Gustaf was talking mean. Thefactor inexplicably switched to median at the end of his post, but I did not read it (as I made the understandable assumption that people to do suddenly and abruptly change their topic in the middle of their posts) as such, so I posted the link to tell him that no, the mean center of population has not been in Indiana since 1950.
Gustaf wrote "centre of population" (sic). The Census Bureau defines both a "mean center of population" and a "median center of population".
It is not clear which measure Gustaf was referring to, since it is not true that the center of population by either measure has always been in a state carried by the presidential winner.
Since the median center of population has indeed been in Indiana since 1950 (and also 1900 and 1910), then the clear implication is that thefactor was covering both cases, whichever Gustaf meant.
Right. In any case, it's still a pretty remarkable phenomenon. Since 1840, the state with the mean center of population has gone with the winner in all except for 1916, which was kind of a strange election because Wilson narrowly won on the basis of his incumbency, which he only had in the first place because of Roosevelt's ultra-successful third-party challenge in 1912... probably the most successful "third party" challenge since 1860, at least.
About the 1910-1930 loopback... the closing of the frontier, urbanization, and immigration all probably contributed. This was the period when northeastern cities were the most dominant in terms of their percentage of the national population.
Before 1930, there was clearly a gradual northward movement of population, again due to both immigration and internal migration, since the south had been devastated so much economically by the Civil War. The New Deal I think played a role in jumpstarting true industrialization in the south and since then the population center has been moving in that direction. That's a case where being in the electoral majority translated into clear economic gains for a region.