Competence Based Medical Education (CBME) is a popular buzzword in contemporary medical education. Nonetheless,  it seems that at times it is difficult to see the wood for the trees. There is a fundamental issue regarding  BCME. Without appropriately defining competency as well as its relationship to future clinical practice, CBME may not serve its purpose. 
 The word ‘competency’, means: ‘The ability to do something successfully or efficiently’. CBME is based on presenting the ability to perform appropriately, yet it does not consider the likelihood to perform appropriately. 
To address the shortcoming of CBME there is a need to apply a number of measures. First assessing trainees should focus on dichotomous judgement of observed clinical practice, that is has the trainee performed well (the required task); second, there is a need to accumulate the longitudinal data of the trainee’s performance in each task they are assessed for; third, there is a need to consider the poor /unsatisfactory performance observed and  based on the longitudinal trend to predict the likelihood that the trainee will perform poorly in the future.  The focus on poor performance is critical since appropriate performance is not risky, whereas poor performance may compromise patient care. The talk will focus on dichotomous judgement of clinical skills as well as on the determination of the likelihood for poor or competent clinical performance.