Hi @ResponseRules56 and welcome. The Custom Test dialog uses the paramaterization as described in this link.
https://www.jmp.com/support/help/en/15.2/#page/jmp/nominal-factors.shtml#
So, that can make it challenging to get the estimates you want. Let's take a relatively simple case like a one-way ANOVA: one factor, 4 levels (A, B, C, D). Say you want to estimate A(mean) - D(mean).
Custom Test column is:
Int
A
B
C
A, D, and A-D are:
1 1 0
1 -1 2
0 -1 1
0 -1 1
It gets more complicated with interactions etc. It can be more easily seen in the Effect Details/LS Means Contrast option within the output (and the Test Detail and Parameter Function under the results therein). Take a look and come back with any questions.