Here's another way of looking at some of your data. A scatter graph with some lines of best fit (5th order polynomials) drawn in.
1. If the
measured data were consistent, the lines would be straight left /right.
2. If the
measured difference between configurations was consistent, the gap between the lines would be equal.
Obviously neither of these in the case.
However, it can be seen that
generally the 2 Bar pressures give lower energy consumption than the other pressures I've graphed.
This is just a subset of the data (Direction A for just these pressures) but this is the sort of analysis I suggest you do for all the data you have collected. It would be interesting to see the results presented in this way for your aerodynamic modifications.