How does the political selection compare across secular and Islamist parties in the Middle East and North Africa region? Even though the political selection literature burgeoned in recent years, we know much less about who becomes a politician in the MENA region and across this important socio-political cleavage. This paper studies this question by focusing on most recent local elections in Turkey and Tunisia. Using original sources of data (detailed and large-scale candidate surveys, accompanying household surveys in the same municipalities, and interviews with party officials), the paper demonstrates that in both countries, the Islamist parties are able to select more competent candidates with higher levels of educational attainment, organization-building experience and prior civic engagement, even though the secular-leaning segments of the society are usually more endowed with these assets. In turn, secular party candidates are more likely to have a family history of political engagement and administrative office-holding, suggesting that they are more connected. I argue that the relative negative selection among seculars likely stem from demand-side reasons: Secular parties suffer from intense intra-party competition, as a result of which secular party elites prioritize loyalty over competence in candidate selection. I provide evidence for this argument from interviews with party officials and a conjoint experiment in both Turkey and Tunisia.