Background: The metacognitive approach by Wells and colleagues has gained empirical support with a broad range of symptoms. The Metacognitive Anger Processing (MAP) scale was developed to provide a metacognitive measure on anger (Moeller, 2016). In the preliminary validation, three components were identified (positive beliefs, negative beliefs and rumination) to be positively correlated with the anger. Aims: To validate the MAP in a sample of mixed clinical patients (n = 88) and a sample of male forensic patients (n = 54). Method: The MAP was administered together with measures of metacognition, anger, rumination, anxiety and depressive symptoms. Results: The MAP showed acceptable scalability and excellent reliability. Convergent validity was evidenced using the general metacognitive measure (MCQ-30), and concurrent validity was supported using two different anger measures (STAXI-2 and NAS). Conclusions: The MAP has promising potential to assess anger regulation problems by providing a framework on angry rumination as well as the belief structures that proposedly drive the selection of this maladaptive processing strategy as suggested in the metacognitive model. These findings may have implications for clinical interventions. For example, conducting functional analyses on anger rumination could increase the understanding of dysregulated anger processing and lead to new interventions focused on shifting thinking style.