Throwing out that word as a weapon doesn’t work anymore, let’s stay on the logical chain of the conversation. Having a proposed explanation for statistical cor

Throwing out that word as a weapon doesn’t work anymore, let’s stay on the logical chain of the conversation. Having a proposed explanation for statistical correlations doesn’t change the amount of danger a driver or passenger is in. So why would one set of statistics be used to allow discrimination but another not?