I agree, with some reservations. For monsters your concept is true, but the minion effectively reduces one side of the equation to an irrelevant state, greatly changing the durability probability. For example a minion has at minimum of a 5% chance of outright death per attack, which is obviously absent from other creatures.
Yes, it changes the "curve" describing his death chance, but if you were really just interested in "average death probability by round", you could create a fixed point.
I agree. I was referring to explaining minions as the poster I quoted explained them. Minion's cannot be said to be the same creature "with which you struggled at lower level (you needed 4-5 hits) is very easy to kill at 9th level (1 hit)." Such an explanation places monsters in a state of uncertainty. That said, this is the most common role-playing reason that I have heard used to support the minion concept.
I see it as a "modelling" thing. We use models in a lot of contexts in sciences. Sometimes we use different models based on the scale of what we look into. In Solid State Physics, a lot of succesful models accurately predict the observations without relying on quantum states of atoms or molecules. Trying to use those would be theoretically more accurate, but it is most of the time beyond our available processing tower. The "higher scale" model is far faster and not noticeable less precise.
Some scientists even say that this is what we should do. I picked the example of Solid State Physics because I read an article about it where that is exactly what the authors wrote. Their model works great, doesn't use quantum mechanics, so don't bother with them until you have to.
And some consider it even a possibility that this is how the "science" in the end will work out - there are different models for different scales or contexts. For certain situations, a model fundamentally (not just because we lack brain power) doesn't work and we need a different one. The first time I read about this was when reading on the Grand Unifying Theory, where people noticed that there is still no testable theory that unifies Relativity Theory and Quantum Mechanics - it's possible we need to keep those seperated, describe large scale events with RT, small scale events with QM, and maybe we need a third one that only works where the scales "meet" (The Big Bang, Black Holes). Of course, that's still up to debate. It doesn't change that models can be useful depending on context or scale.
In RPGs, our own brain is usually the processing power. So we should look to use models that work reasonably fast with our brain without losing too much precision.
One scale in D&D is level. Yes, you can pit level 20 heroes against CR 1 or Level 1 or HD 1 monsters. But you can also try to use a different model for them that's easier to use. Minions are such a model.
The game rules are not the physics of the game world. They are an abstract model of it.
Now people could say that switching to the Minion model is too imprecise for their taste. But then, they don't have to use the model. If they think that using level 1/CR1/1HD monsters against 20th level PCs gives them better results, they can do so. They can use Minions to model something different (for example, unique creatures that are inherently "weak" on some level - Decrepit Skeletons or Lich Vestiges, but never Level 10 Kobold Minions). Or just not at all.
