If I use HypothesisCoding(), I can label everything nicely (yay!), but it insists on re-weighting my matrix for me, so if I put in the matrix I want to use, it changes it into something else. I have to put in a different matrix, with the positive term as 1 and the negative terms as fractions adding up to 1, to trick it into reweighting the matrix into what I actually want it to be.
I LEARNED HOW TO BUILD CONTRAST MATRICES
BY HAND JULIA! Don't try to do it for me. You're just breaking it!