BACKGROUND Acute subdural haematomas (ASDH) occur commonly following traumatic brain injury and may be evacuated by either craniotomy (CR) or decompressive craniectomy (DC). We reviewed a series of consecutive patients undergoing evacuation of a traumatic ASDH at a regional centre, comparing observed clinical outcomes (assessed by Glasgow Outcome Scale at six months) with those predicted by the CRASH-CT prognostic model. METHODS Retrospective review of prospectively collected data. RESULTS Ninety-one patients were identified (51 DC and 40 CR ). Eighty-five had available admission data sets from which predicted outcome could be calculated. The DC group were younger than the CR group (p = 0.015). The DC group also had a greater proportion of patients whose pre-intubation GCS was ≤8 (p = 0.001), with significant extracranial injuries (p = 0.001) and obliterated basal cisterns (p = 0.001) on their pre-operative CT scan. Bone flaps in the DC group (n = 45) were longer (mean 11.6 cm; 95 % CI: 11.1-12.1) in comparison to bone flaps in the CR (n = 34) group [(mean 10.2 cm; 95 % CI: 9.35 - 10.9); p = 0.0024] The mean CRASH-CT predicted risk of 14-day mortality and of unfavourable outcome at six months was significantly higher in the DC group compared with the CR group. Eighty-eight patients had available 6-month Glasgow Outcome Scale scores. Favourable outcomes were observed in 42 % of DC versus 45 % of CR (p = 0.83). The overall mortality rate was 38 % in DC versus 32 % in CR (p = 0.65). The standardised morbidity ratio (observed/expected unfavourable outcomes) was 0.75 (95 % CI: 0.51-1.07) for DC and 0.90 (95 % CI: 0.57-1.35) for CR. CONCLUSIONS CR and DC for traumatic ASDH are both commonly used for primary evacuation of ASDH. Primary DC may be more effective than CR for selected patients with ASDH. Class I evidence is required in order to refine the indications for DC following evacuation of ASDH.