OBJECTIVE: To determine which of the 21 general pediatrics milestone subcompetencies are most difficult to assess using traditional methodologies and which are best suited to simulation-based assessment. METHODS: We surveyed 2 samples: pediatric simulation experts and pediatric program directors. Respondents were asked about current use of simulation for assessment and to select 5 of the 21 pediatric subcompetencies most difficult to assess using traditional methods and the 5 best suited to simulation-based assessment. Spearman rank correlation was used to determine a correlation between how the 2 samples ranked the subcompetencies. RESULTS: Forty-eight percent (29 of 60) simulation experts and 20% (115 of 571) program directors completed the survey. Few respondents reported using simulation for summative assessment. There are clear differences across the pediatric subcompetencies in perceived difficulty of assessment and suitability to simulation-based assessment. The 3 most difficult to assess subcompetencies were "recognize ambiguity," "demonstrate emotional insight," and "identify one's own strengths and deficiencies." The subcompetencies most suitable to assessment using simulation were "interprofessional teamwork," "clinical decision making," and "effective communication." Program directors and simulation experts had high agreement for both questions: difficult to assess (rho = 0.76, P < .001) and suitable to simulation-based assessment (rho = 0.94, P < .001). CONCLUSIONS: Several general pediatrics milestone subcompetencies were identified by pediatric simulation experts and pediatric program directors as difficult to assess using current methodologies and as amenable to simulation-based assessment. The pediatric simulation community should target development of simulation-based assessment tools to these areas.