The shared language of the rail system.

also called EPH · peak equity · high-water mark

The highest realized account equity ever reached, used as the anchor from which drawdown is measured.

It is a ratchet: it rises when equity makes a new high and never falls. Anchoring to the peak rather than to a period start is what makes the gate ladder a statement about capital condition rather than about the calendar — a 12% drawdown is 12% whether it happened this month or across a quarter.

Owned by Cycle Command Console → Throttle Control Panel

Current equity expressed as a percentage below Equity Peak High. The single input that routes the gate.

(Current equity − Equity Peak High) ÷ Equity Peak High

Note the sign convention: drawdown is negative and deeper drawdown is a more negative number. The SDE inverts drawdown-drift semantics for exactly this reason — worsening drawdown is bad, recovery is good, which is the opposite of how a rising number usually reads.

Owned by Gate Map

also called gate state · capital state · drawdown gate

A capital-state band derived from drawdown depth. It caps the maximum tier permitted and nothing else.

The ladder runs Growth · Recovery · Buffer · Floor · Deep-Floor · Ground-Floor · System Lock, deepening as drawdown deepens. A gate is a ceiling, not an instruction: sitting in Growth permits a high tier, it does not require one.

Owned by Risk Gate Map

also called brake state · brake tier · brake tone

A triggered restriction on further trading — daily or weekly — associated with a gate's behavioral risk mode.

The manuals describe brake tones running from controlled exposure and early defense, through mild contraction, to survival-mode compression and terminal slowdown at minimum exposure. Unlike a gate, a brake is an action taken, and the response should be predefined rather than improvised at the floor.

Owned by Gate/Brake State · Cycle Command Console

also called the lock · shutdown boundary · terminal gate

The terminal boundary. In the standard configuration it triggers at drawdown worse than −40% from Equity Peak High; locked paths stop trading.

In simulation, locked paths are excluded from target completions — which is why a fast median hit week means nothing read on its own. ⚑ The boundary is a configured threshold, not a market fact: move it deeper and the reported lock rate falls immediately without the account having become any safer.

Owned by Gate ladder · Monte Carlo Lab

also called risk tier · final tier · T1–T7

The deployment-intensity level, T1 through T7, selected inside the ceiling the current gate permits.

Growth caps at T7 and Recovery at T6, with lower gates capping further. T6 and T7 remain rare by design — the top of the envelope is a boundary, not a target. Every layer in the stack may lower the tier; only some layers are licensed to raise it.

Owned by Seven-Tier Risk Model · Throttle Control Panel

also called authorized cycle pool · pool % · risk pool

The total percentage of capital authorized for deployment across one cycle, set by the gate-and-tier pairing.

Pool geometry is defined per gate by tier: a Growth gate runs a full growth envelope while Floor holds a fixed minimum survival exposure. The pool is the cycle's entire budget, and the per-trade average is what remains after dividing it across the cycle's slots.

Owned by Throttle Control Panel

One four-trade concurrent operating block — the primary unit of deployment inside the MARS weekly rhythm.

The cycle, not the trade, is what gets sized. The fourth slot is not optional filler: the budget is designed to be spent across four, and a cycle run as two large trades has departed from the profile even if the total risk matches.

Owned by Cycle Command Console

also called open exposure adjustment · smart capacity

The adjustment layer that accounts for risk already live at the stop when computing how much fresh deployment remains.

Without it, an authorized pool would be spent twice: once on positions already open and again on new ones. The layer converts authorized pool minus open risk into effective fresh trade count and fresh per-trade risk. Raise open risk and the fresh count drops with nothing else having moved.

Owned by Throttle Control Panel · Cycle Command Console

Capital still exposed at the active stop across all open positions.

Sum of (entry − current stop) × size, across open trades

Owned by Cycle Command Console

A documented exception permitting full-tier deployment outside what the standing state would allow.

Overrides are by exception and by record. The doctrine's value is not that it permits deviation but that it forces the deviation to be written down with its reasoning, which converts an untraceable judgement call into an auditable event.

Owned by Throttle Control Panel · Cycle Decision Log

The authorized average risk for a single slot, derived by dividing the cycle pool across the cycle's trades.

Owned by Throttle Control Panel

How many new positions may still be opened this cycle once open risk has been accounted for.

