Bruce Murphy says that there is a paradox in the pattern for voting in the Supreme Court race. The area with the most severe crime - Milwaukee County - voted for the candidate that was criticized for being more sympathetic to criminal defendants while those area with less crime - e.g., the counties surrounding Milwaukee voted for the candidate who was - although he was also criticized for being lenient toward criminal defendants - was probably perceived to be less sympathetic.
I think the campaign was about more than crime. Both sides criticized the other candidate as "soft on crime" and I think that the claims made by the Gableman camp came to stand in for the type of cue that we normally get from partisan affiliation. He was the "conservative" and the public is smart enough to know that, at the level of the supreme court, ideology matters even if they don't always understand how it matters.
In that sense, there really is no paradox in the fact that Butler carried heavily Democratic areas. He was perceived as the "liberal."
But Murphy's claim of a paradox in attitudes about crime is nevertheless valid. I think his explanation that residents of high crime areas are "more likely to know someone who has run into trouble with the police or is unjustly accused, and thus more likely to value those “loopholes” that our constitution established to protect the innocent" requires some evidence.
Another hypothesis is that black voters in particular have come to associate calls for "law and order" with racial insensitivity. Some of the reasons for this are valid (cops can act like a occupying force) and some are not (politicians stoke resentment over these issues).
Murphy says that those in low crime areas are "removed from the harm done by erroneous convictions or police brutality" and that "[i]t’s easy to support getting tough on all those bad people you’ve never met, and harder perhaps to see the value of constitutional protections for an accused criminal."
Again, I'd like to see some evidence that attitudes on these issues do differ among these populations. But even if they do, another hypothesis might be that those in high crime areas come - not to accept it - but to see it as inevitable; as the way that things are. They don't really think it can be different - nothing in their experience suggests this - and so are not attracted to those who claim to be able to make it so. Indeed, the fact that communities can come to believe that public safetly is impossible and that this actually contributes to further crime and disorder is one of the tenets of the "broken windows" approach to law enforcement.
Murphy's "paradox", though, should be attractive to those on the left. We all like to think that the position or canidate that we favor arises from our superior understanding. Gnosticism is not just for religion.