[mc32]

If you look at per-police interaction statistics the narrative changes. All kinds of people die during interactions with police.

Now, what we can claim is that historical injustices have led to increased ratios of police interaction by different populations. That’s true. However, per interaction, the police are not more likely to be fatal to one group vs another. That’s a misrepresentation.

See: Simpson’s Paradox.

Given this it should behoove media and leaders to institute root changes that improve the “interaction” frequency.

[perl4ever]

What does the percentage of fatalities per interaction have to do with justice?

For the same given percentage, they could all be people who were genuine threats, or all people reaching for their wallets in their own garage.

The metric is not measuring what matters.