Football managers are too frequently fired when their teams underperform. That highlights a general bias to attribute credit / blame on the basis of realized outcomes. Because realized outcomes will to some extent be a function of luck, efficiency would recommend instead attributing credit and blame on the basis of process.

The football manager example is mirrored by a tendency in journalism: “But minister, now that x has happened, do you not owe it the electorate to resign?”

One result is that talent gets wasted – high quality decision makers get ditched because luck runs against them and low quality decision makers get promoted because they happened to catch a lucky break.

A more insidious result is that decisions become more conservative – high quality decision makers will shy away from that high-expected value long shot.

One way to overcome these negative results is to educate the public in expected value. That’s a big ask but it might be made easier if langauge made the ideas more intuitive.

How about distinguishing between a SAD outcome versus a BAD decision?

Statistically Anticipated Damage (SAD) is the price we knowlingly and willingly pay for walking, driving, having electricity in our homes, sending our children to school etc. etc. A SAD outcome will occur occasionally because lightning strikes, accidents occur and shit happens.

A BAD decision should not happen. It is Biased And Destructive. It speaks to poor process that leads the decision maker to choose a course of action that is strictly inferior to some alternative course of action.