You will be able to judge a decision separately from how it turned out.
Remember Wei Ling from Module 1, who mapped her manager Kumar's proposal for a new warehouse scanning system? On her advice, the company waited until after the year-end peak to install it. In January, Kumar hears that a competitor installed the same system in October and had its smoothest peak season ever. He forwards the news to Wei Ling with a single line: "Should have listened to the vendor."
Was waiting a bad decision? It turned out worse than the alternative, at least for one competitor. But that is a different question. This module is about telling the two apart, and it starts with the most common mistake people make when they look back.
Annie Duke, whose Thinking in Bets you met in lesson 7.1, Decisions are bets, borrows a term from poker players for this mistake: resulting. It means judging the quality of a decision by the quality of its outcome.
Duke opens her book with a famous example from American football. In the final seconds of the 2015 Super Bowl, the Seattle Seahawks' coach, Pete Carroll, called a passing play near the goal line instead of the expected running play. The pass was intercepted and Seattle lost. The next day the call was widely described as one of the worst in the game's history. Duke argues that the decision had sound reasons behind it and that the commentators were judging it by how it ended. If the pass had been caught, the same call would have been praised as clever.
Resulting is natural. The outcome is vivid and easy to see. The reasoning behind a decision, and the information available at the time, are invisible afterwards. So people grab the visible thing and treat it as the verdict.
Lesson 7.1 said that every decision is a bet with luck involved. That means outcomes and decision quality can come apart in both directions.
A sound decision can turn out badly through bad luck. You check a car carefully, get it inspected, pay a fair price, and the gearbox fails two months later. You hire the strongest candidate after a careful process, and she leaves after six months because her spouse is posted overseas.
A careless decision can turn out well through good luck. Someone drives home after several drinks and arrives safely. A colleague puts most of his savings into a single stock a friend tipped and it doubles. Nobody would call either of those a good decision, but the outcome alone would say they were.
It helps to think of four boxes: good decision and good outcome, good decision and bad outcome, bad decision and good outcome, bad decision and bad outcome. Resulting collapses these into two. It calls everything with a good outcome a good decision, and everything with a bad outcome a bad one.
If you judge by outcome alone, you learn the wrong things from both kinds of luck.
After a good decision with a bad outcome, you may abandon a sound process. Wei Ling might conclude that she was too cautious and that next time she should go along with vendor timelines. But the competitor's smooth peak could have been luck, too. Another company that installed in October might have had a disastrous peak, and nobody would have forwarded that article. Only the success story reached Kumar, which is the survivorship problem from lesson 3.4, Survivorship bias and missing data.
After a bad decision with a good outcome, you may repeat a careless process. The colleague whose tipped stock doubled may do it again with more money. The lesson he took from his result is the opposite of the one his process deserved.
Teams and companies fall into this too. A project that succeeds gets its methods copied, even if it succeeded because of a lucky market. A manager whose risky call worked gets promoted and makes more risky calls. Over time, resulting rewards luck and punishes caution, and the organisation slowly learns to gamble.
The fix is to grade a decision on its process, and to judge that process only by what was known, or could have been known, when the decision was made. Three questions do most of the work.
What did I know at the time? Write down the information that was actually in front of you then, and leave out anything you learned afterwards. Wei Ling knew the vendor offered no evidence of smooth installations near a peak, and that their own error data came from a single busy month.
What could I have known? Was there information that was available at a reasonable cost and that you did not get? This is where real mistakes often hide. Perhaps Wei Ling could have asked the vendor for a list of clients who had installed just before a peak.
How did I reason? Given the information, was the reasoning sound? Did you consider the options, the probabilities and the worst case? Did you fall into any of the traps in your bias checklist from lesson 5.4?
If the answers show a good process, a bad outcome is mostly bad luck and the right response is to keep the process. If they show a gap, fix the gap, whatever the outcome. Wei Ling's answer to Kumar is short: "Our decision was based on the risk of a failed install at peak. They took the risk and it worked. I'd still make the same call with what we knew, but next time I'll ask the vendor for installs near peak so we have better data."
You will need one decision that turned out badly and one that turned out well for the activity below. Pick ones where you can still remember what you knew at the time.
Pick one past decision that turned out badly and one that turned out well, and grade the decision process for each separately from the result.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).