Numpy has some pretty powerful bin counting functions.
>>> binplace = np.digitize(avgs, bins) #Returns which bin an average belongs
>>> binplace
array([1, 6, 2, 3, 5, 4, 1, 4, 6, 6, 2, 3, 5, 6, 3, 1, 2, 3, 5, 7, 5, 3, 4])
>>> np.where(binplace == 1)
(array([ 0, 6, 15]),)
>>> np.where(binplace == 2)
(array([ 2, 10, 16]),)
>>> avgs[np.where(binplace == 1)]
array([ 11.02, 11.3 , 11.01])