Mike Meier · Lessons Learned and Murphy's Corollary
Anything that did not go wrong could not have gone wrong.
Nobody believes Murphy's Law. But we are surprisingly susceptible to its corollary: we take a positive outcome as evidence that a positive outcome was inevitable.
Would you play?
Imagine that you need a significant amount of money, with no obvious way to get it, and with serious consequences if you don't. Someone offers you ten million dollars to play a single game of Russian roulette. One bullet, six chambers, one pull of the trigger. If the gun fires, you're dead. If not, you get the ten million.
Most people say no. Yet a go decision that is 99 percent safe, made again and again, carries the same odds sooner than you'd think.
Try the numbers
A go decision that succeeds % of the time, made times, fails at least once of the time: about one chance in 6.
The overall chance of success is the chance of each success raised to the power of the number of trials. At 99 percent, eighteen go decisions leave a one-in-six chance that at least one fails, the same odds as the revolver.
The framework
Two premises, three factors that work together against good decisions, and a remedy.
- Two premises. The primary determinant of safety is the quality of decision making. And when a negative outcome could be catastrophic, the go decision must be made only with a very high degree of confidence in a positive outcome.
- Factor I · Deceptive mathematics. High odds produce long strings of success, and each success seems to confirm the decision. But small risks, repeated, add up: 99 percent, eighteen times, is one chance in six of failure.
- Factor II · Murphy's Corollary. If nothing did go wrong, we come to believe nothing could have gone wrong. A positive outcome is taken as proof that the decision was sound, and standards slowly slip.
- Factor III · The pressure to go. Deadlines, budgets, expectations and ego push us to go when we have doubts. A go decision relieves the pressure; a no-go only increases it the next time.
- The remedy · Analyze your successes. Stay aware of these tendencies, set a quantitative safety metric and a target beforehand, and measure every success against it, to find the ones that were really luck.
Extension: Lucky or Good – Further Thoughts (2021) on skill, judgment and luck.