What is interesting about that table is that it presents not raw counts or counts scaled to the proportions of people of color in our communities, but instead c

What is interesting about that table is that it presents not raw counts or counts scaled to the proportions of people of color in our communities, but instead counts "per 10,000 stops". So, if one black guy gets stopped 30,000 times (because he "looked urban and suspicious") and is only shot in 5/30,000 times then he is obviously much better off than the 1 pinky-tan guy who is only stopped twice ever but gets shot to death in that 1/2 times).