to be clear: you could take a large, random sample of people that have been vaccinated and watch them get infected by a variant and then rather easily beat it and conclude the vaccine must have helped, at least a bit.

but, there's no reason to think they wouldn't have beaten it easily, anyways - and hence no reason to think the vaccine did anything at all, if you're seeing routine contraction in vaccinated people, which we apparently are.

rather, what we're doing is assuming the vaccine worked at least a little bit, and i'll even hold to that - that's a reasonable assumption.

but, we're going to need some major studies before this line that it protects against "severe illness" can be reasonably evaluated. and, in the end, i suspect it will turn out lacking.

but, i want to repeat the point - if you're high risk, you should get vaccinated, anyways, because even a little bit of protection might make the difference, in the end. it might not. but, you want to take that guess - that's a bet you want to make, even if it turns out to be specious, in the end.