[1/n] Well @gummibear737 pushed me into a position to have to explain our theory of how universal mitigations (for ex. lockdowns) kill over time more people than what Sweden did.
[2/n] Disclaimer ahead, this is early work and not even the central point of our paper with @LDjaparidze. So don't read too much into the details yet.
[3/n] The gist of it is: Like many things in physics and biology, the death toll is path-dependent. One way or another, you always eventually get to community immunity and accrue unavoidable deaths. The question is always HOW you got there.
[4/n] Either by infection, vaccination (at least for some time), or death you will remove a certain percentage of the population through immunity. Meanwhile, over time, you will unavoidably accrue deaths (that's a sad fact).
[5/n] The central argument is that if you have a 2 class system (like with COVID): Vulnerable, and Healthy. When you reach your destination (endemicity) one class will accrue on expectation N deaths over a number of infections (usually 100K),
[6/n] while the other will accrue N'. Here the expected deaths for Vulnerable are higher than for the Healthy, any path that risks Vulnerable more than the Healthy will have higher total deaths over time (and the longer the path to herd immunity, the worse the total fatalities).
[7/n] Clearly the optimal strategy is always for the Healthy to get infected first, their IFR can be more than 2 orders of magnitude lower; and in this fashion a country achieves community immunity as fast as humanly and ethically possible, minimizing total fatalities.
[8/n] Lockdowns are universal mitigations. It means: If you were the virus, whenever you are forced to choose your victim (because we couldn't achieve extinction), you are going to pick whoever happens to make an isolation mistake.
[9/n] In probability we call that a random walk. You assign the same probability to choose Vulnerable and Healthy. The probability of choosing a Vulnerable is p=0.5. 
[10/n] On the other hand, targeted mitigations like being careful around Vulnerable, makes it very difficult for you to infect the Vulnerable and far easier to find Healthy victims which in turn will contribute to community immunity with minimal risks. p going to 0
[11/n] In practice, p is not exactly p=0.5 or p=0 because people have to work, demographics and policies influence that number; so the immediate question arises: "As a country, Is it possible to estimate your p number?". And that is what our paper is about.
[12/n] To summarize: If we are not able to extinguish fast, over the lifetime of the epidemic, universal mitigations force the probability of getting infected for Vulnerable and Healthy to equalize. Therefore, we will have a much higher death toll.
