>where you got that 30% from

According to https://coronavirus.app/ we have 34955 infected, 726 dead, 2287 recovered as of now. Total resolved cases are dead + recovered, which is 3013. Fatality rate is 726 / 3013 = 24.1%. At the time of my original post it was 30%.

Just checked the last Situation Report from Feb. 7th from the WHO:

31,211 confirmed cases and 637 deaths in China. Makes a rate of roughly 2.04%. And I trust the WHO to get the calculation behind that right.

That's exactly what I don't understand. Logically, wouldn't deaths typically lag diagnosis by some amount? If so, we would expect the death rate to be higher than current deaths/cases because we need to use the confirmed cases from some number of days ago. That number of days being the average time between someone being diagnosed and dying.

Another stat is that there are roughly 700 deaths and 2400 recoveries, which would indicate a mortality rate of 23%. Of course, this suffers from the same delay issue as the 2% calculation (T_death < T_recovery). But the possible range is 2% < M_rate < 23%

https://www.worldometers.info/coronavirus/#repro

Your intuition is right but it's complicated.

This paper gives a discussion of different biases:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4504518/

There's a standard paper on this topic the authors I forget atm but all estimates will be biased. The best estimates are sort of unintuitive and account for estimated time-to-death and the time since diagnosis.