I remember in the beginning of this, many saw the numbers in China going from 41 one day to 210 and then going up by a 1000 per day and by calculation they projected that by the end of February the numbers would be around 4 millions infected....
That's a different subject. You are talking about contagion here.
The number of extra infection caused by one person or R0, sets the aggressivity of how much the virus is passed from one person to the other. Contagion can be reduced dramatically by quarantine efforts (what seems to have happened in China, but we don't have a lot of transparency).
vs.
The mortality rate (how much of the infected die), can only be truly calculated 6 weeks after a group of person infected on the same day is either cured or dead.
- This is why, when you see 7 people dying and only 20 cases, you should expect a lot more cases in the community (they have been there for while you didn't detect them).
- This is why you could see a location with 50 000 infected and 0 deaths, while another location has 50,000 infected and 2000 death.
Location A is a test after 1 week, Location B is a test after 4 weeks = they are likely to get to the same mortality rate over time.
You have to factor time when looking at the ratio of infected vs. dead.