The literature contains several reports of Mie-like fits to angular-domain elastic scattering measurements from multiple cells or isolated mitochondria. In these studies, the sampling volume typically contains hundreds or thousands of mitochondria, allowing for the size distribution of mitochondria to be modeled as a smooth function, (e.g. Gaussian or log-normal) with a small number of free parameters. In the case of a single-cell volume containing significantly fewer mitochondria, the true size distribution will no longer be as smooth. Increasing the number of free parameters can lead to unstable fits, however, as the forward-directed angular scattering pattern from such a population illuminated with 785 nm light is a monotonically decaying radial function with few distinct features. Using simulations, we have investigated the limitations of modeling single-cell mitochondrial scattering using smooth population distributions of Mie scatterers. In different instances, the fidelity of the estimated size information can be limited by the number of organelles, the angular detection range, or the non-ideality of the data (both speckle and shot noise). We will describe the conditions under which each of these effects dominates. We will also discuss whether mean and standard deviation are the best sizes to report from such Mie modeling, or if there are other size parameters that have greater fidelity to the true, non-smooth size distributions.