You would expect that to be the case, however, by default
seaborn.pointplot uses the average estimator to calculate the number for each hour. So the numbers you are seeing on the y-axis is the average number of bikes shared for each hour. Since the number of bikes shared is not equal for category
workingday=1 the two averages for those categories do not add up. Using
seaborn.pointplot does give the expected results.