While the formal valence and charge state concepts have been tremendously important in materials physics and chemistry, their very loose connection to actual charge leads to uncertainties in modeling behavior and interpreting data. We point out, taking several transition metal oxides (La2VCuO6, YNiO3, CaFeO3, AgNiO2, V4O7) as examples, that while dividing the crystal charge into atomic contributions is an ill-posed activity, the 3d occupation of a cation (and more particularly, differences) is readily available in first principles calculations. We discuss these examples, which include distinct charge states and charge-order (or disproportionation) systems, where different "charge states" of cations have identical 3d orbital occupation. Implications for theoretical modeling of such charge states and charge-ordering mechanisms are discussed.Spin ordering, and often orbital ordering, is normally unambiguous, as these properties are subject to direct observation by magnetic and spectroscopic measurements, respectively. Charge ordering (CO) and the actual charge of an ion is rarely measured directly, and the formal charge of an ion in the solid state can be a point of confusion and contention. Valence, oxidation number, and formal charge are concepts borrowed from chemistry, where it is emphasized they do not represent actual charge [1,2] and have even been labeled hypothetical.[1] As the interplay between spin, charge, orbital, and lattice degrees of freedom become more closely watched [3] and acknowledged to be a complex phenomenon, disproportionation and CO have become entrenched as the explanation of several high profile metalinsulator transitions (MIT). The possibility that CO in the charge transfer regime is associated with the oxygen sublattice, with negligible participation of the metal, has been raised[4] and considered as an alternative. [5] Charge density is a physical observable of condensed matter, and the desire to assign charge to atoms has evident pedagogical value, so theoretical approaches have been devised to share it amongst constituent nuclei. Mulliken charge population, which socializes shared charge (divides it evenly between overlapping orbitals) is notoriously sensitive to the local orbital basis set that is required to specify it. Born effective charges are dynamical properties and are often quite different from any conceivable formal charge or actual charge. Integrations over various volumes have been used a great deal, but dividing the static crystal charge density into atomic contributions is, undeniably, an ill-defined activity.A possibility that has not been utilized is that, taking 3d oxides as an example, there is a directly relevant metric that is well defined: the d occupation n d . This quantity is in fact what the physical picture of formal charge or oxidation state brings to mind. 3d cations, in their various environments and charge states, have maxima in their spherically averaged radial densityρ(r) in the range 0.6-0.9 a o . At this short distance from the nucleus, the only oth...