Objective: Despite high-Quality paediatric head trauma clinical prediction rules, the management of otherwise asymptomatic young children with scalp haematomas (SH) can be difficult. We determined the risk of intracranial injury when SH is the only predictor variable using definitions from the Pediatric Emergency Care Applied Research Network (PECARN) and Children's Head Injury Algorithm for the Prediction of Important Clinical Events (CHALICE) head trauma rules. Design: Planned secondary analysis of a multicentre prospective observational study. Setting: Ten emergency departments in Australia and New Zealand. Patients: Children <2 years with head trauma (n=5237). Interventions: We used the PECARN (any non-Frontal haematoma) and CHALICE (>5 cm haematoma in any region of the head) rule-Based definition of isolated SH in both children <1 year and <2 years. Main outcome measures: Clinically important traumatic brain injury (ciTBI; ie, death, neurosurgery, intubation >24 hours or positive CT scan in association with hospitalisation ≥2 nights for traumatic brain injury). Results: In children <1 year with isolated SH as per PECARN rule, the risk of ciTBI was 0.0% (0/109; 95% CI 0.0% to 3.3%); in those with isolated SH as defined by the CHALICE, it was 20.0% (7/35; 95% CI 8.4% to 36.9%) with one patient requiring neurosurgery. Results for children <2 years and when using rule specific outcomes were similar. Conclusions: In young children with SH as an isolated finding after head trauma, use of the definitions of both rules will aid clinicians in determining the level of risk of ciTBI and therefore in deciding whether to do a CT scan. Trial registration number: ACTRN12614000463673.