Updated 2020-05-28 07:00 UTC - Situation is improving. Lock downs may have been too much for the price. There is enough info out there, so I will shutdown this site on 2020-06-01.
Real-time prediction counter, an interactive chart of 4 day forecast, comparison of countries, probability of your infection, latest news, quiz, and discussion about the novel coronavirus. Updated daily. Predictions are blue points.
Above is real-time intraday prediction based on the model below.
The prediction of the outbreak progression is based on a very simple assumption that the counts will at first follow an exponential curve, at least at the beginning of the outbreak. Over time, instead, the s-curve shape will become dominant. The log scale chart of the time-series shows that growth of infected count currently is loading ... per day (geometric progression). Since the parameters of the geometric progression are changing, the prediction is calculated using only the latest 5 data points.
The time-series graph above tracks 2019–20 Wuhan coronavirus COVID19 (2019-nCoV, SARS-CoV-2) outbreak. The basic reproduction number is around 2.6 new infected per one sick. Common symptoms include fever, cough, and shortness of breath. Time from exposure to developing symptoms (incubation) of the virus is 2 - 14 days.
🔍 Compare countries
Add your countries of interest via dropdown and button below. I recommend to use log-scale to contrast and compare growth acceleration of different countries. The data is thanks to Hopkins.
Fill in your local numbers and click calculate. This is just a simple estimate and there may be a bug in the calculation. But it is the best I have as of 2020-03-12. It is calculated by estimating probability of not meeting any of the infected, but unconfirmed (not quarantined) people. The infected people are estimated based on claim that 80% of cases have mild symptoms and are not reported.
Total infected not quarantined:
Daily probability meeting one or more infected person:
If 1/n people are infected with a disease, a gathering of .7n people is likely to include at least one infected person. E.g. if 1% of people are infected, n = 100, so any gathering of 70 or more people probably includes an infected person.P. Graham
Currently all data comes from Hopkins. On days prior 2020-04-05 the data was sourced from Wiki as described below.
The data is updated daily from Wuhan coronavirus outbreak wiki page.
The very early data till 2020-02-25 are sourced from various web pages since I found no good single source with both infected and dead historical data. The historical infected were collected from a wiki based on the expectation that before 2020-01-28, almost all infected and dead were in China.
The data point for 2020-01-29 comes from the global estimates on the wiki linked refered to as "As of 30 January 2020". The historical data for the death toll comes from an page I cannot find now. And are then combined with the data point from NY Times 2020-01-27 and the wiki one for 2020-01-27.
📬 Report issues to: corowrrw [at] server135.web-hosting.com