Lesson 7 of 7
Process versus outcome
Judge a decision by the reasoning available when it was made, not by how the price moved afterward.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- A decision should be judged by the reasoning and information available at the time, not by how the price later moved.
- Judging a decision only by its outcome is called resulting, and it can teach the wrong lesson as easily as the right one.
- Over a handful of decisions, luck can look exactly like skill. Only a longer record separates them.
An outcome tells you what happened. It does not tell you whether the decision that produced it was good. A sound decision can lose money, and a careless one can make it, and the price cannot see the difference.
Process and outcome, judged separately
RowsProcessColumnsOutcome
Sound, Lost money
Bad luck, not a bad decision
Sound, Made money
The expected case
Careless, Lost money
The easy lesson to over-learn
Careless, Made money
A dangerous case
Resulting: judging the decision by the outcome alone
Duke calls this resulting: treating the outcome as proof of the quality of the decision, in either direction. A good outcome does not prove the reasoning was sound, and a bad one does not prove it was careless. The decision has to be judged on its own terms, using only what was knowable when it was made.
Why a handful of decisions is not enough
Mauboussin argues that separating skill from luck requires enough repetitions for the luck to average out. An investor who makes a handful of decisions a year has, after three years, perhaps a dozen outcomes. At that sample size a result is mostly noise.
The quadrant that teaches the wrong lesson fastest
A careless decision that made money is the most dangerous quadrant, because the gain looks like confirmation. Nothing about the outcome distinguishes it from a sound decision that also made money, and only a review of the process, not the price, can tell them apart.
Reviewing your own record
A decision journal is what makes this review possible. Without a dated record of the reasoning at the time, a later review is just another exercise in resulting. It grades the decision by the ending price instead of the reasoning that led there. Outcomes still count as evidence across many decisions: a long run of losses is a reason to re-examine the process, not only a single one.
Check yourself
Why is it hard to separate skill from luck over only a handful of investment decisions?
- Because outcomes are never recorded accurately
- Luck's effect on a small number of outcomes is large relative to skill's
- Because skill does not exist in investing
Show the answer
Answer: B.
Mauboussin's point is about sample size: with only a few decisions, luck has not had a chance to average out, so a small record cannot reliably show skill.What does Duke call judging a decision by its outcome alone?
- Anchoring
- Resulting
- Base rate neglect
Show the answer
Answer: B.
Resulting treats the outcome as proof of the decision's quality in either direction, when a good or bad result can happen regardless of how sound the process was.Why is a careless decision that made money the most dangerous quadrant to learn from?
- Because it is the rarest of the four outcomes
- Because it always loses money eventually
- The good outcome can be mistaken for proof that the careless process was fine
Show the answer
Answer: C.
The outcome looks identical to a sound decision that also succeeded. Only a review of the process, not the price, tells the two apart.Sources
- Duke, Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts, 2018.
- Mauboussin, The Success Equation: Untangling Skill and Luck in Business, Sports, and Investing, 2012.
What this leaves out
- The worked example uses two invented decisions to illustrate the point; it is not a claim about how often sound decisions lose money in practice.
- Separating skill from luck needs a real record over time. A single review, of one decision, cannot do it.
- Outcomes across many decisions are still evidence: a long losing streak is a reason to re-examine the process, not to dismiss it.
- Nothing here is advice about any security.