"if other sentient species exist, we should expect ours to have an unusually high population"
Except that sentience evolving is, as far as we know, extremely tied to population numbers and then afterwards sentient agents will have strong incentives to manage their population in some way (or you could say: the odds of them deciding to exactly match their pre-sentient population levels, now that they can make a choice, is small).
The underlying explanation of the principle of mediocrity and a mathematical intuition for it is pretty good.
Maybe people in 2021 are just extremely skeptical about "if we assume the most likely thing every step and pretend each choice is independent we can build a prediction that we can assume is maximally accurate, which means it is meaningfully accurate" chains of reasoning? That's how you get "Roko's Basilisk scenarios scaring people" type outcomes. It certainly is the source of my distaste and bias against it if I'm being honest with myself.