> 11% (just adopting the number from your comment, no idea if that is the real figure) is huge. It would turn a spread 1.05 into one of 0.94 assuming that number has a direct impact like that.
No, because that's not what the 11% number was measuring. You can't just take an 11% relative reduction in population seroprevalence at a given time, and then blindly multiply the virus R0 by that factor. They observed a drop in symptomatic seroprevalence that corresponded to about 1 case for for every 1300 people who wore a mask. That is nowhere near an 11% drop in R0 (which you'll recall is an exponent).
> That's exactly what is wrong with all these discussions, the cost of wearing a mask is negligible compared the potential cost of not wearing a mask:
How do you know? Have you measured it? No, of course you haven't. You're just asserting things.
> so wear a mask. It doesn't matter if it is proven to help yet or not, the fact that it may help and won't harm is the only thing that matters when deciding whether or not you should wear one.
Fine. You should take Ivermectin, then. And vitamin D. And flush your sinuses daily with a neti pot. And get your chakra aligned. It's essentially the same argument. Can't hurt, might help! By your logic, it all must be done.
...or do you just pick-and-choose which unsupported interventions you prefer to believe in?
(for the record: I do not believe you should take Ivermectin. But it is the same argument.)
> the evidence quality for Ivermectin is substantially lower than that,
It really isn't. Just like mask studies, there are a few weakly suggestive good ones, and a few bad ones that make ridiculous claims, and a whole lot of nonsense in between.
> there are substantial downsides to using it in quantity.
No, not really. It's a pretty common, safe drug. We probably know more about the downsides of Ivermectin than we do about the downsides of (say) masking young children in school.