They should do value at risk and factor based %s to evaluate how much risk the models are taking on unintentionally

A good outcome doesn't mean the decision making process isn't bad