Do mass mobilizations drive social change? This paper explores this question by studying how the Black Lives Matter protests following George Floyd’s death influenced the 2020 U.S. presidential election. Using rainfall as an instrument for protest participation and complementary difference-in-differences analyses, we show that protest activity significantly increased Democratic vote share in affected coun- ties. Our research makes three key contributions. First, we show causal evidence for the effect of one of the largest protest movements ever recorded on electoral out- comes. Second, we provide evidence of novel temporal dynamics: while protests ini- tially triggered a conservative backlash, they ultimately generated progressive shifts in voting behavior. Third, we identify mechanisms driving these effects, showing that rather than merely mobilizing existing Democratic voters, protests substantively shifted political preferences and beliefs about racial inequality.