Yes that's true. I was being somewhat charitable in affording it that benefit, but I did so because the article is the one that brought up the 1.9 legs example

Yes that's true. I was being somewhat charitable in affording it that benefit, but I did so because the article is the one that brought up the 1.9 legs example so that was kind of an implied benefit. And it's also true that you can use a "median" that picks one of the two, instead of splitting them, without any real downside if you want to preserve the feasibility of the statistic.