It is not conjecture since my theories are based on the factual observations of how the Bitcoin network functions today. To simplify the argument for you however it is based on the fact that the vast majority of miners do not run full nodes for the purpose of mining. The pools run the full nodes for mining instead, that is why miners are not effected by the increased difficulty of running a full node, because miners do not run full nodes for the purpose of mining. This is why increasing the blocksize does not lead to increased mining centralization.
The pool operators have no choice but to include their running costs as a part of their fees. That means that the full costs of expanding network resource requirements are still referred to pool users, just indirectly. You're clearly not competent to assess this sort of thing, despite inviting people to read countless paragraphs of your "factual observations".
In the case of the pools, the running costs would be no more then say 100 full nodes for all of the pools combined, this number would include backup nodes as well. I am sure you can come to understand that running such a small number of full nodes is completely negligible in terms of the running costs of mining operations globally.