The reliable calculation of the phosphorescence energies of phosphor materials is at the core of designing efficient phosphorescent organic light-emitting diodes (PhOLEDs). Therefore, it is of paramount importance to have a robust computational protocol to perform those calculations in a black-box manner. In this work, we use Domain Based Local Pair Natural Orbital Coupled Cluster theory with single, double and perturbative triple excitation (DLPNO-CCSD(T)) calculations to attain the phosphorescence energies of a large pool of Pt (II) complexes. Several approaches to incorporate relativistic effects in our calculations were tested. In addition, we have used the DLPNO-CCSD(T) values (i.e., our best theoretical values) to assess the performance of different flavors of density functional theory including pure, hybrid, meta-hybrid, and range-separated functionals. Among the tested functionals, the M06HF functional provides the best values as compared with the DLPNO-CCSD(T) ones, with a mean absolute deviation (MAD) value of 0.14 eV. In its turn, and thanks to the increased accuracy achieved in the calculation of phosphorescence energies, we also demonstrate that not all the investigated complexes emit from their lowest lying triplet-state (T1). The outlier complexes include different complex photophysical scenarios and both Kasha and anti-Kasha types of complexes. Finally, we provide a general computational protocol to pre-screen whether T1 is actually the emissive state and to accurately calculate the phosphorescence energies of Pt (II) complexes.