Well, TBH, the CFR is the best indicator of how deadly. Is this from the “cases” or from the entire population?
That is an assertion. "Dead" is the best indicator "deadly."
Case fatality rate had usually been used -- until this 'pandemic' -- to identify which groups (more than one) within a population are most at risk, comparing one within a population group with another within the same population group.
The lockdown strategy initially proposed by Imperial in the UK (as a proxy for China) ignored age-related risk to assert all were at risk, which was false from the beginning.
It remains false today, as all the "official" data from China with their many "public media" lockdowns. If your choice of CFR is the "best," then by "official" sites China's CFR is most odd.
Here's the comparison, done for you in the old-fashioned "show you work" way which so many today refuse to do:
China's "official worldometer" CFR as of 13 February 2022,
( 4,636 total deaths / 106,930 total cases ) x 100 = 4.33 % CFR.
China's "official worldometer" mortality rate as of 13 February 2022,
( 4,636 total deaths / 1,439,323,776 Chinese ( x 100 = 0.00032 % mortality rate.
"You wrote to ask: "Is this from the 'cases' or from the entire population?" Entire population and entire deaths. And this is the classic definition for the mortality rate.
Which is the "better" rate as a descriptor for "deadly" for the simple and "official" data for China's 4,636 total deaths in a nation of 1.4 billion?
The "deadly" CFR of 4.33 % of China "dead" or the "deadly" 0.00032 mortality rate, aka "death rate?"
The media has harped on CFR because the calculation is the larger. If it bleeds, it leads, so goes the media adage.
Here are both United States rates calculated for you.
For the US as of today -- a comparison of CFR and mortality rates --
( 919,255 "Johns Hopkins" dead / 77,707,694 "Johns Hopkins" confirmed cases (JHU) x 100 = 1.18 % case fatality rate expressed as a percentage.
( 919,255 "Johns Hopkins" dead / 331,002,651 Americans ) x 100 = 0.278 % mortality rate expressed as a percentage.
Which "better" tells of 0.278 % of the population "dead" from Covid in a nation of 331 million?
This is why CFR appears front and center in all the media as regards the 'pandemic.' It appears "deadlier."
But applying 1.18 % case fatality rate as if it were a measure of the population would yield "dead" in the number, 39,058,312.
Now THAT'S deadly. It is also untrue.
It is as untrue as is the "official" less than five thousand Covid dead in China from the twenty-five month long, lockdown-decorated panic.
https://www.worldometers.info/coronavirus/#countries