No, it's actually correct. You've got the logic here backwards, so please think about this a bit. How much impact can we see at the time of introduction and re

No, it's actually correct. You've got the logic here backwards, so please think about this a bit. How much impact can we see at the time of introduction and removal? Zero. In all those graphs the curve continues exactly on its previous trend line. Your point would be valid if there was a visible impact and we were trying to establish how much of that came from masks. But we aren't trying to disentangle cause and effect here, because there's no effect to establish a cause for. If there's no effect then masks cannot be the cause, can they?