Coronavirus Disease 2019 (COVID-19) Pandemic

Your video is quite interesting but you can get a more realistic approach by combining it with this one:

The main idea is that there is a delay between the time when someone is infected - and becomes able to spread the virus (in the video they insist on the time when symptoms appear but new contaminations are possible before symptoms appear) - and the time when the case is confirmed. So you're underestimating the number of cases during the phase of exponential growth and you also don't immediately see the effects of confinement because people contamined before are still detected and also because you test more people and detect cases that you wouldn't have seen before (people that are asymptomatic or have very lean symptoms). This is what happened in Italy a few days ago when the number of new confirmed cases started to grow again (but the number of tests raised at the same time). About the end of the video, I use other statistics that are more accurate: the number of people who are hospitalised and the number of people in intensive care units. More exactly, if datas are available, I evaluate the number of people that were hospitalised a given day (or week) thanks to the number of hospitalised people and the number of people who left the hospital (alive or dead). The reason is simple: severe forms generally appear between six and eight days after the first symptom so when someone arrives at the hospital you can assume that he was infected 12 days before. So if you use these data instead of the number of deaths, your delay is 12 days, not 20.

About the mortality rate, it depends on two factors: the ability of the healthcare system to avoid deaths as much of possible - so a place must be found for every patient - and the composition of the population: in a country like Italy, elderly people represent a bigger part of the population than in Iran for example. The best way to estimate the mortality rate in a given country would be to use the example of Diamond Princess, because we know exactly how many people were infected there. 10 people died out of 712 cases so the mortality rate was 1.4%. But that mortality rate is only true for the population of Diamond Princess (with a higher rate of elderly people than in the rest of the population). Making a proper estimate would require detailed data but we can make a rough estimate by reasoning as if only people over 65 years old died and if everybody on the Diamond Princess was over 65. So the mortality rate for a given country would be the product of the mortality rate of Diamond Princess by the proportion of people over 65 years. So for example for Japan (28% of people above 65) we would obtain 0.392%. The rate would be 0.317% for Italy, 0.301% for Germany, 0.278% for France or European Union in general,  0.256% for UK, 0.224% for USA, 0.157% for China or 0.078% for South Africa for example. I insist on the fact the this rate includes all asymptomatic cases, and in the video it's more about the cases with symptoms. And also this rate grows if the healthcare system is overwhelmed.