“The thing I have noticed is that when the anecdotes and the data disagree, the anecdotes are usually right. There is something wrong with the way that you are measuring it.”

— Jeff Bezos

Trusting your eyes over people with replication rates in the 20-50% range (ie, social scientists) is probably a rational choice.

There’s plenty of evidence that polls have systemic inaccuracies — question phrasing, people giving “right” answers, etc.