This review examines the policy and practice of interventions with male perpetrators of domestic violence in light of the widely accepted principles of evidence-based practice. Thus far, these policies and practices have enjoyed immunity from the external, empirical accountability available through implementing the findings from evaluations research and other empirical practice analyses. This immunity is supported by a policy framework where, for example, the state certifying agencies may presumptively forbid methods of intervention, with no obligation to empirically assess their efficacy or safety, that contradict the approved model. Based on the review of findings from both explanatory research and interventions research, evidence-based recommendations for policy and program change are proposed.