A framework is developed for modeling the resonance Raman (RR) intensities of metalloporphyrins, with a view toward rationalizing the enhancement patterns observed in the spectra of heme proteins. The geometry of the S-2 excited state of nickel(II) porphine is computed using INDO/1s methods, and the structural changes resulting from S-0-S-2 photoexcitation are projected onto the ground-state normal modes to calculate the intensity of each Raman-active vibration. The RR intensities derive mainly from expansion of the CalphaCm and CbetaCbeta bonds in the excited state, with the relative intensities strongly influenced by the phasing between CalphaCm and CbetaCbeta stretching coordinates. Analysis of the vs overtone shows the INDO predicted geometry changes to be about 25% too large. Results are compared at successive levels of approximation, demonstrating that inclusion of displacements along; bending coordinates in the excited state are essential, as are frequency-dependent scaling factors which are determined from the absorption spectrum by the transform approach to RR scattering. Finally, the activation of non-totally symmetric modes by an A-term mechanism is modeled by distortion of the excited state along a b(1g) coordinate. Enhancement of the experimentally observed non-totally symmetric modes is correctly predicted, although quantitative modeling of their intensity requires the inclusion of non-Condon coupling.