The number that actually governs the next order. It falls as open risk rises even when the gate, tier and pool are unchanged — which is why a directive can tighten without anything appearing to have happened.

Owned by Cycle Command Console

also called growth gate · expansion zone · full compounding zone

The gate occupied when drawdown from the equity peak high is shallower than -7%. The expansion and full-compounding zone.

The top of the gate ladder, and the only state carrying no brake tone. Growth permits the highest tier the rest of the engine can justify — it does not require one. A gate is a ceiling, never an instruction.

Owned by Risk Gate Map

also called recovery gate · stability brake

The gate occupied at roughly -7% to -13% drawdown from the equity peak high. Its brake tone is stability and mild contraction.

The first braked state. The account is still deploying meaningfully; the system has simply stopped treating conditions as expansionary. Recovery is reached by descending from Growth and is left by ascending back toward the peak — gates move in both directions as equity moves.

Owned by Risk Gate Map

also called buffer gate · defense brake

The gate occupied at roughly -13% to -19% drawdown from the equity peak high. Its brake tone is defense and controlled exposure.

The doctrine line most often quoted about this gate is that Buffer blocks casual aggression: aggressive variants stop being a default option and start needing a reason. The pool-budget row compresses at the same time, so the same tier number authorises less capital here than it does in Growth.

Owned by Risk Gate Map

also called floor gate · containment brake

The gate occupied at roughly -19% to -27% drawdown from the equity peak high. Its brake tone is containment and capital preservation.

Floor and below block aggressive variants entirely — this is the compression event rather than a mild downgrade. The name is the point: the system is now defending the account rather than growing it.

Owned by Risk Gate Map

also called deep floor · deep-floor gate · suppression brake

The gate occupied at roughly -27% to -34% drawdown from the equity peak high. Its brake tone is suppression and survival mode.

Deployment is compressed to the low end of the ladder. The operating question at this depth is no longer which tier is justified but whether the account survives long enough to use one.

Owned by Risk Gate Map

also called ground floor · ground-floor gate · seizure brake

The gate occupied at roughly -34% to -40% drawdown from the equity peak high. Its brake tone is seizure and terminal slowdown.

The last state before the system stops authorising fresh deployment altogether. Minimum exposure only.

Owned by Risk Gate Map

also called pool budget · gate row · pool envelope

The row of the pool-budget panel belonging to the active gate, from which the selected tier reads its cycle pool percentage.

This is why the same tier means different capital in different gates. A tier number is a column and the gate is a row; the authorised cycle pool is the cell where they meet. The gate compresses the pool, the gate cap limits which columns are reachable, and the tier engine chooses among what remains.

Owned by Throttle Control Panel

also called gate number · gate position

The gate's position on the ladder expressed as a number, so that gate state can be averaged or trended over a period.

A label cannot be averaged; a rank can. Ranking the ladder is what allows a month to be described as having spent most of its time in one capital state rather than another.

Owned by Risk Gate Map

also called carryover risk · carryover trades · open carryover

Risk already committed by trades still open when a new cycle begins.

Carryover is the reason fresh deployment is not simply the authorised pool. Open trades have already consumed capacity, and the smart-exposure layer subtracts what they hold before sizing anything new. Treating the pool as fully available while carryover is live is how an account ends up double-deployed.

Owned by Throttle Control Panel

also called compressed · risk compression

The narrowing of authorised risk as capital state deteriorates or as open exposure consumes capacity.

Compression arrives from two directions at once and they are not the same mechanism. The gate compresses the pool row as drawdown deepens; open exposure compresses what remains of that pool as trades stay live. Both reduce fresh risk, and the system applies them in sequence rather than choosing between them.

Owned by Throttle Control Panel

also called risk unit · initial risk

One unit of initial risk on a trade. Every outcome in MARS is denominated in R rather than in currency.

R normalizes across account sizes, position sizes and instruments, which is what allows a trade from a $5,000 account and a $50,000 account to appear in the same expectancy calculation. A +2R trade returned twice what it risked, whatever that was in dollars.

A management structure a trade is run under: Normal, Trend Partial, Trend No-Partial, or Overflow.

