Total Cases/Population |
Weekly New Cases/Population |
---|---|

Total Fatalities/Population |
Weekly Fatalities/Population |

Total Case Fatality Rate |

All data obtained from the COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University GitHub repository.

There is no guarantee of the accuracy of the data provided here. While many metrics are similar to what states are using for re-opening guidance, there can be differences in the details of the calculations or the dates in which the metrics are calculated. Please refer to your state's health department data for specific guidance. For example, general COVID-19 data from the Arizona Department of Health Services can be found here, while school opening specific data can be found here.

Data tables provide totals, total over past week, and average of last two weeks. Statistics include cases, fatalities, and fatality rate and can be viewed as raw values or per 100,000 population. For US states, statistics also include tests and positivity rate. All tables can be re-sorted by clicking on the column header.

Plots provide cumulative and previous-week totals. Worldwide plots show all countries with at least 1000 fatalities reported. Highlighted in color for the worldwide plots are the nine countries with highest previous-week death rates, as well as the USA. State-by-state plots show all 50 states plus the District of Columbia. Highlighted in color for the state-by-state plots are the nine states with highest previous-week positivity rate, as well as Arizona. Full-size plots can be viewed by clicking on the thumbnails. Vertical lines represent the first of each month.

Note that the statistics, especially weekly totals and two-week averages, can be artificially high or low based on adjustments in reporting standards, which can result in significant increases or decreases in statistics from one day to the next. The accuracy of the data is limited by the timeliness, accuracy, and consistency of the input data sources, which varies state to state and country to country.

While the above plots show cumulative and weekly cases and fatalities (and for the US, tests and positivity), one may have noticed that the weekly fatality rates are omitted. This is because fatalities in any given week generally correspond to positive cases in previous weeks (that is, the cumulative data available for download do not identify which fatality corresponds to which positive case).

To attempt to model weekly fatality rates, I have developed a simple model to insert a delay function (a log-normal distribution where both median and standard deviation are solved using maximum likelihood) onto the US positive cases data, and then estimate what fraction of those positive cases ultimately died. The quality of the fit can be assessed by comparing the two lines below in the left plot. Caveat: unlike the above plots, the plots below involve modeling. The modeling approach and results are for this website's author's interest only, and are not peer reviewed or otherwise independently verified.

Note that the plot of fatality rate superficially appears to indicate the likelihood of someone who is ill with COVID-19 perishing, but it should be emphasized it is the likelihood of a positive case resulting in a fatality. However, other factors related to the accuracy and completeness of the data are likely in play as well. Examples include:

- Individuals who became ill from COVID but did not test positive (either did not get tested or had a false negative)
- Individuals who tested positive but were not ill from COVID
- Risk factors among the positive cases
- Uncertain/multiple causes of death

US Weekly COVID-19 Fatalities (Model vs. Reported) |
Estimated US Fatality Rate |
---|