60/30/10: the weights are the doctrine, written as arithmetic.

The ranking

Six times more weight on following the plan than on ending in a better gate.

The spread between the three weights is wide, and the width is the message. Compliance at sixty percent means that more than half the score is determined by whether the operator did what the plan said, before any question of how the capital state fared. Compression at thirty acknowledges that the severity of the capital squeeze is real information about the month. Gate improvement at ten is present but deliberately minor — it is the most outcome-adjacent of the three, since ending in a better gate depends partly on what the market delivered, and the system is consistent in giving outcome-adjacent inputs the least authority.

Why compliance dominates

It is the only input that is entirely the operator's.

Compression is a property of the drawdown path, which the operator influences through sizing and selection but does not control. Gate improvement depends on whether the account recovered, which depends substantially on the market. Compliance is the one input with no external component at all: the plan was followed or it was not, in conditions the operator did not choose but under rules they did. Giving it the dominant weight makes the score primarily a measure of the thing the operator can actually change, which is what makes an instruction to improve it a coherent instruction. A score dominated by market-dependent inputs would fluctuate for reasons no behaviour could address, and a metric nobody can move is a metric nobody attends to.

The consequence at the margin

A compliance failure cannot be offset by a good recovery.

The arithmetic has a practical edge worth understanding before it is discovered accidentally. Because compliance carries six times the weight of gate improvement, a month with meaningful plan breaches cannot be rescued by ending in a better capital state — the recovery is worth ten points at most and the breach can cost far more. This is intentional. The alternative, where a strong finish launders a badly conducted month, is exactly the accounting most traders already perform informally, and the score exists in part to refuse it. The corollary is also true and less obvious: a fully compliant month that ended in a worse gate still scores respectably, because the system judges that following the plan through a deteriorating period is the correct behaviour and should not be penalised for the deterioration.

• Sixty percent means compliance alone can carry a month to a passing composite, or sink one on its own.
• The ten percent on gate improvement is deliberately too small to buy absolution.
• This mirrors the wider law that outcome-dependent evidence ranks below behavioural evidence.

Why fixed and unweighted by regime

The weights do not move with conditions, for the same reason the score does not.

A natural refinement would be to shift the weights by context — to lean harder on compression during a stressed month, or on compliance during a quiet one. This is not done, and the reasoning is the same one that keeps difficulty out of the score entirely. Weights that move with conditions require a judgement about what conditions are, made by the party being measured. Fixed weights are cruder and produce a number that means the same thing in January and in a crisis, which is the only property that makes a twelve-month series of scores comparable. Refinement here would buy sensitivity at the cost of the comparability that gives the score its use.

The key idea

Composite weights are where a system says what it believes.

Any composite has to decide what counts more, and that decision is doctrine whether or not it is ever written in prose. 60/30/10 states plainly that MARS grades conduct above circumstance and controllable inputs above market-dependent ones. An operator who disagrees with those priorities does not disagree with a formula — they disagree with the system, which is a much more useful thing to discover.