@LittleBeefy and
@ntsqd
Nerd alert received!
Can we please stick to the physics? It turns out that
@ntsqd is correct it his original theory. BECAUSE his theory assumes that the caliper is sufficiently rigid so as to prevent "cocking" of the piston. So the forces that are imbalanced with respect to the centroid of the pad are countered by bending moment in the piston to bore interface. So we all can agree that it is a horrid design concept...but there is no reason why the piston needs to be centered on the pad. (in either plane)
It's just moronic to do so.
For the same reason that moving the caliper radially in and out will impact braking torque, moving the pad in and out will have the same effect. Obviously, the act of reducing the brake pad area necessarily changes Mu. (because of the way Mu is derived...) But if the brake dyno applies the same nominal hydraulic pressure on the same piston, the clamping force will be constant. The same clamping force will deliver the same force on a either pad. If you draw a free body diagram of the pad, you will see the same clamp force, on a smaller pad area, located farther from the axis. The brake torque will be increased, the pad will see greater pressure, and the pad will also see some very weird (and foolish) caliper binding, piston binding, and pad twisting, and a bunch of other horrid things.
At least, that's the way it works in theory. But as you correctly point out: In Theory...I am a millionaire on a beach. In practice, I am here with you.