>If there is one thing the US election of 2020 showed us is that there is zero relationship between crowd/rally attendance and voter turnout for the candidate.

>If there is one thing the US election of 2020 showed us is that there is zero relationship between crowd/rally attendance and voter turnout for the candidate. I'm not sure if this is the strongest argument in favor of your hypothesis. In most instances it would rather be an indicator of election fraud.