Have you tried mapping all the other causes of death?

Also, you're comparing two disparate stretches of time leading up to the same endpoint. That seems like bad practice.

I'm also not seeing where the wapo link you provided gives data on the party affiliation of the people who were shot. Is that data just a reflection of the location, then?

I'm not a data scientist, but this feels very pegboards-and-strings-across-the-room kind of scenario.