I think you have moved the goalpost from:
"There’s 100 years of science that says masks aren’t effective."
to:
And you appear to be saying that because there isn't 100% reduction of penetration then they can't be effective at all, unless you have changed your story.
Thing is, we also know that there appears to be more than a simple plus/minus of exposure to whether or not someone becomes ill.
There is also dose, length of exposure, individual susceptibility, and the infectivity of the viral particle/variant among others. And these may be moving targets as a virus mutates. Which, again, makes clinical controlled trials more difficult.
If the mask just reduces the exposure some, it may be and probably often is enough to prevent clinical illness. It might allow, as you clearly recognize, the exposed person's immune system to clean up without clinical illness and very possibly blessing them with a degree of immunity.
It needn't be all or nothing. It's not as if one viral particle and you are in the ICU.
At any rate, I will happily grant you that a procedure mask is not going to do much in a high particle environment, which is why we wear PAPRs or N95s in those environments when we know. I am sure many nurses caught Covid because they were given procedure masks and were exposed before they suspected a given patient had covid. And yes, there are breakthroughs in any event.
And, to make things more fun, a noninconsequential percentage of people don't have the right ACE2 receptors for the virus to take hold of, so they are highly resistant and make studies harder, since we don't know for sure who they are, just that they exist.
Anecdotally but with assurance, I can tell you that my partners who caught it (we've all been real careful to wear N95s rather than the company provided procedure masks) did so from their partners or kids, who all came home with it first. We are a small place, and it's a small sample size, but...