Yep. well put.. the same can be said of any hinge when side lean is involved. And you'll see a group of fibers that stand higher than the rest in many hinges. These fibers are thought to have held the best. They are actually at the mid-point between tension and compression. In a balanced tree, they tower-up right in the middle of the hinge. In leaners they will indicate how much lean a tree had by how close they are to the edge of the hinge.In the case of an usual limb (not a rigged limb, nor a well balanced free standing axis), the bottom is compressed and the top is stretched. In between, the fibers are less and less stressed as long as you progress toward the middle. Until a point (line actually) where the constrain falls to zero and then inverts to the other category.
Cutting fibers at the top or at the bottom only reduces the actual thickness of the layer of working fibers, but you find always the three areas previously cited, top stretched, middle quite and bottom compressed (its a continuous gradient actually). The global load stays the same, but less and less fibers are here to hold it, so locally the stress on the fibers increases dramatically, the maximum being at the outer fibers top/bottom. Top increases its stretch, bottom increases its compression, no matter where you cut. The zero line lives the middle of the limb to stay in the middle of the remaining layer of fibers. You can't suppress a category by cutting, it's only a progressive reallocation of the roles, driven by the quantity of survival fibers.