First off, this is a news article - not a peer-reviewed study. You can't reasonably expect complete detail to be included - have a look at the linked papers. S

First off, this is a news article - not a peer-reviewed study. You can't reasonably expect complete detail to be included - have a look at the linked papers. Second, there's substantially more detail in here than you have stated. For example, the correlation between the decrease in violent crime and the use of lead is reproduced in areas with differing crime policy, and shows a similar kind of decline in areas which stopped using lead earlier or later than others. Third, "it just kinda-sort-of fits" is all you will ever get in anything outside of pure mathematics. The real world is pretty noisy, and exact correlation is never going to happen.