When looking at stats, you need to consider sample size, and where the samples are coming from. If their pool was all republican, the data tells you something about Republicans. If their pool was all voters, it gives you information about all voters. Then there are other factors about socioeconomic stratification, gender, and race that could impact how you sample. Data is Data, it doesn’t give a shit about narratives. It’s up to the reader of the data to critically think about what they are looking at to consider how the data collection methods could skew the data. Like Republicans sampled from one county in the deep south might skew the data. But republicans sampled from all over the country may do a better job of removing factors that skew the data. So try thinking about what you read instead of having a knee jerk reaction.

“Why do people with advanced degrees in statistics understand stratification and statistical modeling better than i do?! I better make an angry post online about it!!1”