Triage aims to identify patients in the emergency department who need immediate care and those who can safely wait in order to decrease mortality and morbidity. The Manchester Triage system (MTS) is based on expert opinion. It is based on flowcharts and specific decision rules which allocate urgency level (1 to 5) for a broad population visiting the emergency department. The reliability is dependent of the clinical experience and MTS training of the triagist, unambiguous of the system and diversity in flowcharts and discriminators to match the patients. The fundamental problem in studies to validate triage tools is the lack of consensus of the reference standard for “true urgency”. Does the MTS urgency predicts true urgency? A proxy reference standard can be a combination of vital signs, disease severity, resource use and follow-up or just hospitalization. The validity of the MTS in a large prospective study of 17600 children in the Netherlands was moderate and low for febrile children compared to the combined reference standard. Age related fever modifications were validated in a new sample and improved specificity while sensitivity remained similar. The modified MTS agreed in 37% with the reference standard of urgency and 36% were overtriage and 13% undertriage (2% by > 1 category). Over and undertriage need to be balanced to have a safe triage system. Performance of the MTS in different settings and in large populations need to be studied. New modifications on flowchart or subgroup level might further improve the MTS.