what makes you think that an additional 66% would do anything but cause them to use even more resources on Bitcoin?

Well let's speculate that we're heading into a high fee future eventually again. In the Ordinals wave we saw the BRC-20 sub-wave, where millions of transactions were created for the hope of some meme token getting some value. If transaction fees are at $5 for a token mint or transfer, then you might be more prone to create such tokens massively, than if the minimum threshold is 15$ because the buyer of the token will have to pay that fee too, so the needed profit is actually $30+. I can definitely see scenarios (based on past experience with Ordinals) where that can make the difference between a NFT "spam" wave really taking off or not.
I've seen the debate with Luke-Jr and his friends (and not so good friends) now. I understand Luke's point a bit more now I think. But I don't agree with him.
In the end, I think the debate is between a "theoretic" position - Luke's - and a "practical" position - Core's. Luke's theory that spam encoded in financial data can be still seen as financial data is not incorrect. But practically, it doesn't make a difference if there are websites and easy to use tools that happily decode all the "fake financial data" for you and transform it into images, videos and audios.
The Core position instead focuses on the practical consequences of "spam" waves, like the UTXO set pollution and the compact block issues, and tries to find a way to make node operation cheaper for nodes and thus more decentralized.
So while I can respect Luke's theoretical position, at the end it is a good example for the well known phrase "the road to hell is paved with good intentions". His ideas would lead to higher node operation cost and more development cost.
All of that assuming that the NFT/token market stays the same, which is the most reasonable thing to do. And trying to predict market changes because of an "official approval" of large OP_RETURN data chunks is already taking the debate to another, much more speculative level -- we could also talk them about "the attraction of prohibited things", e.g. NFT guys often describing themselves as "degens". Would "degens" be happy to be "officially" allowed to do something? That's equally speculative.
Thus I stay with my stance: the Core 30 change is reasonable and has probably the least harmful practical consequences even if new spam waves occur in the future. Stampchain-based spam waves would be worse (and not even BIP 444 could
stop them).