Mindset
Resulting Fallacy
A good result can follow a poor decision, and a sound decision can end badly. Review the choice using the information that was available when you made it.
- Available information
- Decision
- Possible outcomes
- Review choice
One decision leads to several possible results, while decision quality is reviewed using the information available before those outcomes.
Winning can teach the wrong lesson
You rush an entry with poor position and somehow score because the opponent slips. Was the entry good? The score tells you what happened. It does not tell you that your reasoning or preparation was sound. Repeating the same choice against a different response could go very differently.
Annie Duke uses resulting for judging decision quality through outcome quality. The underlying outcome bias has a research history, including experiments by Jonathan Baron and John Hershey in 1988. Skill First applies the distinction to training reviews. The task is to separate a decision from the luck, opponent behaviour and execution that also influence its outcome.
Review the choice before revealing the ending
When watching a grappling exchange, pause before the result is clear. Ask what options the athlete could reasonably see at that moment. Did the chosen action fit the position, rules, available time and likely responses? What did it risk? Then watch the outcome and examine execution. This makes it harder for a lucky success to erase an obvious weakness.
Use the same approach after a failure. A well-timed attempt can meet an excellent defence. That does not automatically make the original choice foolish. It may show that the action needs better execution, another follow-up or a narrower set of conditions. Keep these possibilities separate rather than describing every loss as a bad decision.
Keep the result as evidence
Avoid the opposite mistake of protecting a favourite approach by saying the process was good whatever happened. Repeated outcomes are information. If the same choice repeatedly fails under conditions you thought favourable, something in your reasoning, implementation or assumptions needs attention. Separating decision and outcome lets you investigate that issue more accurately; it does not excuse ignoring it.
A coach and athlete can disagree about what information was available. Say what cue was visible and what remained uncertain. Video may help, though slow motion can make recognition look easier than it was in real time. Evaluate the actual decision environment instead of expecting knowledge that only appeared afterward.
Try a two-part session review
Choose one exchange and write two notes. The first covers the decision: the cue, options, purpose and expected risk. The second covers what happened: your movement, the partner's response and the result. Look for the smallest correction that follows from both notes. You might keep the choice and adjust execution, or change the conditions in which you select it.
Confirmation bias decides which evidence you favour. Resulting fallacy gives the outcome too much authority over your judgment. Probabilistic thinking allows several possible outcomes, while Bayesian updating changes confidence as evidence accumulates. Together they support a review that is neither harsh after every loss nor complacent after a lucky win. Discuss a good choice even when the score goes against you; that helps keep the appropriate action recognisable.
Sources and further reading
- Baron and Hershey: Outcome Bias in Decision Evaluation
1988 experiments on outcome information influencing judgments of decisions.
- Annie Duke: When mistaken beliefs lead to bad bets on data
Author account of resulting and the difference between decisions and outcomes.
See an idea alongside another
Related models
Probabilistic Thinking
Several outcomes can follow the same reasonable choice.
Bayesian Updating
Repeated evidence can change confidence without turning one result into proof.
Confirmation Bias
A preferred explanation can distort which outcomes you take seriously.
Investment in Loss
A temporary bad result may be part of a worthwhile learning decision.