The studies trying to find the link between genetics and IQ are flawed in that they don't have a way to handle the interactions between genes. Instead they are

The studies trying to find the link between genetics and IQ are flawed in that they don't have a way to handle the interactions between genes. Instead they are based on correlations. If you will be good at baseball if you have A and B, OK with neither, but terrible with only one, those genes will be identified in most studies as having nothing to do with baseball, maybe even identified as harmful. I can offer a concrete example. Autism is more often found in high-IQ families, but itself is associated with decreased IQ. That suggests one case where either A and B alone are harmful, having neither makes you fine (and unlikely to bear children with A or B), and having both makes you smart. I am not sure if we want people to figure out how to make connections involving specific genes, because a lot of us benefit from living in a world where we can game LeetCode by working longer and harder in a way that's unrelated to what Google is looking for.