Conversation:

You state that 50% will always be above or below average. However in your example data 2 only 1 is above the average of 17. This shows how only 1 very high or low number can skew the results

Actually, he stated that IF 50% were above the average THEN 50% must be below the average. Not that the average is at the 50% point for all populations. Still, I can see how you might find that part of the article a bit confusing.

oops