Normal is the core branch and the simulation's centre — the highest-sample, highest-weight structure. Trend Partial is the hybrid fat-tail accelerator, Trend No-Partial the true one with geometric growth and high variance, and Overflow is the stability branch. Every trade is classified into exactly one.

Owned by Branch Logic

A specific, explicitly-stated parameterisation of a branch, with its management rules fully written down.

Variants are what the Branch Variant Selection Matrix compares. A comparison is only valid if seed, trade sample, period and path count are held identical between candidates — otherwise the measured gap contains the effect of the variant and the effect of the drift, inseparably.

Owned by Branch Variant Selection Matrix

also called blended EV · All Blended

The weighted combination of branch expectancies, using the active branch weights, into one system-level figure.

SUMPRODUCT(branch EVs, active weights)

The blend is the headline expectancy number and the most misread figure in the system: a healthy blend can contain a branch that is quietly failing, carried by a strong neighbour. Read the branch layer before the blend, always.

Owned by EV Scorecard · Weekly Trading Scorecard

also called EV · expected value

Expected return per trade in R, computed from checkpoint probabilities and payoff structure rather than from an average of outcomes.

Normal branch: EV = p1 × (0.5 + p2) − (1 − p1)

Building EV from checkpoint probabilities rather than averaging realized results is deliberate: it makes the estimate auditable — you can see which probability is carrying it — and it degrades honestly on small samples instead of producing a confident number from four trades.

Owned by EV Scorecard · Advanced EV Analytics Lab

also called p1 · p2 · hit probability

The probability of a trade reaching a defined milestone — P(1R), or P(0.75R) and the conditional conversion beyond it.

These are the inputs an operator actually enters. Everything downstream is derived, which is why a probability typed as 60 instead of 0.60 produces an impossibly large EV rather than a subtly wrong one — the manuals list that exact symptom.

Owned by Weekly Trading Scorecard

A governance minimum on branch usage, ensuring the fat-tail structures are actually run rather than quietly abandoned.

Quotas exist because the accelerator branches are psychologically expensive to run: they give profit back repeatedly in exchange for occasional large outcomes. Left to preference, an operator drifts toward the comfortable branch and the system stops being the system that was simulated.

Owned by Weekly Trading Scorecard · Variant Usage Audit

The realized R across all closed trades in the selected cycle.

Sum of Final Net R across closed trades in the cycle

Owned by Cycle Command Console

The full decision snapshot for a cycle, including any override and the reason recorded at the time.

Its value compounds. Written before outcomes are known, it is the only genuine record of what the operator did under pressure — worth considerably more than any recollection formed afterwards, when the outcome is available to reshape the memory.

Owned by Cycle Command Console

CP3's period rollup — the engine that converts the week's journal into graded status.

Computed last and therefore read last. The week badge summarises the layers beneath it, and reading the badge first is how a quietly failing branch stays invisible inside a green week.

Owned by Compliance Panel 3

The Saturday pass transferring captured trade evidence into CP3 and confirming nothing is missing or duplicated.

Reconcile before you review, or the review audits itself. Any diagnostic run against an unreconciled period is describing a record that has not finished being defined.

Owned by CP3 Journal

also called capture sheet · paper capture

The paper record filled immediately after execution, holding MAE, MFE, pips risked, duration, fees, swap and notes until CP3 entry happens.

It exists because CP3 is entered weekly and memory is not a data source. Excursion values in particular cannot be reconstructed days later — they are captured while the trade is fresh or they are lost.

Owned by Trade Capture Document

also called normal · normal trades · static branch

The branch whose management path banks a partial at 1R and takes the remainder to a static 2R target, with no trail.

One of the four branches — Normal, Trend Partial, Trend No-Partial and Overflow — selected before entry and never retro-fitted once a trade is moving. Normal is the structurally static one: it has a fixed destination rather than a runner, which is what makes it the baseline the trending branches are measured against.

Owned by Normal Branch

also called trend partial · partial branch · TP

The branch that banks a partial at 1R and then trails the remainder once the trade unlocks past roughly 1.6R.

The hybrid structure: it takes the psychological cushion of an early partial AND keeps a runner exposed to continuation. That first monetisation event is what makes the branch stable to operate — which is also why delaying it is the most consequential change any variant can make to it.

