With the costs of political repression rising, non-democratic governments around the world rely on subtler strategies of preventive protest repression. In recent decades, a restrictive protest permit system has been used to dampen dissent from Hong Kong and Singapore, to Russia and Kazakhstan. Yet, due to a lack of appropriate data, we still know very little about the effect of the protest permit system on the opposition’s ability to forge broad coalitions in electoral autocracies. Do subtle strategies of repression, imposed through the protest permit system, bolster coordination among ideologically distant opposition groups, or do they trigger a ‘sorting’ effect instead, with opposition groups of the ‘toughest’ sort and highest resolve being more likely to take to the streets, despite repression? And, how does the decision to protest without a permit influence voters' perceptions of the opposition in these regimes? To gain traction on these questions, I rely on unusually detailed, subnational data on the organizers of protest in Russia from 2017 to 2018, combining them with novel survey evidence. Findings, which suggest that the protest permit system may trigger splits among more and less risk-acceptant opposition groups, and dampen voter support for the ‘extremist’ protest organizers, have implications for debates on the changing character of repression and authoritarian resilience.