I think basically there’s a whole lot of ‘very democratic names’ that are relatively unique. I’ll bet if you made a histogram of all the names, with most common on the left out to least common on the right, there would be a very long, mostly blue tail out to the right. So, all you’ve got to do is have a pretty common name, and it’s going to appear that a fairly large % of names are ‘more democratic’ than yours, just because a whole lot of democrats out there have very rare names (I’m sure you can think of some…) and ‘Bob’ counts just as much as a category as, say, ‘Latoya.’

Just try putting in a few names in that website, and the skew is obvious. ‘Raymond’ is (a slim) majority democrat name, but ~60% of names are still somehow ‘more democratic’ than Raymond. If that’s the case, how much does Scott Alexander’s statistic even mean? It seems to me to say more about uniqueness of name choice among democrats than anything about republicans, except maybe conventionality of name choice.

I’m not sure “douchebag” is particularly “tribally coded” as you suggest — there are plenty of douchebags in San Francisco, and I doubt that more than 20% are Republican.

From what I’ve seen, “douchebag” = “youngish man who has more apparent success with women than I think is appropriate”. So does that mean Republicans are better at providing what women really want, or that Democrats are more likely to be envious little cunts?