If you look at the actual shootings, it's a bunch of tales of people engaging in mortal combat with the police. You might think, oh, police are still reacting more violently in similar situations. But we can check for that: In the past decade, police have been feloniously killed by black criminals at 3.7 times the rate of other demographics (199 out of 535, iirc, see the Law Enforcement Officers Feloniously Killed section of the Uniform Crime Report, Table 42). If black guys were fighting police at the same rate as other groups, you'd expect a much much lower number.
What you might still have an argument for is that police could do a better job overall of not killing people. I mean, I would make that argument, in specific ways.
You can't assess racism in the criminal justice system if you start by assuming that there is no racism in the criminal justice system’s determination of who is committing crimes.
In reality, almost all of the people killing cops are committing crimes.
In fact, the statistic for felonious killings of LEO doesn't even require specifying the exact perp -- it's just the race of the perp. One category is "unknown race."
I should add, the biggest hole in your comment is this notion that they're not only convicting the wrong guy, they're getting the wrong race and that's throwing off the numbers (by a factor of 3? Or 8, depending on crime?) As if the victim's getting attacked by a white guy and she's saying a black guy did it, to the point of a 3x crime rate. And the security cameras got the color wrong, too.
It's like you're imagining this great holocaust of white assault victims suffering from having the wrong guy prosecuted for attacking them, with half the perps getting replaced by some random black guy.
What you might still have an argument for is that police could do a better job overall of not killing people. I mean, I would make that argument, in specific ways.