This is a delusion - they are at least 2 orders of magnitude more likely to be assaulted or killed by another (non-police) black person.
This is yet another scenario where a relatively minor source of risk gets vastly disproportionate coverage and almost everyone falls for it.
ETA: it’s funny that this straightforward statement of objective fact is being so poorly received.
But the more important factor is that in most homicides the victim and the perpetrator know one another. On that basis minorities, and particularly middle-class and up members of minority groups, have good reason to be more concerned about abuse and/or murder at the hands of authorities as a more important consideration than truly random violence.
Also, "at least 2 orders of magnitude" sounds like a big overestimate. In 2018, 2570 black people were killed in the US by other black people. About 250 were shot and killed by the police, in cases where the police self-admitted to killing them. This does not include the non-shooting deaths, either those that occur during an arrest, or while in custody.
Even if it were only 250/2570, that's still only about 1 order of magnitude.
You may be getting downvoted because of your false equivalences and your incorrect statistics, not because of your "objectivity."
There are (at least) two different conditional probabilities we might be interested in here:
1. If a black child is assaulted or killed, who was responsible?
In this case you are partially correct: it is (by about one order of magnitude, not at least two) more likely that it was another non-police black person.
2. If a black child interacts with a particular person (e.g. police, or a non-police black person), how likely are they to be assaulted or killed?
If we are willing to assume (as is certainly the case) that black children have more interactions with non-police black people than they do with police, by more than one order of magnitude, then we can conclude that police are more likely to assault or kill than a non-police black person, conditional on an interaction. It is thus extremely reasonable for black parents to teach their children to avoid police and to try to keep them away from police.
Cops take more in assets than robbers. [1]
[0] https://granta.com/violence-in-blue/
[1] https://www.washingtonpost.com/news/wonk/wp/2015/11/23/cops-...
Nota Bene: your down votes are from a lack of evidence combined with an aggressive demeanor. Neither are a good look.
My mistake - this does not change the nature of my argument at all. That's still a massive difference in risk.
Why do you think it makes sense to ignore the majority of homicides (which don't fit a popular political narrative)? That's like saying "Most car crashes happen near home, so on that basis people have good reason to be more concerned about plane crashes."
If you are middle class and non-white your odds of being hassled by the police are quite high and probably validated by experience. The odds of such an experience escalating into violence or death may be empirically small but each and every instance of interaction with authorities is an instance of greatly increased risk and a risk that is in large part out of your control. This is something worth being concerned about and that concern has an impact on your day-to-day behaviour and well-being.
Being more concerned about police violence than random violence makes perfect sense to me. I suspect that if you dig into it that concern is also empirically more rational.