Human beings, despite being good at a lot of stuff, are very bad at probability. A famous demonstration of this is Amos and Daniel’s 1983 study often called the “Linda question.” Imagine there’s a woman named Linda with a degree in philosophy who is concerned with social justice. Is it more likely that Linda is a bankteller or that she is a bankteller and active in the feminist movement?
It turns out most people say the latter, although it is strictly more likely that Linda is just one of those things rather than both. Our sense of probability is essentially vibes-based, and although this worked well for us evolutionarily speaking, it can lead to some confusion is modern life. We hardly even agree on the basic words we use to indicate probabilities.
When someone says “probably,” they could be anywhere from 45% to 90% confident! Clearly we need more precise terminology.