The assumptions as originally conceived hold up just fine. They hold up as well as other checks and balances on democracy: the requirement to be a property owner in order to vote, the assumption that the Senate will “know better” than the House on many matters, the assumption that state legislatures will “know better” than the common man when it comes to selecting Senators, etc. Electors are known quantities who are generally active in and knowledgeable about politics and at least in theory can be subject to much greater scrutiny than a common voter who is subject to none at all.
Not saying that this approach necessarily makes sense, but that seems to have been the thinking. Some of the above restrictions, like property ownership or appointment of Senators, have become more democratized, others, like the EC, have not. But ISTM that the EC was undermined by years of tradition where electors almost always followed the popular vote, setting expectations that they would always continue to do so.
When those few densely populated counties contain most of the population… yeah. Yeah, I do. I certainly don’t think we should weigh miniature counties like Loving County as heavily as we should weigh massive counties like Los Angeles.
Clothy would think the same thing if densely populated urban centers happened to vote Republican, instead of sparsely populated hayseed country being their primary base.
The reality is that while it may make some sense to give extra weighting to sparsely populated states to give them more say in the process, it is fundamentally people that vote, and not acres of farmland and wilderness, so population must be the primary determinant. About one-third of the entire US population is concentrated in the northeast and west coast blue states, and they suffer a disproportionate weakening of their franchise in favor of vast empty tracts of flyover country.