Workflow Overview · The Feedback Principle
How MARS uses this
MARS never trusts a single blended figure. Weekly branch probabilities produce a separate EV per branch, weighted by the live profile (45 / 33.75 / 11.25 / 10 by default) into the blend. Reviews read the decomposition first: which branch produced the edge, which one leaked it, and whether the weights still match the evidence.
How it benefits you
You stop repairing the wrong thing. When results dip, the decomposition shows whether the whole system weakened or one branch is dragging - so you fix a branch instead of overhauling a working system, and you spot a single strong branch masking weakness everywhere else.
Branch-level EV decomposition with blend weights. The blended number alone would hide exactly this picture.
The key idea
The feedback principle
The loop ends where most traders begin: changing rules. In MARS, a rule change is tested in the EV Lab and scenario tools first, validated against evidence, then deliberately promoted to production. Impulse edits after a losing streak are exactly what the loop exists to prevent.
Where it lives
02Within Workflow Overview, this is one load-bearing idea — worth its own page. The parent module frames it this way: MARS operates on a repeated rhythm across eight horizons — pre-trade, post-trade, daily/cycle, weekly, monthly, quarterly, annual, and structural. The loop is what turns raw activity into governed intelligence.
Promotion friction
The feedback principle deliberately slows improvement traffic: a validated change enters production at the next scheduled review, never mid-cycle. The queue does more than prevent hot-patching — it batches changes so each review window measures a known configuration, keeping the evidence base interpretable while the system evolves.
Further illustration
Every decision in MARS is routed through this hierarchy in order. The plan defines doctrine; gate state defines capital posture; the throttle converts posture into deployment numbers; evidence and diagnostics inform from below; R&D stays sandboxed at the bottom until validated. When layers disagree, the higher layer wins - always.
The MARS authority stack. Authority flows down; evidence flows up; nothing lower may override anything above it.
Before you go deeper
Connected inside MARS
Go deeper
Deep dive — 01
A promoted change doesn't end the loop. It resets what every subsequent measurement is measuring.
Read the full brief →
Deep dive — 02
Not a pipeline that ends in a report — a loop that ends in changed behavior, which generates the next evidence.
Read the full brief →
Deep dive — 03
The alternative is faster — exactly until the first untested change meets its first live drawdown.
Read the full brief →
One price. Eleven workbooks, three TradingView indicators, and the full manual library — $497.
