It turns out that economic cost of labor and inputs also factors into energy considerations. It's rather soft thinking to ignore them, and leads to very bad conclusions.
I see this from nuclear advocates when they talk about "energy density". Energy density (or EROI) are only relevant insofar as they affect cost. Why not just use cost? (The answer is that cost isn't giving the nuclear advocates the answer they want, so they have to lie and use some substitute metric.)
EROI cannot be bad, since energy is only a small part of the cost of manufacturing. If EROI was unacceptable, then overall cost would be too.