Owned by Trend Partial Branch

also called trend no partial · no-partial · TNP

The branch that takes no partial at all, holding full exposure and trailing after the unlock.

Structurally the most aggressive of the four before any variant is applied, because nothing is banked on the way up. The manuals are explicit that it must be earned rather than chosen by mood: gate permission, branch EV support and trend context all have to agree, and no-partial quota pressure is a known way for the branch to get used emotionally.

Owned by Trend No-Partial Branch

also called overflow · catch-all branch

The catch-all branch for trades that do not belong to the other three structures.

Overflow exists so that the other three stay clean. A trade forced into Normal or a trend branch it does not fit corrupts the evidence for that branch, and branch expectancy is what the whole governance chain reads from — so the correct home for an exception is a branch that expects exceptions. It carries its own row in the rolling condition tables and its own attribution, exactly like the others.

Owned by Branch Logic

also called standard · baseline variant · v1

The baseline management variant — the branch's original exit structure, used when nothing justifies a defensive or aggressive modification.

One of five variants: Time-Conservative, Exposure-Conservative, Standard, Exposure-Aggressive and Time-Aggressive, ordered least to most aggressive. The two axes are separate — Time governs when protection starts, Exposure governs how much size stays uncapped. Standard is the benchmark the other four are judged against.

Owned by Variant Attribution Module

also called variant choice · selecting a variant

Choosing which of the five management variants a branch will be traded with, before entry.

A management-selection decision, not a risk-sizing one. It never overrides gate, throttle, trading-plan rules or system lock. The governing asymmetry is that aggressive variants must be harder to qualify than conservative ones — an aggressive variant that becomes a feeling-based override raises variance without raising expectancy.

Owned by Variant Attribution Module

also called promoted · tier promotion

A move up the tier or variant ladder, authorised by evidence rather than by confidence.

Promotion is the direction the system makes hardest. Every path upward requires the reading that justifies it to already exist in the record; nothing is promoted on the expectation that it will be earned afterwards.

Owned by Throttle Control Panel

also called demoted · tier demotion · downgrade

A move down the tier or variant ladder, applied as soon as a reading deteriorates.

Demotion needs no confirmation period and no second opinion. A cap is a cap the moment the condition producing it appears.

Owned by Throttle Control Panel

also called maximum adverse excursion · heat

Maximum adverse excursion — the furthest a trade moved against you, in R.

MAE is how much adverse pressure a setup routinely absorbs. Reducing it should raise expectancy by cutting stop pressure and variance. An acceptable average MAE can still conceal a cluster of trades sitting in the worst quadrant, so the distribution matters more than the mean.

Owned by MAE/MFE Execution Efficiency Lab

also called maximum favorable excursion

Maximum favorable excursion — the furthest a trade moved in your favour, in R.

MFE measures whether the market offered opportunity at all. Low MFE with clean entries is a market problem rather than a management one, which is precisely the distinction the lab exists to draw.

Owned by MAE/MFE Execution Efficiency Lab

The share of available favourable movement that was actually realized.

Realized positive outcome ÷ MFE

High capture with high MFE is clean alpha — entry, opportunity and management aligned. ⚑ Capture cannot be maximised: driving it toward 1.0 means exiting at the extreme every time, which is unachievable and would destroy the runner structures the fat-tail branches depend on.

Owned by MAE/MFE Execution Efficiency Lab

Favourable movement that was reached and then surrendered before the exit.

MFE − realized outcome, floored at zero

Giveback is not automatically a fault. Runner structures give back by design — that is the price paid for the occasional outsized outcome. It becomes a finding when it appears on branches whose rules do not call for it.

Owned by MAE/MFE Execution Efficiency Lab

also called friction · fee drag

Total trading friction expressed as a fraction of the initial risk dollars.

Total fees ÷ initial risk dollars

Denominating friction in R rather than currency is what makes it comparable across trade sizes. Eleven dollars of spread and commission means nothing on its own; the same eleven dollars against a small initial risk can be a material share of the edge.

Owned by CP3 Journal · MAE/MFE Lab

The trade's result after friction is deducted.

Outcome R − Fee R drag

The distance between gross and net is itself a reading. A system whose gross expectancy holds while its net expectancy decays has a friction problem, not an edge problem, and the two require completely different repairs.

