Structural Diagnostic
Performance Efficiency Gradient (PEG)
“Is the machine getting better at converting effort into result?”
PEG measures the direction and credibility of the live-versus-benchmark efficiency gap in one signed number. A positive gradient means live deployment is outperforming the fixed frame — weighted up when RAER class is strong (the outperformance is backed by real conversion) and damped when RAER is weak (the gap is probably noise).
The formula, derived
The live reading (PF ÷ actual average risk) minus the benchmark reading (PF ÷ range median), scaled by the RAER class weight (Critical 0.1 → Elite 1.0) — so the gradient only speaks loudly when capital conversion says the efficiency is credible.
How MARS reads it
Technical explanation
Aggregates the slopes of capture efficiency, RAER, fee-adjusted outcomes, and stop-efficiency scores across rolling windows.
Positive PEG during flat P&L periods is one of the most encouraging states in MARS: the machine is improving before results show it.
Negative PEG during profitable periods is the mirror warning: results are coasting on prior quality while conversion erodes underneath.
Interpretation bands
Strong. Neutral. Weak.
Strong
Efficiency components improving together — the machine compounds skill, not just capital.
Neutral
Flat gradient — stable machine; improvement requires deliberate intervention, not more sample.
Weak
Eroding conversion beneath acceptable results — schedule execution review before P&L confirms the decay.
Use cases
Where it earns its place
- ▸Verifying that execution repairs from the MAE/MFE Lab actually moved the gradient
- ▸Distinguishing plateau from decay in mature systems
- ▸Prioritizing review effort toward the fastest-eroding efficiency component
- ▸Evidence for promoting or rejecting R&D changes based on measured improvement
Edge cases
Where it can mislead
- !Regime-driven gradients: efficiency naturally shifts across volatility regimes; PEG is read against regime context to avoid false alarms.
- !Repair lag: execution changes take sample to surface — gradient windows must match realistic adaptation horizons.
Example scenarios
The metric in the wild
Improvement before profit
Three flat weeks of P&L, but capture efficiency climbed from 52% to 64% and fee drag halved. PEG reads strongly positive — the system holds posture instead of abandoning a machine that is visibly improving.
Monte Carlo connection
Simulation assumes static efficiency parameters. A strong live gradient means the benchmark is becoming conservative; a negative one means it is becoming optimistic — either informs refresh timing.
Monte Carlo Lab →Live benchmark comparison
Benchmark deviation is decomposed with PEG: improving efficiency explains benign outperformance; eroding efficiency explains drift under the median.
7-Tier MC Benchmark →Go deeper
Operator briefs on this territory.
Deep dive — 01
Expectancy arithmetic: what a trade is worth before it happens.
The EV formula walked end to end — why per-trade worth beats realized P&L as the master metric.
Read the full brief →
Deep dive — 02
Reading the fan: Monte Carlo without self-deception.
What the percentile bands actually promise, why P10 governs sizing, and where simulation authority ends.
Read the full brief →
Deep dive — 03
MAE/MFE: what your trades did while you weren't looking.
Reading maximum adverse and favorable excursion to audit stops, targets, and exits with evidence.
Read the full brief →
PEG is calculated for you — automatically.
Every reading on this page is produced, tracked, and interpreted inside the MARS workbook ecosystem.