It still works out if we assume young people are just as likely as old men to be killed by police. Restricting us to young people would probably reduce the deno

It still works out if we assume young people are just as likely as old men to be killed by police. Restricting us to young people would probably reduce the denominator of "300/19740000" enough to counteract the fact that the exponent of "69" is reduced. In fact, if you assume that people are young for x% of their lives, and x% of black people are young, this exactly cancels out and you still get `0.001048` odds of being killed by police.