Objective To demonstrate first that a ventrodorsal radiographic (VDR) projection is inadequate for obtaining accurate acetabular cup position (ACP) measurements after total hip replacement (THR), and second to validate use of the lateral radiographic (LR) projection to measure ACP. Study Design In vitro study. Methods Eleven VDRs were obtained after insertion of a cup into the acetabulum of a Sawbones pelvis fixed at an approximated angle. For each VDR, the pelvis was placed in varying degrees of craniocaudal pelvic tilt (CCPT). Fifteen LRs were obtained of a Sawbones pelvis with varying ACPs. Five blinded observers made previously described measurements of ACP on each VDR and angle of lateral opening (ALO) and coronal retroversion (CR) on each LR. Results An almost perfect level of agreement was observed between 5 observers for the repeatability of both VDR and LR measurements. When varying degrees of CCPT were introduced, there was no agreement in the measurements of ACP from VDRs. In all cases, and at all ACPs, measurements taken from LRs were in agreement. Conclusion Undetectable CCPT causes significant variability in ACP measurements on VDRs predisposing postoperative radiographic measurements to inaccuracy. Measurements obtained from an LR of CR and ALO are accurate and consistent between observers and it should be used to measure ACP. The VDR should be performed to assess bone and implant integrity but not to obtain measurements of ACP because of CCPT causing measurements to be unreliable. Measurements of ACP obtained from an LR in conjunction with femoral anteversion may allow for investigation of the influence of ACP on coxofemoral joint luxation after THR.