Operator brief · 178

Five conditions, all of them at once, or it is called something else.

The key idea

The structure of the claim

Conjunctive definitions are rare because they are expensive.

Most performance vocabulary is generous by construction: a strong month needs one strong number, and the rest is context. Defining alpha as the simultaneous satisfaction of five conditions inverts that, and the cost is real — the label becomes rare, and months that felt excellent will not qualify. That cost is the point. A term awarded whenever the headline is favourable stops carrying information within about two quarters, because it will have been applied to over-risked months, override-driven months, and lucky months alongside the genuine ones. Making it hard to earn is the only way to keep it worth earning.

FigureOne favourable month against the five conditions

Schematic: a month clearing the benchmark on return while failing two conditions. The headline is unchanged by the failures; the classification is entirely changed by them.

Failing on drawdown

Above the model with adverse risk is over-risk, and the playbook says so.

The scenario playbook's ruling is direct: live equity above median with drawdown worse than the adverse band is fast-but-risky performance, and the response is an audit of sizing, overrides, gate compliance, and open exposure rather than a celebration. What has happened in this case is that the account bought its outperformance with risk the simulation says it did not need to spend — the same return was available inside the modelled envelope, and the excess was consumed producing volatility. The renaming matters because the two states demand opposite responses. Alpha suggests continuing. Over-risk suggests finding out which authorisation was exceeded.

Failing on governance

Outperformance with heavy override use is leakage, not edge.

A month can clear the benchmark while its deployment record shows repeated use of the manual override that exceeds smart-capacity sizing. The override stress table exists specifically to keep this visible — it counts how often forcing full tier risk would exceed the authorised pool, and it exists, in the standard's own framing, to stop override from becoming a habit. Results produced this way are not evidence about the system, because the system was not what produced them. They are evidence about a series of exceptions, and the exceptions carry no expectancy the model can vouch for. The correct name is governance leakage, and its response is the deployment record rather than the performance record.

Failing on expectancy or quality

A favourable month with weak underlying expectancy is most likely variance.

Two of the conditions concern what sits beneath the result. If the period's measured expectancy is weak while the equity outcome is strong, the most probable explanation is that ordinary variance delivered a favourable draw — which is neither a problem nor an achievement, and specifically is not a reason to increase deployment. If profit quality is poor while conversion looks efficient, the return is leaning on outcomes the model does not expect to repeat at that frequency. Both cases produce the same practical instruction: record the month, do not extrapolate from it, and let the next period's evidence say whether anything actually changed.

Failing on acceleration

Growth velocity has to be sustainable to count, and sustainable has a definition.

The fifth condition asks whether the account's growth is tracking modelled velocity in a way that could continue. Acceleration achieved through gate-authorised deployment at earned tiers is sustainable by construction — the ladder would compress it if drawdown demanded. Acceleration achieved by sitting at high tiers more often than the model, without the structural confirmation that earns them, is a different phenomenon with the same shape on a chart. The tier usage comparison is what separates them, judged against the gate cap rather than raw frequency, and it is the reason this condition cannot be assessed from the equity curve at all.

The key idea

The definition is strict so the label can stay informative.

True alpha in this framework means the system did what it was designed to do, at the risk it was authorised to take, in the manner the model assumed, at a pace it could hold. That is a demanding conjunction and it will be satisfied in a minority of favourable months. The remainder are not failures — they are results with more accurate names, each pointing at something specific to check. A classification scheme that awarded alpha to all of them would be more pleasant to read and would tell an operator nothing about which of their good months were actually good.

Connected inside MARS

Classifying Deviation →

Live-Trade Benchmark Comparison →

Smart Open Exposure →

Every brief documents the same shipped system.

The complete MARS package — eleven workbooks, three TradingView indicators, the full manual library — $497.

Get MARS Pro — $497 Explore the Platform