Perhaps the blocks aren't implemented as naively as you assume. It wouldn't be too difficult, as per your example, to distinguish between an IP that is shared

Perhaps the blocks aren't implemented as naively as you assume. It wouldn't be too difficult, as per your example, to distinguish between an IP that is shared by many devices and one with more limited use. There are also other ways to block besides just IP-based. I do agree with you that the nature of the internet makes blocking people from free (non-paid) sites inherently difficult though.