One problem is that the "50% factor" has no mathematical or theoretical basis - it's "what works".
It fails to account for the fact that almost all of the rod's small end is not pure reciprocating weight, that material above the pin center is traveling opposite the rod beam, that the difference in forces between long and short rods is completely absent, and that other very large forces (cylinder pressure and vacuum) are completely ignored.
De-constructing over and/or under-balancing to justify an observed result is not science, it's rationalizing.

Read my article here: http://victorylibrary.com/mopar/crank-bal.htm