What the indicators are allowed to do — and what they never touch.

Where they sit

Squarely inside the operator's territory, not the system's.

The discretion/governance split defines two territories: the operator decides which setup is valid, which branch a trade belongs to, whether volatility context is acceptable, and whether the trade is worth taking; the system decides gate, tier, pool, exposure, and whether deployment is permitted. All three indicators serve exclusively the first column. They help answer 'is this a valid setup in a real trend with confirming structure?' — a question the governance layer never asks and never answers. That placement means an indicator can be spectacularly right and still not change a single number in the throttle's output.

The three tools

Trend, structure, and risk mechanics.

Each tool covers a different question inside the discretionary layer. The Trend Continuation Indicator answers directional validity — is the trend real and confirmed across the timeframes that matter — and it's the workhorse behind the continuation strategies. EW Convergence answers structural context: where the wave phases align across timeframes, and where a corrective leg is likely terminating. The ATR BE Assistant answers the regime and risk-mechanics question — Normal or Trend, and therefore which breakeven and management doctrine applies — which is why it doubles as a Volatility-category module. Together they cover direction, structure, and mechanics without overlapping much, which is why the trio is bundled rather than sold as alternatives.

The boundary in practice

A perfect signal in a restricted gate is still a restricted gate.

The rule that governs every evidence module governs these too: a favorable read cannot override a restricted gate or System Lock, and any position it informs must size inside throttle-approved risk. Concretely — the Trend Continuation Indicator lighting up across every timeframe while the account sits in Floor produces exactly one legitimate response: take the trade if the setup genuinely qualifies, at the tier the gate authorizes, which may be T1. The indicator's confidence is not a reason to deploy more. It never was an input to that question. This is worth internalizing precisely because indicator conviction is emotionally the most persuasive signal in the stack — it arrives visually, in real time, at the moment of decision.

• Indicators can block a trade in the operator's judgment; they can never expand one in the system's arithmetic.
• The launch checklist places regime and setup checks before size and risk checks — the ordering encodes this boundary.
• An indicator-driven override of throttle output is an override like any other: logged, reasoned, and counted.

Why they're built in-house

Tools that match the doctrine they serve.

Generic indicators answer generic questions; these were built to answer the specific ones MARS asks. The Trend Continuation tool exists because the continuation strategies need multi-timeframe alignment confirmed the way the plan defines alignment, not the way a default template does. EW Convergence exists because the three-stage breakout strategy needs structural confirmation that a trend will continue in the intended direction. And the ATR BE Assistant exists because the branch doctrine's breakeven points and trail unlocks are specific values that a generic ATR tool has no knowledge of. Building them in-house is what lets the indicator layer and the governance layer share vocabulary.

The key idea

Better inputs to a governed decision — that's the whole offer.

The indicators don't make the system work; the rails do that. What they do is improve the quality of the one thing the rails deliberately leave to the operator — trade selection — and improve it in the specific dimensions the strategies depend on. A trader with better trend confirmation and better structural context still deploys exactly what the gate and throttle authorize. They just do it on better trades, which over hundreds of trades is precisely where discretionary skill compounds.