Let v = high violence, d = dem mayor, r = republican mayor. You seem to want to compare P(v|d) to P(v|r). We have little to no data for comparison because ther

Let v = high violence, d = dem mayor, r = republican mayor. You seem to want to compare P(v|d) to P(v|r). We have little to no data for comparison because there are so few examples of cities with republican mayors. Those that exist are amongst the smaller American cities. This means you need to disentangle the effect of city size on violence rates from the effect of the mayor's party. I repeat: There are 10 Republican mayors amongst the 50 largest cities, and furthermore, these are heavily concentrated at the low end. https://en.m.wikipedia.org/wiki/List_of_mayors_of_the_50_lar https://en.m.wikipedia.org/wiki/List_of_mayors_of_the_50_lar... It turns out the relationship between crime and population is non linear, and that nonlinearity is also true outside of the US and it's political context. https://crimesciencejournal.biomedcentral.com/articles/10.11 https://crimesciencejournal.biomedcentral.com/articles/10.11... So, what you're seeing is bigger cities with higher per capita rates due to the underlying relationship between population density and crime. It also happens to be the case that in the US, urban areas strongly prefer Democratic mayors.