First, there is active research and we demonstrably have learned something. See
https://aricjournal.biomedcentral.com/articles/10.1186/s1375...,
https://www.nature.com/articles/s41598-020-72798-7, and
https://www.pnas.org/content/118/4/e2014564118 for several examples.
Second, your simplistic analysis demonstrated that you, personally, are ignorant of the real tradeoffs involved in whether masks work.
Wearing a mask reduces how much virus leaves your mouth. But when you breathe out, most of the virus is in larger droplets that quickly hit the ground. However breathing out through a mask creates perfect conditions to create an aerosol, which can allow more of the virus to stay in the air for an indefinite period of time. So there is a tradeoff, and there were reasons to question whether cloth masks were better than simple social distancing.
It turns out that what matters most is not that you get exposed, but rather the initial viral load that you get. You see, the virus will go on an exponential growth until the relatively fixed time it takes the immune system to figure things out and start shutting it down. If the virus gets a solid head start, the odds of serious illness go up. Therefore the lingering aerosol from a mask is (except if it accumulates in poorly ventilated indoor spaces) of less concern than an unmasked person talking directly to you.
So the result is that masks work. Even crappy cloth masks work.