Feynman is warning against assigning special significance to an event simply because it appears improbable after it has already occurred. Once an outcome is known, it is easy to focus on its unusual features and calculate how unlikely that exact combination would have seemed beforehand. But many other equally specific outcomes could also have occurred.

The point concerns the difference between a prediction made in advance and a probability calculated after selecting an observed result. Scientific reasoning is stronger when the possible outcomes and the test for significance are defined before the evidence is examined. Otherwise, chance patterns can be mistaken for meaningful discoveries. Feynman’s remark therefore highlights a basic problem in statistical reasoning: surprising events are not automatically informative merely because they are rare in retrospect. Their significance depends on how the question was framed and whether the probability assessment was established independently of the observed outcome.

Full quotation

“I now turn to another kind of principle or idea, and that is that there is no sense in calculating the probability or the chance that something happens after it happens. A lot of scientists don’t even appreciate this.”