Two typed numbers decide the gate, and one of them is remembered rather than observed.

The computation

A ratio, a band ladder, and a gate — with nothing discretionary in between.

The console takes current equity against the previous peak, expresses the shortfall as a percentage, and locates that percentage on a fixed ladder of drawdown bands to produce the gate. Shallower than roughly seven percent is the growth state with the full ladder available. Each successive band steps down through recovery, buffer, floor, deep-floor and ground-floor, and past forty percent the system halts rather than trading at minimum size. The arithmetic is deliberately unclever: one subtraction, one division, one lookup, no interpretation. That simplicity is a feature — it makes the gate reproducible and impossible to argue with — and it means the entire quality of the output rests on the quality of the two inputs, because there is nothing downstream capable of noticing that either was wrong.

Why the peak is the fragile one

Current equity can be checked against reality; the peak cannot.

The asymmetry between the two inputs is what makes one of them a risk. Current equity is verifiable in seconds against the account, and an error in it is likely to be noticed because the operator has a reasonable sense of what the account holds. The peak is a claim about the past — the highest equity ever reached — and there is no surface anywhere in the console that can confirm it. It is maintained by the operator, carried forward from cycle to cycle, and updated only when a new high is made. If it drifts, nothing objects. Every subsequent gate, tier ceiling and pool percentage is computed against it faithfully, and the whole deployment chain runs correctly on a reference point that no longer corresponds to anything that happened.

Both directions hurt

A peak set too high over-restricts; one set too low removes the brake entirely.

The two failure directions are not equally dangerous but both are real. A peak recorded above the true maximum — a transient intraday figure captured instead of a settled balance, or a peak never revised down after a deposit reversal — makes every measurement read as a deeper drawdown than the account has suffered. The account then operates one or two bands lower than it should, deploying conservatively for reasons nobody can see, which is a bounded cost paid indefinitely. A peak set below the true maximum is the serious one: real drawdown reads as shallow, the gate stays permissive, and the mechanism that is supposed to compress deployment as damage accumulates simply never engages. The account is then sized as though healthy at precisely the point the ladder was built to intervene.

Withdrawals

Capital movements shift the reference and the arithmetic cannot tell.

The case most likely to introduce drift is not carelessness but ordinary account activity. A withdrawal lowers current equity without any trading loss having occurred, so the computed drawdown deepens and the gate compresses on evidence of nothing. A deposit does the reverse, lifting equity toward or past a peak the trading never earned. The system has an established convention for this elsewhere — the milestone rail measures adjusted equity precisely so that transfers are not read as performance — and the same reasoning applies with more force here, because the gate governs deployment rather than merely reporting progress. Any capital movement is therefore an event that requires the peak reference to be reconsidered deliberately, and it is the one moment where leaving the field untouched is the active mistake.

The one check available

Confirm the gate the console produced, every cycle, as its own step.

The checklist ends step five with confirming the final gate used, and that confirmation is the only integrity check the chain has. It costs one glance: read the computed drawdown, locate it on the ladder, and confirm it produces the gate now sitting in the field. What it catches is a mismatch between the two, which is the visible symptom of a stale peak, a mistyped equity figure, or an undocumented override. What it cannot catch is a peak that is wrong but internally consistent — the arithmetic will be flawless. That case is caught only by periodically verifying the peak itself against the account's own history, which is worth doing after any capital movement and at each weekly setup rather than never.

• Current equity is verifiable; the peak is a maintained claim with no cross-check.
• Too low is worse than too high — the brake never engages when it is needed.
• Any deposit or withdrawal makes the peak reference a deliberate decision.

The key idea

The most authoritative input in a system is often the least instrumented.

The gate outranks every other layer in the deployment chain, cannot be promoted past by any amount of favourable evidence, and is computed from two numbers a person types. There is nothing wrong with that construction — the alternative is a wired calculation that would still depend on a correctly maintained peak somewhere — but it does mean the strongest link in the authority stack rests on the weakest kind of input in the workbook. Recognising which numbers carry that much weight, and treating them as maintenance rather than as entry, is most of what keeps a governed system governing the account it actually has.