The US collects racial information in crime statistics, right? So it should be one evaluation of Bayes' Theorem to demonstrate actual increased risk based on ra

The US collects racial information in crime statistics, right? So it should be one evaluation of Bayes' Theorem to demonstrate actual increased risk based on race, if any. Seems like it should be pretty trivial in some areas to demonstrate exactly what the probabilities are (and if they are even relevant if background rate of crime is low).