Risk-Tier Performance · Review Cadence
Four cadences, four different questions.
Cadences
4
Tier read
MONTHLY
Judged
CONDITIONALLY
Read this first
Further illustration
The Dynamic 7-Tier Benchmark converts the drawdown distribution into modeled tier-usage expectations: given the gate transitions the evidence implies, T1-T3 should host most deployment while T5-T7 stay exceptional. Live tier dwell is audited against this profile every cycle.
The seven-tier ladder with modeled dwell: how often each tier should be in use given the gate history. High tiers exist to be rare.
The conditional read
01Usage is judged against the gates that actually happened.
A month spent mostly in the lower tiers is not automatically timidity. Under a compressed gate history it is exactly correct behaviour, and grading it against an unconditional average would penalise the operator for obeying the ladder. The review therefore conditions expected dwell on realised gate states before scoring any gap, so the operator is graded on how they behaved inside the month they actually had rather than an average month that never occurred.
Three cross-checks
02Tier usage is never read alone.
Tier behaviour arrives at the monthly slot alongside gate dwell and throttle behaviour, and the three are interpreted together. Heavy upper-tier usage during a growth-dominant quarter is normal. The same usage during extended defensive dwell is a governance breach in progress, and the difference between those two readings is not in the tier row at all.
- Tier-conditional R: each tier must show outcomes proportional to its extra risk.
- Override log cross-check: elevated tier usage plus frequent full-tier overrides is the aggression-drift signature.
- Under-usage gets equal scrutiny, because unmonetized authorised expectancy is a real cost.
Refresh discipline
03The ruler is re-marked on a schedule, with a version number.
Tier usage distribution is refreshed monthly. Completion probability and lock rate are refreshed quarterly or after a major model change. Every refresh is versioned, and the operating standard is updated alongside the workbook. An unversioned refresh is the quiet failure mode of any benchmark: the numbers improve, nobody can say when the ruler changed, and last quarter's comparison becomes unreadable in retrospect.
What the annual slot is for
04One question, asked once, with no partial credit.
The full-horizon review compares completion, lock, final equity, maximum drawdown, and risk-adjusted performance against the simulated baseline, and returns one of three answers. It is the only slot where the benchmark is used as a verdict rather than as a monitor, and it is deliberately annual, because a survival-adjusted comparison run over a shorter window is dominated by whichever part of the distribution the account happened to be traversing.
How MARS uses this
Monthly review places live tier usage against the model conditionally — usage is judged given the gates that actually occurred, not against an unconditional average. This chart is the review's exhibit: the gap between modeled and lived deployment, tier by tier, read alongside gate dwell before any conclusion.
How it benefits you
Aggression creep gets caught while it is still a pattern on a chart rather than an oversized loss. Equally, unnecessary timidity shows up as a measurable gap, so you deploy the edge you have actually earned instead of leaving modeled expectancy on the table.
The monthly exhibit: live tier deployment against conditional expectation, the gaps being the agenda.
Reference
The review routine
| Cadence | What is compared | Decision output |
|---|---|---|
| Weekly | Cycle results, gate state, open exposure, early equity path, throttle behaviour | Is this week inside normal variance or creating early stress? |
| Monthly | Ending equity, return, max drawdown, gate dwell, tier usage, regime, smart exposure | Is the month inside, above, or below benchmark? |
| Quarterly | Completion pacing, drawdown persistence, tier behaviour, throttle efficiency | Is the system still aligned with the benchmark model? |
| Annual | Completion, lock, final equity, max drawdown, risk-adjusted performance | Did live results beat, match, or fail the simulated baseline? |
Connected inside MARS
This module doesn't work alone.
Go deeper
Operator briefs on this territory.
Deep dive — 01
How often tier performance gets re-examined — and why not more.
Slice a small sample seven ways and something always looks broken. Cadence is the cure.
Read the full brief →
Deep dive — 02
Tier usage is graded against the gates that actually occurred.
Heavy T5 in a Growth quarter is normal. The same usage under Buffer dwell is a breach in progress.
Read the full brief →
Deep dive — 03
The governance layer is required to justify itself, on the same evidence as everything else.
With-throttle against no-throttle, on the same paths. The governance layer does not get a pass.
Read the full brief →
Every module ships in the complete MARS package.
One price. Eleven workbooks, three TradingView indicators, and the full manual library — $497.