Live vs Benchmark · Classifying Deviation
How MARS uses this
Classifying a deviation as alpha, variance, drag, or failure starts with the drawdown axis. Returns above the model with drawdown outside these bands is over-risk wearing an alpha costume; returns below the model with drawdown well inside them is often drag, not decay. The distribution is the classifier's first witness.
How it benefits you
You get a pre-agreed definition of normal pain, decided calmly before the pain started. That single artifact prevents the two classic errors - panic de-risking inside ordinary variance, and denial when damage is genuinely abnormal. You always know which situation you are in.
The classifier's first check: where live drawdown sits against the simulated distribution decides which deviation story is even plausible.
Classifying deviation
Returns above the model with worse drawdown is probably over-risk, not alpha. Above the model with excessive overrides is governance leakage. Above the model with weak EV is likely variance. Above the model with poor RAER means risk is being wasted. True MARS alpha is benchmark outperformance with controlled drawdown, positive expectancy, efficient conversion, clean profit quality, and sustainable acceleration.
Where it lives
Within Live vs Benchmark, this is one load-bearing idea — worth its own page. The parent module frames it this way: The live comparison is deliberately not a shallow equity race. Five metric-level questions are asked against the model — EV, drawdown, RAPF, RAER, acceleration — plus behavioral checks on gate dwell, tier usage, throttle behavior, and open exposure.
The waiting rule
Single-review gaps stay unlabeled by rule. A deviation only receives its classification — alpha, variance, drag, or failure — after it survives the next evidence window, because most gaps close on their own and premature labels trigger premature interventions. The queue of 'watching, not yet named' items is a feature of the process, not indecision.
Doctrine
Connected inside MARS
Go deeper
Deep dive — 01
Variance heals, drift trends, breaks step-change. Misclassify the gap and every response is wrong.
Read the full brief →
Deep dive — 02
Beat the model on four of five and you have not produced alpha. You have produced a diagnosis.
Read the full brief →
Deep dive — 03
Most gaps close on their own. A label applied before that happens creates a response the evidence never asked for.
Read the full brief →
One price. Eleven workbooks, three TradingView indicators, and the full manual library — $497.
