When we observe a success or a failure, we are observing one data point, a sample from under the bell curve that represents the potentialities that previously existed. We cannot know whether our single observation represents the mean or an outlier, an event to be on or a rare happening that is not likely to be reproduced.
This is another passage from The Drunkards Walk by Leonard Mlodinow. It highlights a common bias, especially in the public space; that we judge actions purely by their results.
We readily assign praise for success and blame for failure, despite not knowing the probabilities and tradeoffs, or how the decision was made.
Take basketball for instance. The NBA regular season is just about to start and so we can expect plenty of ooing and aahing. But a shot going in is not what makes it good. Just as a missed shot is not necessarily bad.
A good basketball shot is one that maximises the expected value – taking into account both the probability of scoring (the player’s skill, whether they are guarded etc.) and the value of the shot (one, two or three points). A good shot is one that you can take again and again, regardless of whether you miss one or even a sequence, leaving you ahead in the long run.
A good shot is unlikely to be the one that makes you gasp, or that you remember later. A high degree of difficulty isn’t what we are looking for. Hitting an off-balance shot with time running out should be the exception.