You don't really need to understand genetics very well to do eugenics see animal husbandry and except for certain dog breeds the unintended consequences of have proven to pretty manageable in animals (except in some breeds where targeting a certain appearance rather than temperament, physical ability or intelligence is the primary goal).
The difference is that with animal breeding you have a clear distinction between the people doing the breeding and the animals being bred. Humans breeding humans any attempt at being "scientific" is likely to collapse in the face of the resulting signaling games.
I mean look at the current state of social science, do you think it would somehow magically improve if eugenics were to be embraced? In fact, it's likely to get worse since the results of social science would then have even more significance.
I recently read an article by Steve Sailer that reminded me about something I have been puzzled by for a long time. Relevant paragraphs:
Poor people having fewer children means that the children have more resources available per capita making the children better off. Rich people having more children actually increases equality in society since it reduces the per capita resource advantage their children have. Rich people giving to their children is also one of the few cases where the redistribution of wealth doesn't reduce incentives for wealth creation. Rich people care about their children too.
Since programs aimed at reducing teen pregnancy rates do seem to have had some effect, we known something like this is possible without being horrible to the potential parents it targets.
Yet a policy of "poor people should have fewer children, rich people more" sounds heartless despite increasing general welfare both by making poor children better off and by reducing the privilege of rich children thus increasing equality which we seem to think is ceteris paribus a good thing.
Why is that?
Edit: To test the source of the reader's intuiton (assuming he shares it with me), I encourage the consideration of two interesting scenarios that may depart from reality.