A precise calculational analysis of the INEL manganese bath experiment to measure beryllium neutron multiplication has been performed. The goal throughout the analysis was the minimization of all sources of error due to the calculational model and method. An extremely detailed three-dimensional Monte Carlo geometry model was developed for use with the code MCNP. Calculations were performed for a bare-source and four beryllium sample configurations for both DT and 252Cf neutron sources. The primary objective of the analysis was the calculation of various neutron-economy parameters applied as experimental corrections, either directly or as verification of measured values. The most significant of these were the tank leakage, duct streaming, structural absorption, fractional bath capture in manganese, high-energy parasitic bath absorption, neutron multiplication in other materials, and indirect absorption and multiplication in beryllium.