Colors/legend could be much more clear. Initially it looked like each color should have one descriptor in the lower right legend area. However, it became clear that the descriptors are for all of those runs.
Is the difference between the average of the runs with and average without intake, 10 hp, 12 hp or something else?
It appears that the without runs show a 10 hp spread and runs with a 4 hp spread.
Thus, most importantly you can't exactly claim a true gain equal to the measurement variation...
I'm certainly not sold on a 10 or 12 hp gain.