It assumes that when _you_ make up a scale, the people that deserve to be at the top of it tend to be there. There's no reason for it to imply there's one universal scale of good people versus bad people.
For instance, if you want to understand why some people make more money than others (a trait that _is_ well-ordered, and can be put into a list), then somebody who believes in meritocracy would say that the people who make the most money earned it. If you want to understand why different people hold different positions in a company, then the meritocrist would say that between all the people who wanted each position, the person who was best for it tends to be the one that gets it.
Each of those scales may be more or less sympathetic to an explanation by way of meritocracy. But anyone who wants to make that argument doesn't have to construct an a priori global list of goodness for all people to give you an answer.
That's the thing nobody deserves anything. How do you prove one scale is better than another? GDP - arbitrary. GINI index - arbitrary. Violent submission? I guess at the end of the day there's your answer.
When people say things like that, what I hear is that "the scale on which things are ranked should be opaque and only understood by the elect, i.e. the people I prefer", not that there is an alternative to ranking things to choose between tradeoffs.
There's different perspectives and trade offs.