I'm following you, I agree with all this. I'd say the study I posted does somewhat account for this because it looked at many different geographic areas and the

I'm following you, I agree with all this. I'd say the study I posted does somewhat account for this because it looked at many different geographic areas and therefore income levels, and found a higher proportion of Black deaths in all of them. But I agree it doesn't directly consider what you're asking. I'm not sure if the data exists for this. You'd need to record a huge sampling of police interactions and somehow classify the behavior of the civilian independent of the police to see if they were "aggressive" or "escalated", which is quite difficult. And part of the issue is that many behaviors that police often claim are aggressive are obviously not, so who gets to define these words? Very tough. Overall, I think I agree that given the data we've discussed in this thread, there's no way to prove whether or not disproportionate deaths are caused by police bias. I started this conversation to respond to a question in the Twitter thread you linked: "I'm going to keep tweeting this until someone can explain to me how this is possible if there is truly pervasive racial bias in policing". I maintain that it is absolutely possible, and seems likely given the data we have here, but definitely not proven.