A study has come out analyzing police shootings in America. The prevalence of fatal police shootings by U.S. police, 2015–2016: Patterns and answers from a new data set. The article was written by Jon M. Shane, Brian Lawton, and Zoë Swenson .
Some pertinent details:
In other words, whites are just as likely to be killed when dealing with the police, and are more likely to killed than blacks when not attacking the officer. These statistics of course run counter to the BLM narrative of pervasive racism that causes blacks to be endangered when dealing with police. This is a good example of why we need to look at statistics, rather than relying on emotional anecdotes. Seeing a black teenager murdered makes us emotionally want to agree with BLM, and it makes other people look askance at us if we do not. However, BLM’s narrative is now not merely unsupported by evidence, but contradicted by evidence.
Of course, there may be some cities where there is an anti-black disparity. More study would be needed. But the idea that there is a nation-wide problem is certainly shown to be incorrect by this study.
Why would whites be more likely to be killed when not attacking the officer? Perhaps it is because white officers are more careful when dealing with black suspects, because they do not want to be accused of racism if anything goes wrong (were I an officer, I think that would be a motivating factor in determining how I dealt with a situation).
There is one point in the research which does show support for some claims by BLM. Namely, although blacks are not more likely than whites to be killed during a stop, they are more likely to be stopped. See http://www.latimes.com/science/sciencenow/la-sci-sn-cops-race-injury-20160725-snap-story.html for details about this aspect of the research.
Whether the frequency of stops is itself the result of race, or whether it simply correlates with race due to unrelated causes, is not clear to me.