid19 wave of deaths had barely begun, and it was not yet obvious to enough people that this really was the start of a significant ramp in death (red vertical line). But the case count increase was already clear (green vertical line) and the red is essentially pre-ordained to follow the green curve at this time. Having analyzed, I released this paper with the prediction of the next wave. And this is why the data has tracked so closely my prediction, which I most recently updated Nov 26 here. Similarly we can see for the next 18 days from today where this is going. We can expect a lag in some reporting over the Thanksgiving Day weekend, but this would be expected to catch up the following week. There is little doubt that covid19 will be #1 killer within the next few days. The green curve predicts the red curve. The red curve follows the green curve retroactively.

Will Thanksgiving gatherings or the recent SCOTUS ruling removing restrictions on religious gatherings affect deaths? Almost certainly. These are not yet reflected because there are new cases. New cases appear some time after the specific superspreader events. Also note the decline in how fast cases are increasing post Nov 5. One potential speculative explanation is that the superspreader events prior to the Nov 3 election are no longer happening.

It is very clear that covid19 deaths, and associated hospitalizations, are essentially pre-ordained by the case count (green curve) looking 18 days forward. Masks and physical distancing starting today can affect the new cases, and therefore start to have a positive (or negative) effect on deaths, but not until more than 18 days have passed. Yet the sooner the US starts, the sooner the US can bend these curves.

What can we do?

  1. Wear masks and practice physical distancing.
  2. If you work in a hospital, brace for the onslaught and do everything you can to prepare. It’s going to get even worse for more than the next 18 days.
  3. Share widely, please. This is, I hope, clear information and unimpeachable data/analysis.
  4. Contact your local officials who can influence policies. If you feel inclined, contact your local newspapers/reporters/etc. — get the word out.

Recap of my recent work, leading to this one graph

I have been correctly predicting this for the last 11 days, based on what has already happened.

I’ve made quantitative predictions on Nov 15 going out to Dec3 about how many will die of covid19 “tomorrow” which have tracked since said prediction, based on this methodology of causation/correlation, after having quantitatively associated covid19 cases and deaths with existing policy,

And this is why, says the data, that covid19 cases, hospitalizations and deaths will get even worse over the next 18 days, and then even from there if these gatherings go ahead.

I realize that many don’t like this message. The data, though, is quite clear.

Sources of primary data




Larry is a semiconductor physicist by day and a musician (piano/voice/guitar, “Dr L’s Music”) evenings/weekends. He should someday update his LinkedIn profile.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Conspiracy Chrissy presents “My Personal Conspiracy Theory”: A Quarantine Tale

We’re only Now Starting To See This Critical Piece of Coronavirus Vaccine Data

Things are Starting to go Back to Normal for College Students Thanks to the COVID-19 Vaccine.

What is the COVID-19 vaccine and How Safe is it?

Coronavirus: Police get access to NHS Test and Trace self-isolation data

What You Need to Know About The Covid Vaccine

The Light at the End of the Tunnel

COVID-19 collateral damage: Our drinking water

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Larry Tarof

Larry Tarof

Larry is a semiconductor physicist by day and a musician (piano/voice/guitar, “Dr L’s Music”) evenings/weekends. He should someday update his LinkedIn profile.

More from Medium

The Sunday Feed

Have we become too fragile to survive?

A fallen tree blocks my road

DUH! Nuclear War is Bad for Business

The Paradox of Choices