Owned by CP3 Journal

Favourable excursion divided by adverse excursion — opportunity against pressure.

MFE ÷ MAE

Owned by MAE/MFE Execution Efficiency Lab

A composite grade combining MAE behaviour and capture efficiency into one execution-quality reading.

Composite scores are convenient and lossy. The score is a prompt to investigate, never a diagnosis — it names that something is worth looking at and is structurally incapable of naming the cause.

Owned by MAE/MFE Execution Efficiency Lab

The trade's realized result in R, before friction is deducted.

Owned by CP3 Journal

Whether the trade was managed the way its branch rules specify, recorded independently of whether it made money.

The most uncomfortable field in the journal, because a profitable breach is the expensive one — it teaches the operator that deviation works. Adherence is recorded against the rules, never against the outcome.

Owned by MAE/MFE Lab · CP3 Journal

Overnight financing charged or credited on a position held past the rollover point.

Tracked separately from spread and commission because it scales with holding time rather than trade count — which means it penalises exactly the runner structures the fat-tail branches depend on, and belongs in the friction reading rather than being dismissed as small.

Owned by CP3 Journal · MAE/MFE Lab

also called execution friction · drag

The gap between the R a trade theoretically offered and the R actually realised, once execution and cost are accounted for.

Friction is where an edge quietly disappears without any losing trade appearing. Fee drag, swap, slippage, late entries and early exits each subtract a fraction of an R; individually invisible, cumulatively decisive.

Owned by MAE/MFE Execution Efficiency Lab

also called aligned · alignment check · authority alignment

Agreement between the readings that must concur before an aggressive path is authorised.

Alignment is what the system requires instead of conviction. When branch EV, gate permission, structural diagnostics and volatility condition all point the same way, an aggressive variant is earned. When they disagree, the disagreement itself is the answer, and the conservative path is the correct one.

Owned by Structural Diagnostic Engine

also called percentile fan · the bands · equity bands

A range within the simulated distribution describing where a given share of paths landed.

The standard reading protocol: P50–P75 is above the centerline and healthy with no need to force aggression. P25–P50 is below median but still normal — monitor, check execution, fees, branch mix and sample size, and avoid panic. P5–P10 is danger-zone underperformance warranting formal diagnostic review. Below P5 is outside normal behaviour and is treated as serious.

Owned by Monte Carlo Lab · Dynamic 7-Tier MC Benchmark

The share of simulated paths that reach the target within the horizon.

Owned by Dynamic 7-Tier MC Benchmark

also called lock probability · risk of ruin

The share of simulated paths that reach the System Lock boundary and stop trading.

⚑ Conditional on where the boundary is set. Deepen the threshold and the rate falls instantly with no change to the strategy, the sizing or the discipline — the paths formerly counted as failures become survivors carrying very deep drawdowns. A lock rate quoted without its threshold is unreadable.

Owned by Monte Carlo Lab

How quickly successful paths reached the target, expressed in weeks by percentile.

⚑ The single most misread number in the suite. Hit-week percentiles are calculated among successful paths only — failed and locked paths are excluded. A fast median hit week does not mean the system is safe; it must be read beside completion probability and lock rate or it becomes a highlight reel.

Owned by Dynamic 7-Tier MC Benchmark

The unfavourable tail of the drawdown distribution — typically the 90th or 95th percentile of maximum drawdown.

A live account can sit above median equity and still be structurally unhealthy if its maximum drawdown is worse than the adverse benchmark band. Equity placement and drawdown placement are separate readings and can disagree.

Owned by Drawdown Percentiles

The value fixing the random draw a simulation run uses, making the run reproducible.

Two variants tested on two different seeds are two anecdotes: part of the gap between them is the difference in the draws rather than in the structures. Same seed, and whatever separates them is attributable to management. It is the cheapest control in the exercise and the one most often skipped, because skipping it produces no error message.

Owned by Monte Carlo Lab · run log

How much of an outcome is attributable to the order in which trades arrived rather than to their composition.

The same set of trades in a different order can produce a comfortable path or a locked one, because drawdown depth depends on clustering. This is why a single backtest cannot answer a survivability question — it is one ordering out of an enormous number.

Owned by Monte Carlo Lab

One simulated account history — a single ordering of outcomes run through the full rule set to the horizon.

