Because that's the standard I'm calling for.
"Is there evidence of 99% of supposedly non-complicit blacks standing with victims of black crime, insisting on justice for those guilty?"
If you assume that everyone in a group is guilty of some crime committed by a tiny fraction because there isn't evidence that 99% of them condemn the guilty, then you're participating in a conspiracy and you're validating a dangerous line of reasoning that can be weaponized against anyone, including groups that you care about and/or belong to.
It's not the same at all.
The idea that police nationally get together and conspire violence against the public or even a particular racial group is beyond absurd. This is among the most dubious of conspiracy theories, and I don't know how such theorists can hope to be regarded as credible, reasonable people.
"Bad apple" arguments only work if there's actual evidence of "good apples" working to expel the bad ones.