This study evaluated the use of different degree day (DD) models, developmental thresholds and developmental data sources for estimating postmortem interval (PMI) based on developmental rates of blow flies (Diptera: Calliphoridae). Blow fly evidence was collected from three mock crime scenes and sent for blind analysis of PMI. PMI estimates were calculated using averaging, single sine, double sine, single triangle, and double triangle models of DD calculation with minimum developmental thresholds of 0, 6 and 10 degrees C and Anderson's, Byrd and Allen's, Greenberg's, Nishida's and Kamal's developmental data sets. These estimates were compared with the actual PMI to determine which combination of DD model, developmental threshold and developmental data provided the most accurate estimate. For all three cases, at least one method gave a PMI estimation that was within 1 day of the actual date of death. There was no variability between PMI estimated using the five models of DD calculation. Overall, as lower developmental threshold increased, the PMI interval estimates increased. Depending on the developmental data set used, increasing the lower developmental threshold caused some overestimations of the date of oviposition, with estimates given that were prior to the actual death. Future PMI estimates should include error ranges, so that overestimation of PMI is avoided.