> Why didn't that sort of analysis find fraud in Philadelphia in 2014, 2015, 2016, 2017, and 2018? You can’t look at votes a local candidate for city office re

> Why didn't that sort of analysis find fraud in Philadelphia in 2014, 2015, 2016, 2017, and 2018? You can’t look at votes a local candidate for city office received in Massachusetts or Iowa because those candidates are only on the ballot in Philadelphia. The type of national vote trend analysis I’m describing only exists for national elections.