you can't completely predict who it's going to be mild for. 14-year olds with no known underlying issues have died, 25-year olds have lost their sense of smell and taste permanently or suffer from long covid months later.
Covid has killed 42 people from age 5-14 in the US according to the CDC, for example. That alone is really not something we should be making huge policy decisions around. It's sad, of course, but it's extremely isolated. All it does is distract us from making good policy decisions.
If I was making policy decisions I'd order vaccine rollout as follows:
Essential workers => 80+ years old => 70+ => 60+ => etc. as we produce and distribute the vaccines.
At some point we might look at the data and say "that's good enough, we can resume normal activities while we work to vaccinate the rest of the population".
IIRC the IFR is ~0.2% for people in the 20-39 age bracket. I still find that risky (though I'm risk-averse), but the better argument is the lingering effects of moderate/severe cases, which I'm not sure is counted.
Also, I assume your policy decision would include younger people with risk factors like heart disease and diabetes, rather than going strictly by age. I'm chronically ill and I want my vaccine.
My main point is that there are sub-populations that have a small enough risk that we shouldn't stress over them too much. And talking about them is a distraction.
I'm not sure if this is a concession for argument's sake, but imo it does take away from it. Why do you think it doesn't?
There are plenty of valid arguments to vaccinate everyone we can.
EDIT: After re-reading I understood the sentence, my bad.