> Specifically, if large-scale mandates did work then it would be clearly visible in the case curves of each country that introduced them, by definition, because creating clearly visible drops in those curves was the only reason they were introduced. Yet no such drops are visible. You can't look at a country's data and work out when masks were introduced or removed, it's impossible, people have tried. Thus mask mandates failed on their own terms and no further studies or experiments are actually necessary to answer the question of whether they're effective.

I am sorry, but this is a BS argument. First of all, COVID-19 (classic variant) had a 7 to 14 days incubation period. Upon infection rate curves it acts as a low-pass filter (aka smoothing filter). You will not see sharp drops even if contagion abruptly stops. The curve will decay over one or two weeks. Also, mask mandates were not introduced at random moments of time. Typically they were enacted or enforced when infection was on the rise. At best the expectation was that masks will slow down the up-trend.

Obviously, masks do not fully protect against inhaling the virus. But they do limit the blast radius from infected people coughing and sneezing. This is why in Asia people are routinely wearing masks during a regular flu season. Masks do not stop infection but they reduce spread to some degree. Common sense is oftentimes better that shaky science on top of questionable data.