♪alt13
Well-Known Member
Interesting topic, but the information is not culled from a survey of people outside arenas or ticket holders. It says right on the posted graphic that the sources are Nielsen (television viewers) and Google trends. Furthermore, Nate Silver, author of the article, says upfront that it's just a crude estimate. His methodology, in his own words:
I used Nielsen estimates of the racial composition of the population in each of the United States’ 210 media markets;
I used the frequency of Google searches for each NBA team in each media market as a proxy for its per-capita popularity;
I multiplied each team’s Google search frequency in each media market by the population by race there, then summed the totals to produce an overall estimate of the racial distribution of its fans;
I recalibrated the estimates to ensure that the whole matched the sum of the parts. In other words, I added or subtracted from the fans assigned from each racial group to each team such that the sum total matched my estimate of the overall distribution of NBA fandom throughout the country (e.g. 46 percent white, 31 percent black, and so forth). The estimates were weighted by the overall popularity of NBA teams, according to their number of Google searches.
So he made it up then.