True ...
But there is a caveat. I'm a clinical nurse specialist who has been working in a Covid-19 ICU ward for about the past six months. From a purely statistical point of view, you are correct. Only a tiny fraction of people infected with Covid-19 die from it, and yes, the majority of those who die are elderly and had underlying health conditions. Not exclusively, but generally speaking.
What we know is that about 80 to 90% of people infected by Covid-19 will either have no or minor symptoms and will recover within a few days. About 10 to 20% will require some hospital treatment, and a small fraction of those will require an ICU bed.
My job is to keep my patient alive. I don't care what age or underlying condition they have, I will do everything I can to give them the best chance of recovery. For a single ICU bed you will have about six nurses on rotating shifts to provide 24 hour care. In the ward you will have doctors and specialists also on rotating shifts. It requires a lot of people to staff an ICU bed, let alone a ward.
If the public, through their actions (or inactions) don't get the viral spread under control, that means more people presenting to hospitals, and more people in ICU wards. There is only a finite number of beds and staff. We don't kick people out of beds.
If we end up with a situation where Covid-19 patients are occupying most of the ICU beds and capacity is stretched then that has consequences ... if you have a car accident and need critical care we can't magically create new beds and staff to support them.
The point is the statistical numbers are being used to suggest Covid-19 isn't a problem. It is. If too many people contract it too quickly, the system doesn't have the capacity.
Every person in ICU as a result of Covid-19, regardless of their age or health condition, represents an ICU bed not available for you, your family, or your friends should you or they need it. That is what people need to remember before suggesting Covid-19 isn't serious.
Additional: Just to respond to GraphicConception ... the argument he makes is common, but completely wrong. One way of calculating excess mortality is to take the mortality figures from previous years, average them, then compare the current year to that. This has to be done on a month by month basis, and has to be performed for individual US states since reporting methods are different. Another way of doing it is to model for known variations ... a particularly bad flu season one year due to a specific strain can be accounted for in the baseline. The CDC does not agree with GC's conclusion. Studies in JAMA and the BMJ disagree. Studies in individual nations do not agree. The upshot is the numbers of deaths attributed to Covid-19 is not only real, is not only above the norm for previous years, but is probably an underestimation. The other point I would make, since I have seen this virus in action, is that in about 90% of cases, there is no doubt medically that Covid-19 was the reason these people died prematurely. That might not suit the narrative some want to believe, but we aren't just making stuff up on forms because we do actually have sufficient skill, diagnostic data, and a patient's prior history and medical charts to be able to make that determination on an individual basis.