Week 20: Ex Ante Vs Ex Post Analysis

Was it right to buy that lottery ticket? Well, it depends on whether you're talking ex ante or ex post.


What is ex ante and ex post thinking?

Ex ante means "before the event". When you’re making a prediction, you’re doing so ex ante. The opposite of ex ante is ex post, which means after the event. This is a useful framework because people often conflate the two in their reasoning. The ‘Expected Value’ entry made the claim that buying a lottery ticket was a bad idea, but I never specified the point at which you were deciding it was irrational. Buying a lottery ticket loses you money ex ante (in expectation), but if you win, it was the right decision ex post

This distinction means we should focus on making the right decision with the information we have available to us, not on making the perfect decision. 

Examples of ex ante and ex post thinking

If a fund manager succeeds in substantially outperforming the market, ex post they made the right decisions. However, they could have used a monkey throwing darts to pick their stocks, so ex ante they were borderline-negligent, and you probably shouldn’t trust them with your money.

Hindsight bias, also known as the ‘knew-it-all-along effect’ is the inclination, after an event has occurred, to see the event as having been predictable, despite there having been little or no objective basis for predicting it. This is why in science we make testable predictions to guard against our tendency to make up plausible-sounding but wrong post-hoc explanations.

As Philip Tetlock has shown in Superforecasting, we need to make a lot of predictions to track whether people were right ex ante. They could get lucky, and to our brain an event with a 40% and a 20% chance don’t seem that different, so we need to average over a lot of predictions (a Brier score) to figure out if someone was right.

Also check out

  1. Post-hoc, Wikipedia

Get one concept every week in your inbox