Ignoring the importance of sample size in statistical reasoning. Small samples feel just as convincing as large ones.
Leads to false conclusions from limited data, explains why anecdotal evidence feels convincing, and causes poor generalisation from small samples.
People make the same confidence judgments about patterns in samples of 10 as samples of 1000, completely ignoring statistical reliability.
"Reminder: Every decision you make is influenced by biases you don't know you have. Sweet dreams!"