You're asking the wrong question. The question is how chase/no-chase policies would trend against a control. These types of social experiments are very hard to

You're asking the wrong question. The question is how chase/no-chase policies would trend against a control. These types of social experiments are very hard to conduct, because there are so many confounding variables.