I’m typing this before heading off, so if anything seems amiss or unclear, I’ll clarify later:
I went to check to make sure I wasn’t making stuff up, and Sizzle’s on record (multiple times) as saying as much. So yes, NP1 is assumed. I understanding the simple, straightforward reasoning: Not everyone is a whale, the rates suck. Broadest applicability overall.
Personally, I read the list as meant as discretionary rather than rather than prescriptive, and saying this is how things are as simple as that (especially since where SSRs go, they outright call every SSR worth leveling). But that doesn’t necessarily mean they automatically change in ranking, where your second part goes, even though they can. A highly-grailed FSN Cu, and I believe Heracles as well, for example, can both more efficiently solo nodes ungrailed versions could not, if not just have a broader range outright. Never mind Cu (Alter). On a different angle, various Servants like Eresh, Shiki (Saber), Da Vinci, and so on all especially appreciate the NP2 given their otherwise soft NP damage. And if you really want to take someone far as they can go via Fous and Grails, you can push the limit and argue for such-and-such unit ‘deserving’ of such a position.
So that’s what I mean by discretionary, less so “Only raise this Servant” or “These Servants trounce these losers” and more, as mentioned… a comparison tool. All-in-all and assuming you’ve got all the options there, you’re likely going to get the most efficient and practical usage across the spectrum out of raising Merlin before Enkidu, Drake before Shuten, Gil/Tesla/Ishtar before Arjuna (cries), so on; it’s relative and fluid, not all set in stone, because god knows not everyone plays the same or rolls the same.