As they say in the article, they didn't choose the responses to be neutral; they simply used common reasons people were voluntarily sending through their contact page.

I agree that that is poor methodology, since it doesn't include the opinions of anyone who didn't care enough to volunteer a response. But to their credit there is an "Other" option, the responses to which are linked in the article... Many of the answers fall into one or more of the categories already in the survey. It looks like the most common answer that isn't already covered is "I just wanted to try the game out and don't know how much it's worth".