[jstx1]

Kahneman famously wrote in Thinking Fast and Slow that his results can't possibly be statistical flukes and then a bunch of the studies in the book ended up being exactly statistical flukes - https://slate.com/technology/2016/12/kahneman-and-tversky-re...

Anyway, about the books...

Blitzstein's Introduction to Probability and Harvard's Stat110 course are a good starting point if you've taken calculus. There's also good books like All of Statistics by Wasserman and Bayesian Data Analysis by Gelman but they're an absolute slog to get through - definitely not my favorite in any way but they cover a lot of stuff and you can have a copy around for reference.