The standard run is 50,000 paths. A backtest is one path; the simulation's whole contribution is showing the other 49,999 you did not happen to live through.

Owned by Monte Carlo Lab

The time window a simulation runs before stopping and scoring the result.

Completion probability is always conditional on it — reaching target means reaching target within the horizon. A path still alive and short of target at the end is neither a completion nor a lock, which is why those two figures do not sum to one.

Owned by Dynamic 7-Tier MC Benchmark

The share of paths that neither reached the target nor locked — alive at the horizon, short of goal.

The forgotten third of the scoreboard, and often the largest. Reading only completion and lock leaves this population invisible, which makes a system look more decisive than it is.

Owned by Dynamic 7-Tier MC Benchmark

also called p50 · 50th percentile · middle path

The middle outcome of a simulated distribution — half of all paths finish above it, half below.

The median is the honest centre of a Monte Carlo run in a way the mean is not: a small number of extreme paths drag an average and leave it describing no outcome anybody is likely to experience. When the system quotes a typical result, it quotes this.

Owned by Monte Carlo Lab

also called 90th percentile · p-90 · upper percentile

The 90th-percentile outcome of a simulated distribution — only one run in ten finishes better.

P90 is the upper reference point, and it exists to be read as rare rather than as a target. Planning against it is the most common way a simulation gets misused: it describes the good tail, not the expected case.

Owned by Monte Carlo Lab

also called 10th percentile · p-10 · lower percentile

The 10th-percentile outcome of a simulated distribution — nine runs in ten finish better.

The lower reference point, and the one worth planning against. If the account cannot survive the P10 path, the sizing is wrong regardless of how good the median looks.

Owned by Monte Carlo Lab

Average True Range — the raw measure of how far an instrument typically moves per bar on a given timeframe.

ATR on its own authorizes nothing. It becomes a stop distance only after a coefficient is applied, and the coefficient is where the intelligence sits. Sourced from the chart or the ATR BE Assistant on the final authority timeframe.

Owned by Volatility Intelligence Panel · ATR BE Assistant

The reference ATR for an instrument and timeframe — typically a 20-session median or average, or a sampled Distance Matrix baseline.

The baseline is what makes a live reading mean something. The same live ATR is calm against one session baseline and extreme against another, which is why the session lens compares All Sessions, London, New York and Asian baselines separately.

Owned by Volatility Distance Matrix

also called ATR coefficient · trail coefficient

The multiplier applied to ATR to produce an actual stop or trail distance.

Derived from the panel's evidence score mapped to an asset-aware tier ladder at the trigger timeframe, then adjusted by a higher-timeframe authority add-on. The final figure and its source are logged, so a wide coefficient can be traced to the authority timeframe that produced it rather than guessed at afterwards.

Owned by Volatility Intelligence Panel

The converted distance actually used for stop and trail interpretation.

Live ATR × coefficient

Owned by Volatility Distance Matrix

also called zone classification

The five-zone classification of a live ATR reading against percentile thresholds of its own baseline.

Live ATR compared to P20 / P40 / P60 / P80 baseline thresholds

Zones are percentile statements, not absolute ones — a High zone means high for this instrument, timeframe and session, and carries no comparison to any other pair. ⚑ Read sample status first: a zone computed from a thin baseline is a label rather than a classification.

Owned by Volatility Distance Matrix

The higher-context timeframe whose volatility read governs the final coefficient.

Two timeframes are in play on every trade: the trigger timeframe where entry happens, and the authority timeframe that governs management. Confusing them produces a coefficient calibrated to the wrong context — usually too wide, and usually blamed on the panel.

Owned by Volatility Intelligence Panel

Comparing the same live ATR against All Sessions, London, New York and Asian baselines to see whether the read holds across contexts.

Agreement across sessions is a stronger signal than any single comparison. Disagreement is not an error — it says the instrument is behaving unusually for one session specifically, which is itself a finding worth acting on.

Owned by Volatility Distance Matrix

The prevailing market character — Normal or Trend — which determines which branch structures are appropriate.

The regime switch is upstream of the volatility panel: the panel measures distance, it does not decide regime. The Regime Classification Engine reads context and persistence, and a regime that keeps flipping is telling you something different from one that holds.

