>>17696
Scatter plots may be attractive. However, without a curve fit that shows correlation, and without the attendant statistics that show "goodness of fit" I just can't see choosing the scatter plot over the pie chart. Though, maybe there is a curve fit on method A. I just can't see it in the picture. On the other hand, maybe the scatter plot has enough resolution for Twilight to crunch the numbers and do a curve fit herself. That might change her thinking on the which method to choose.