Clean Data for Data Recovery

admin 0

March 21, 2020, I started tracking COVID – essentially daily numbers. I do this job for several reasons: I enjoy data, I don’t go anywhere for a while, but more importantly so I can discuss it intelligently.

Public trends have emerged in the past seven months. “Waves,” as some call them, appear in the charts.

In this white paper I would like to provide the charts developed to track my night. No decisions were made, no comment was given, and no opinion was given. This is pure data for data sake. You must draw your own conclusions, provide your own definition and develop your own ideas.


As mentioned, the information used to develop these charts has been tracked and recorded NIGHT FROM March 21. All information used as part of this tracking program is taken from, Johns Hopkins and many state websites. Although each of these sources of information agrees to provide accurate data, neither they nor I can guarantee the information.

Because one day does not make a trend, these charts use fourteen day rotations. Using an average of 14 days removes the highest bones and valleys in daily statistics and even the bones found on average rolling 7 days. When the average rotation of the previous 14 days is NOT used, the chart description shows that another method is used.

The time period tracked in these charts is slightly more than seven months between March 21 and October 31.

One descriptive information I provide is a description of what information is taken by the chart.

Shall we begin? I hope you find this information useful.

New Case Cases

This makes daily case reading charts, again based on a 14-day rotation scale.

Daily Death

These charts die daily using a 14-day rotation scale.

Daily Death Rate

This raises daily mortality rates. The death rate presented in this chart is calculated by dividing the number of deaths in a day by the number of new events for that day. Although it is not an exact indication of this mortality rate, because this model uses a 14-day rolling average of cases and death, it is a typical definition of the actual mortality rate.

The recovery

Followed by a 14-day recovery period.

New charges of Compare to Recovery

Two databases are compared in the following chart: 1) new daily cases; and 2) daily cures. Daily new cases are represented by blue lines and orange lines. Again, these use 14-day rotation scales.

Combined Comparison

Three key databases are compared in the following chart: 1) Total reported events; 2) Total recovery; and 3) Total working. BEFORE charts, this chart DOES NOT use a 14-day rotation scale – this makes the charts the actual daily numbers.

The blue line is the total reported events; gray tracks active cases; and lunch signifies cures.

US percentage of total deaths

According to the WorldOMeter, COVID has affected 216 countries and territories – internationally in fact. US accounts for approximately 4.29% of the world’s population. This model tracks what percentage of COVID deaths worldwide occurred in the US.

Like the previous comparison chart, this chart is NOT based on the 14-day rotation, which tracks the daily percentage from March 20 to October 31.


The following are three charts that provide monthly information instead of the average 14 day rotation or even daily chart of numbers. The following charts compare new cases, new deaths, and new recovery for the months of April to October.

Monthly New Issues

Monthly Deaths

Monthly New Recovery

Overall Comparison

This chart is essentially a tracking of all COVID numbers from March 21 to October 31. This chart compares all the numbers to the US population. The color code is:

  • Blue Blue: US resident
  • Red / Orange: Total Number of Charges Reported
  • Yellow: Total recovery
  • Gray: Working issues
  • Blue Light: Total Deaths

The most important insurance news, in your inbox every business day.

Get a reliable insurance company newsletter

Leave a Reply