Owned by Regime Classification Engine

The point at which a stop is moved to entry, removing the trade's remaining downside.

Moving to break-even is not free — it converts a trade that could still recover into one that can be stopped flat by ordinary noise. The ATR BE Assistant exists so the decision is made against a volatility threshold rather than against how the position feels.

Owned by ATR BE Assistant

The relationship between expectancy measured over faster windows and over slower or lifetime anchors.

4–6, 4–12, 6–12 and 12-cumulative relationship deltas

Positive drift means the faster EV windows are stronger than the slower anchors. Negative drift means expectancy is weakening. ⚑ Drawdown drift inverts: for DD the sign semantics are reversed, because worsening drawdown is bad and stabilization is good.

Owned by Structural Diagnostic Engine

Drift normalized by structural volatility, testing whether a move is meaningful rather than merely present.

Drift ÷ sigma over the 12-window

Without normalization every wobble looks like a finding. The z-score is what separates a real structural move from noise, and it is why a small persistent deviation can outrank a single dramatic one.

Owned by Structural Diagnostic Engine

Risk-Adjusted Efficiency Ratio — net P&L divided by the risk actually deployed. The capital-efficiency lens.

Composite net P&L ÷ composite risk dollars deployed

⚑ Never average branch RAER ratios. RAER is computed from composite totals; averaging the branch-level ratios produces a different and wrong number. The reading maps to a class ladder — Critical, Weak, Stable, Strong, Elite.

Owned by Structural Diagnostic Engine

Risk-Adjusted Profit Factor — profit factor normalized by actual average risk deployed.

Classified across the same five-class spectrum against a fixed benchmark anchor. Paired with RAER in a matrix whose combinations name states like under-monetized, strong system quality, institutional-grade profile, or instability masking a baseline edge.

Owned by Structural Diagnostic Engine

Weekly return and acceleration behaviour — the compounding-velocity reading.

Affects timing rather than permission: it informs when to press or delay a promotion, and does not on its own authorize a higher tier.

Owned by Structural Diagnostic Engine

Capital freedom expressed as a ratio — how constrained deployment has been.

A system spending most of its time under compressed gates is being told something about its risk configuration, independent of whether the individual weeks were profitable.

Owned by Structural Diagnostic Engine

How many consecutive periods a diagnostic state has held, as opposed to how severe it is.

Persistence beats severity. A modest deviation sustained across many weeks is stronger evidence than one dramatic week, and it is precisely the pattern intuition discounts — because intuition tracks the most recent day.

Owned by Structural Diagnostic Engine · Contradiction Engine

A state in which two diagnostic lenses that should agree are disagreeing about the same period.

A single metric can only be believed or doubted; two can be cross-examined. The Contradiction Engine exists because disagreement between lenses is information, not an error to be reconciled away.

Owned by RAER/RAPF Contradiction Engine

A grading of how much a baseline deserves to be trusted, running from No Data through to Hardened.

The first number to read on any volatility output. A zone computed from a thin baseline is a label rather than a classification, and treating it as the latter is how a confident-looking reading gets acted on before it has earned it.

Owned by Volatility Distance Matrix

A composite reading of the machine's structural condition, assembled from the diagnostic layers rather than from P&L.

A system can be profitable and structurally sick. The score exists to make that state visible while the equity curve is still cooperating.

Owned by Structural Diagnostic Engine

A fragility overlay measuring how much strain the system is operating under, independent of results.

Acts as a cap rather than a selector: it may suppress or hard-veto a higher tier, and is not permitted to authorize one on its own.

Owned by Structural Diagnostic Engine

Where the account actually lived over a period, expressed as the mean gate position rather than the current one.

A gate touched for four days and a gate lived in for eleven weeks read identically on a current-state display. Average rank separates them, and dwell is the second reading every gate row deserves.

Owned by Structural Diagnostic Engine

A diagnostic state meaning the engine has licence to say nothing yet — conditions are changing and no verdict is due.

An honest not-yet rather than a weak answer. A system that always returns a verdict on demand is manufacturing them.

Owned by Regime Classification Engine

also called transition · transitional state · mixed drift

The structural drift reading returned when the horizons neither agree on strengthening nor agree on breakdown.

