Can the quality of a decision be evaluated before its outcome is known?
Which properties make a decision process more defensible under uncertainty, and how can we distinguish those properties from luck, hindsight and emotional reinforcement?
A profitable outcome does not prove a good decision.
When uncertainty exists, outcomes contain both decision quality and variance. Evaluating only the result encourages people to learn the wrong lesson from luck. TQS therefore separates the information available at decision time from information that became available afterward.
Separate concepts before measuring them.
Process ≠ Outcome
A good decision can produce a bad outcome. A bad decision can produce a good outcome. Outcome alone cannot identify process quality.
Information ≠ Relevance
More information does not automatically improve judgment. Evidence matters only in relation to the question being answered.
Confidence ≠ Evidence
Subjective certainty may rise faster than the quality of supporting evidence.
Prediction ≠ Preparation
Decision quality can improve by structuring scenarios and invalidation without pretending uncertainty disappears.
Where reasoning can become distorted.
Seeking or weighting information because it supports an existing thesis.
Allowing recent outcomes or price movement to dominate a broader evidence set.
Judging a past decision mainly by what happened afterward.
Treating equivalent losses and gains asymmetrically in decision-making.
Allowing already-spent resources to influence a decision that should depend on future costs and benefits.
Preferring action over inaction even when waiting may have higher expected value.
Every decision is also a decision not to do something else.
Capital, time, attention and risk capacity are scarce resources. A decision should therefore not only be compared with doing nothing, but with the best available alternative. This links behavioral reasoning directly to economic choice.
And is the expected value of this choice superior to the best alternative after accounting for uncertainty, risk and attention?
Turn the framework into structured reflection.
Decision Quality Engine
Evaluate the structure of a decision before judging its outcome. This is a reflection framework, not a psychological diagnosis and not a trading signal.
I defined my hypothesis before the latest market movement.
I can state specific evidence that would invalidate my thesis.
I actively looked for evidence against my current view.
I compared this decision with the best alternative use of capital, time or attention.
I can clearly separate what is known, inferred and uncertain.
I would make the same decision regardless of whether my previous outcome was a win or loss.
Mixed decision quality
Your weakest current dimension is Hypothesis Quality. The useful next question is not “Was the outcome profitable?” but “Was the process defensible before the outcome was known?”
A score is not a mind.
The interactive engine is currently a heuristic prototype. Self-reported answers can be biased, dimensions are not yet empirically weighted and the output is not a validated psychological scale. Its purpose is to make reasoning explicit, not to diagnose a person or certify a decision.
Future versions should test reliability, question wording, weighting, predictive usefulness and domain dependence before stronger claims are made.