The literature on authoritarian regimes has studied in depth why autocrats decide to hold or not hold elections. However, much less attention has been devoted to the autocrats’ electoral rule choices. This paper seeks to fill this gap by studying i) when autocrats decide to reform electoral laws, and ii) which institutional choices they take. The paper argues that the Regime's Strength and the presence of Ethnic cleavages are the main drivers of autocrats’ electoral rule choices. In particular, I argue that strong autocrats will be less prone to propose proportional reforms than weak ones, who will concede more easily to the opposition’s demands for proportionality. The presence of ethnic cleavages, in turn, will provide crucial information to the autocrat on the territorial distribution of power, thus promoting reforms that include distortive elements of representation (i.e. malapportionment and gerrymandering). To test these arguments, I leverage a novel database containing data for 117 countries and 134 reforms from 1979 to 2023. Results show that regime strength and ethnic cleavages strongly condition autocrats’ electoral rule choices. The paper has important implications for the literature on authoritarian regimes. First, it presents a novel argument for explaining the institutional decisions of autocrats. Second, it goes beyond the distinction between majoritarian and proportional reforms, focusing on whether the reforms have introduced elements that distort representation or not, allowing a better understanding of this phenomenon.