Great study from Harvard in 2017 to reference when debating "police brutality/racism against blacks".

The abstract (page 1) states that "on non-lethal uses of force, blacks and Hispanics are more than fifty percent more likely to experience some form of force in interactions with police." This will be what some people focus on. However, there are huge caveats later in the study:

Page 19 - "Taken together, this evidence demonstrates how difficult it is to understand whether there is potential selection into police datasets. Estimates range from blacks being 323 percent more likely to be stopped to 45.4 percent less likely to be stopped. Solving this is outside the scope of this paper, but the data suggests the following rough rule of thumb if one assumes that police are non strategic in stopping behavior there is bias. Conversely, if one assumes that police are stopping individuals they are worried will engage in violent crimes, the evidence for bias is exceedingly small." Last sentence is key regarding policing strategy.

Page 24-25 - "Racial differences in police of force does not seem to vary with civilian gender or officer race especially for black civilians." No difference in use of force based on race of the officer.