Transitional is a real finding rather than a missing one. It says the short, medium and long horizons disagree, which is information: it is the reading that argues for the benchmark path and against a variant that needs conviction to justify it.

Owned by Structural Diagnostic Engine

The explicit statement, in every module's manual, of what that module may decide and what it may never override.

The most important section of any MARS manual and the most often skipped. It is what keeps eleven workbooks from becoming eleven competing opinions — each instrument's ceiling is documented as carefully as its capability.

Owned by Every operator manual

The doctrine that a module may supply evidence into a decision without being permitted to make it.

The Monte Carlo Lab informs and never authorizes. The indicators inform and never size. The Advanced EV Analytics Lab is explicitly R&D and holds no live authority. Deployment authority belongs to the Throttle Control Panel alone.

Owned by Master System Architecture

The distinction between a value the system derives and a value an operator chooses within permitted limits.

Gate state is computed — drawdown determines it and no one may set it. Final tier is selected inside the gate's cap. Almost every governance dispute in the system resolves to asking which of the two a given number is.

Owned by Cycle Command Console → Throttle Control Panel

The rule that housed engines consume CP3's already-computed outputs rather than deriving their own from raw rows.

This is why the SDE's expectancy figure always matches the EV Scorecard's — not because two calculations happen to agree, but because there is only one calculation and the second engine is reading its result.

Owned by Compliance Panel 3

A research environment whose findings carry no live authority until they pass a documented promotion workflow.

The door opens one way: evidence flows into the sandbox freely, and nothing leaves without paperwork. Sandbox → persistence window → documented verdict → promotion.

Owned by Advanced EV Analytics Lab · Scenario Manager

The rule that every user-facing state must resolve to ok, unconfigured, or error — never to silent failure.

A component that cannot do its job says so. This applies to the workbooks and to this website equally: an unconfigured checkout renders a truthful message rather than a dead button.

Owned by System-wide doctrine

Withholding an output that cannot yet be true, rather than publishing a provisional version of it.

An output computed from insufficient evidence stays dark. Insufficient sample is treated as a legitimate result to display — the console refuses to produce an expectancy status for an unfinished cycle rather than producing a weak one.

Owned by Structural Diagnostic Engine · Cycle Command Console

The pre-entry gate confirming market, system and setup clearance before risk is deployed.

Three clearances in order: Market Clear (session, approved asset, news filter, spread and liquidity), System Clear (gate state, throttle risk, cycle slot, open exposure, brake state), and Setup Clear (approved playbook, structure, trigger, logical stop, management path). A memory gate against hard failures, not a suggestion.

Owned by Pre-Trade Checklist

A written statement of expected behaviour, made before a test runs, against which the result is judged.

Defining expected behaviour in advance is what makes a test falsifiable. Without it, whatever the run produces becomes the expectation retroactively, and the exercise confirms rather than tests.

Owned by Scenario Manager · Certified Validation QA

also called deployment permission · capital authority

The system's permission to put fresh capital at risk, as distinct from the operator's judgment that a trade is worth taking.

Authority descends: drawdown sets the gate, the gate sets the pool row and a hard tier ceiling, the engine selects a tier inside it, and open exposure compresses what remains. Nothing lower in the stack may overrule anything above it, and no reading anywhere in the system converts conviction into permission.

Owned by Throttle Control Panel

also called seven lenses · selection lenses · the lenses

The set of lenses a candidate variant is checked against before an aggressive management path is permitted.

The lenses span branch EV support, gate permission, structural diagnostics, volatility and ATR condition, market structure, execution quality, and regime or variance state. ⚑ The manuals and the live page describe this differently — one as a conjunctive test where every lens must clear, the other as a scored system with hard vetoes. That difference is an open doctrinal item, not a settled reading.

Owned by Variant Attribution Module

also called weekly cadence · saturday review · tuesday · weekly ritual · review day

The fixed weekly rhythm on which evidence is compiled and the next cycle's authority is set.

The cadence is what turns a journal into a governance loop: the week closes, the scorecard grades it, structural readings update, and the following cycle inherits an authority level that was decided from the record rather than from how the last trade felt. A fixed day matters more than which day — a review that floats is a review that gets skipped in exactly the weeks it is most needed.

Owned by Weekly Trading Scorecard