The rate of galaxy cluster eccentricity evolution is useful in understanding large scale structure. Rapid evolution for $z < $ 0.13 has been found in two different observed cluster samples. We present an analysis of projections of 41 clusters produced in hydrodynamic simulations augmented with radiative cooling and 43 clusters from adiabatic simulations. This new, larger set of simulated clusters strengthens the claims of previous eccentricity studies. We find very slow evolution in simulated clusters, significantly different from the reported rates of observational eccentricity evolution. We estimate the rate of change of eccentricity with redshift and compare the rates between simulated and observed clusters. We also use a variable aperture radius to compute the eccentricity, r$_{200}$. This method is much more robust than the fixed aperture radius used in previous studies. Apparently radiative cooling does not change cluster morphology on scales large enough to alter eccentricity. The discrepancy between simulated and observed cluster eccentricity remains. Observational bias or incomplete physics in simulations must be present to produce halos that evolve so differently. Comment: ApJ, in press, minor revisions