It does not stop the exponential spread - that ship has sailed. At this point, it just slows it down by reducing the multiplier - "flatten the curve".

Worse yet is that if we let the measures off, we just get the same curve later. A bit lower at peak, but still exceeding healthcare capacity several times (say, 6x rather than 8x, if we do a 5-month lockdown).

But this still saves lives - just not anywhere as many as we could have saved if we contained it. It also gives us a chance to build more hospitals, manufacture more ventilators etc - although it still won't catch up with demand.

The best endgame is that we either get a vaccine, or enough tests to test everybody, and then do targeted quarantine from there.

The "Best endgame" you describe is exactly the goal. We lockdown now so that we can buy time to get tests and tracing systems setup. Then we can start to open up everything again and control hotspots with aggressive testing and individual quarantine (as opposed to the blanket quarantine we are doing now). With good testing/tracing coverage, we can stop the virus from spreading exponentially for more than a few hops.

Also locking down everything DOES stop the exponential spread. That's the whole point. If you don't see other people, you aren't spreading the disease, reducing the rate of new infections. If each infected person infects less than 1 other person on average, we'll have stopped the spread. China and SK have already succeeded in this.

* If each infected person infects less than 1 other person on average, we'll have stopped the spread.*

This isn't exactly correct. It means that, eventually, the spread would die out even if there were an infinite number of potential hosts. But it's still spreading. If that number is exactly 1, ~everyone will still end up getting infected. If it's just a bit below 1, it will behave as a decaying exponential but will still take a long time to die out.

My point was that a 1-month or even 3-month lockdown doesn't stop exponential growth permanently, it just pauses it. A permanent lockdown would stop it, but it's obviously not sustainable.

And I wasn't trying to dispute your claim that the "best endgame" is the goal, merely clarify it - and also point out that even then, we will still see our healthcare overwhelmed, and a lot of people will still die because they will be rationed out of the system. People who wouldn't have died if we locked things down when containment was still realistic - so I consider them all preventable deaths that we have failed to prevent as a society.

To be absolutely clear: a lockdown is still our best bet, and the lives it'll save are worth it.

There are two sides to an exponential. If R0 is >1, it just keeps going up. However, if R0 is <1, it just keeps going down. The point of the lockdown is to get it <1 so it keeps going down for a while, until the case counts are low enough that we can adopt a different strategy.

Including cycling lockdown on and off without me very reaching a caseload that causes collapse of the healthcare system until we get a vaccine, etc., to change the basic equation. But first we've got to get R<1.

Slowing down infection rate whether exponential or not is good because it lets more time for medical system to prepare and because it spreads cases over long period of time.