The matrix exists to reduce hesitation. A version that produced more of it would be a worse tool with better outputs.

The three questions

The matrix answers exactly three, and the discipline is in what it declines to answer.

What is the live reading in practical pips or points. Is that low or high relative to the right baseline. What does each coefficient translate to in actual stop distance. Three questions, each with an unambiguous answer, each of which an operator would otherwise have to estimate. Everything the workbook could additionally compute — and there is a great deal — has been left out or left downstream, because each additional field arrives at the same moment as the three that matter and competes with them for the same attention.

Why hesitation is the metric

The decision this tool sits inside is time-boxed, and an unanswered question stops the operator.

The pre-execution moment is short. A setup has qualified, the gate has authorised, and the remaining question is what distance to use. If that question is answered in seconds the operator proceeds; if it is answered by five fields requiring reconciliation, the operator either delays or picks one and ignores the others. Both outcomes are worse than a clean answer, and the second is worse than it looks because the ignored fields do not vanish — they leave a residue of having not fully considered something, which is precisely the hesitation the tool was built to remove.

Where the extra rigour went

Nothing was discarded; it was moved to a moment that has room for it.

The three-question discipline is not an argument against depth, and the system contains a great deal of volatility depth — structural classification, static-versus-live comparison, coefficient audit trails, band migration tracking. All of it lives in the panel and the calibration review rather than in the execution-moment matrix. The separation is by time budget rather than by importance: analysis that deserves twenty minutes gets a surface where twenty minutes are available, and the surface consulted between setup and entry carries only what can be read in seconds without being reconciled against anything.

• Depth belongs to the review cadence; the execution moment gets three answers.
• A field's value depends on the time budget of the moment it is delivered into.
• Moving rigour to a later surface is not removing it.

The specific failure

A precise number in a moment of uncertainty will be used, whether or not it was addressed to the question.

The mechanism worth naming is that numbers are attractive in proportion to how uncomfortable the surrounding judgement feels. An operator unsure whether a setup is strong enough will reach for whatever quantitative field is nearest, and if the matrix has helpfully printed an intensity score at that moment, the intensity score becomes an input to a structural decision it has nothing to say about. Restricting output is a defence against that substitution, and it works better than instructing the operator not to make it, because the instruction has to be recalled and the absence does not.

The test for adding anything

Does it change what the operator does at the moment it appears, or does it merely inform them.

The rule that keeps this stable through future revisions is a question rather than a list. Any candidate addition to the execution-moment surface has to name the decision it changes and the moment it changes it. A field that informs without changing a pending decision belongs on the review surface. A field that changes a decision that should be made on other grounds belongs nowhere. That test is strict enough to have kept the matrix at three questions through several revisions of the wider volatility stack, which is the evidence that it is being applied rather than merely stated.

The key idea

Restraint at the point of use is a feature that only shows up as an absence.

A tool judged on its outputs will always seem improvable by adding one, and the cost of the addition lands somewhere the tool's own metrics never look — in the operator's attention at the moment of decision. The matrix's design accepts that trade explicitly: fewer fields, delivered faster, into a moment that cannot absorb more. Nobody ever admires a spreadsheet for what it declined to compute, which is exactly why the restraint has to be written down as doctrine rather than left as taste.