Corruption scandals can reshape political careers in democratic elections. While voters typically punish corrupt politicians, scandals also impact the careers of those indirectly implicated through party affiliations or leadership roles, as voters extend this punishment to affiliated politicians. To mitigate these effects, indirectly implicated politicians often dissociate from overtly corrupt individuals. This paper examines how political dissociation serves as a strategy to reduce electoral punishment and explores its effects on the electoral fortunes of indirectly implicated politicians. Dissociation efforts include switching parties or, within the same party, distancing through public statements, limiting collaboration, or reconfiguring voting alignments. The influence of political and corruption networks—defined by voting patterns, party ties, and personal characteristics—shapes dissociation decisions and their electoral consequences. The paper argues that dissociation within a network moderates the negative electoral impact of indirect corruption involvement. A game-theoretical model of network formation shows how politicians sever ties with implicated actors to reduce electoral risks as the cost of being indirectly linked to corruption increases. To test this argument, the paper analyzes nearly 30 years (1992–2019) of network data from Brazilian congresspersons. Political ties are identified through voting patterns, speech similarities, and electoral data. The findings suggest that stronger indirect connections to corruption lead to greater electoral losses, emphasizing the importance of dissociation as a strategic response to scandals.