Read on Twitter
I don't understand how anyone can look at this and think that lockdowns are some kind of magic bullet. Obviously, there are other factors (that's the point!), but people talk about lockdowns as if they were the only thing preventing the epidemic from quickly reaching saturation.
This article is also very helpful to understand that restrictions have a limited impact. It shows the incidence curve for every state in the US and, in each case, it describes what restrictions have been put in place and for how long. https://www.nytimes.com/interactive/2020/us/states-reopen-map-coronavirus.html
If you removed the information about the restrictions and just left the graphs, you would never be able to infer that information from them, not even very roughly. Again, I get that those states differ in many ways beside restrictions, but this should really make you think...
To be clear, I agree that "lockdown" is a vague term, but my point generalizes to restrictions in general. You simply can't explain the graph I show above if restrictions, especially very stringent ones, have the kind of huge effect many people assume. https://twitter.com/AndrewsAmble/status/1353095925883932678
By any account, restrictions in Sweden have been much less stringent since the beginning of the pandemic than basically anywhere else in the EU, yet mortality over there is about average and there has been quite a lot of convergence since the beginning.
Now, you can say that Sweden is different because of this or because of that, and you may even be right sometimes (though mostly you have no idea), but again unless you think there are anti-pandemic magical faeries in Sweden, restrictions can't be that powerful.
My own view is that, once you have implemented relatively limited restrictions (e. g. banning large public events), other restrictions have rapidly diminishing marginal returns in terms of their effect on transmission, but that's not necessarily true of their cost...
If this were not true, even allowing for the fact that stringency of restrictions is not the only factor that affects the epidemic, there would be a clearer signal in the data, but there really isn't and the chart at the top of this thread is a crude illustration of this fact.
I agree that it looks as though Melbourne's lockdown was successful at least in terms of suppressing the epidemic, but 1) it was *very* long and Australia and it would probably have been harder for a large European country to achieve the same results https://twitter.com/samghonfl/status/1353106524294443008
starting from the same point because there is more international traffic and you have harder coordination problems with other countries to solve and 2) this boat has already sailed a long time ago in Europe anyway since incidence is so high that you're not going to achieve suppre
This point needs to be emphasized again: there is path-dependence. Even if European countries could have achieved the same thing as AU and NZ, which again I doubt because there were affected more earlier, it's probably too late, especially with lockdown fatigue, etc.
