Keep in mind this is the American Community Survey, which, unlike the full census, only counts about 1% of the population. Given that cyclist commuters only o

Keep in mind this is the American Community Survey, which, unlike the full census, only counts about 1% of the population. Given that cyclist commuters only only account for 0.3% of the population already, the large swings become a little less statistically significant. I don’t find it hard to believe that cyclist commuting decreased by a few percent overall, but it is hard to believe that cycling decreased in Oakland by 25% while increasing in Philadelphia by 20% in one year.