Outcome bias (judging decisions by results)

Outcome bias is the tendency to judge the quality of a decision by its result rather than by the information and process available when it was made. In trading, it leads to reinforcing rule breaks that happened to win and abandoning sound rules after trades that happened to lose.

Senzoukria · Glossary · Updated September 2026


The idea

Jonathan Baron and John Hershey described outcome bias in 1988: people rated the same medical decision as better when the patient recovered than when the patient died, even though the decision and its information were identical. Trading is an ideal environment for the bias, because every decision has an immediate, visible result, and the result of a single trade is dominated by chance.

A trade that followed the plan and lost 1R was a good decision with a bad outcome. A trade taken in anger, without a stop, that won 3R was a bad decision with a good outcome. Judging them by result reverses the lessons.

Four combinations

The dangerous cell is the bad decision with a good outcome, because it rewards the rule break and makes it more likely to happen again, usually at a worse moment.

Decision quality versus outcome
Good outcomeBad outcome
Good decision (plan followed)Deserved winBad luck: keep the rule
Bad decision (plan broken)Dumb luck: the dangerous caseDeserved loss

Process-based review

  • Grade each trade on criteria met, entry at plan, stop respected and size correct, before looking at its P&L.
  • Keep the grade and the result as separate fields, and compare them over many trades.
  • Evaluate rules on samples, not on the last trade: a rule is changed when its statistics say so.
  • Review a winning rule break as seriously as a losing one.

In Senzoukria

The journal trade form has a Rating field, one to five stars, separate from the P&L, and a Setup tag that says which playbook entry the trade belongs to; together they allow grading the decision independently of the result. The 'Review in Replay' button opens the market around a journal trade so that what was visible at the time can be compared with what happened next. In Replay reports, the note under 20 trades states that the sample is far too small to conclude anything about an edge, a reminder that results over a few trades are mostly outcome.

In the same section

This page in other languages

Frequently asked questions

How do I tell a bad decision from bad luck?
Check the decision against the rules that applied at the time: criteria, entry, stop, size. If all were respected, a loss is part of the strategy's normal distribution. If some were broken, the trade was a poor decision whatever it returned.
Does outcome bias affect backtesting too?
Yes. Keeping a rule because it produced a good equity curve on one sample, or dropping one after a bad month, judges by outcome. Out-of-sample tests and statistics that account for sample size are the research equivalent of process review.

Keep reading