MINDFORGE INTELLIGENCE
Market Risk Classifier
THE STRATEGY
INTEGRATION BRIEF
[ How MRC Plugs Into Your Book ]
What MRC is
Market Risk Classifier (MRC) classifies each U.S. trading day into one of five risk states (STABLE, SHIFTING, ELEVATED, SHOCK, CRISIS), delivered before the open. The differentiating input is exogenous environmental data — signals not in anyone else's stack and orthogonal to the regime models in the rest of yours.
The classifier also incorporates SPX and VIX as inputs alongside the environmental signals. Walk-forward validated 1990-2025. This brief covers how five strategy archetypes use the daily state as a position-sizing input.
The promise
Your strategy doesn't change.
Your sizing does.
Page 2: risk states. Page 3: find your strategy. Pages 4–8: one archetype each. Backtest: 1990-2025 walk-forward. mindforge.tech/terms
SYSTEMTHE FIVE RISK STATESWhat MRC reports each morning
MRC reports one of these every morning.
Every archetype rule in this brief references these five names. Each risk state has its own classification logic, precision, and historical frequency. There is no mapping between any one strategy and any one risk state. Every archetype uses all five.
STABLE
71%
of trading days
Default state. Markets quiet, VIX median 14. Hedges decay; vol carry earns.
SHIFTING
24%
of trading days
Regime inflection. 93% precision, 2-day median lead before stress shows in price.
ELEVATED
2%
of trading days
Selective heads-up. ~5 days/year (2%), 76% precision. A confirmation tier — pair with other evidence before sizing.
SHOCK
<1%
of trading days
Rapid vol dislocation. 85.1% precision (57 of 67). VIX already elevated at firing.
CRISIS
2%
of trading days
Crisis classifier. 92.3% precision (12 of 13), 1990-2025. Includes COVID and Aug 2024.
Distribution: 1990-2025 walk-forward. Precision figures from versioned performance records.
NAVIGATORFIND YOUR STRATEGYPage 3 of 10
Each line is one archetype: the claim, the historical effect, and the strategies it covers. Pick the one that fits what your desk runs and flip to its page. Every archetype uses all risk states from the previous page.
Archetype 1 of 5
Stop bleeding premium during STABLE.
Half-size the hedge on ~87% of days · Rolling Puts · Var Swaps · Vol Overlays
→ p.4
Archetype 2 of 5
Cut peak drawdown in half.
-52% peak drawdown (COVID) · Trend · Systematic Macro · CTA
→ p.5
Archetype 3 of 5
Close before the regime breaks.
83% of crises end with SHIFTING within 10 days · Short Straddles · Short Variance · Dispersion · Vol Carry
→ p.6
Archetype 4 of 5
Reallocate the premium budget.
~0.5x average notional held · Tail Funds · Long Vol · Crash Protection
→ p.7
Archetype 5 of 5
Tilt factor weights by regime.
Methodology framework · Momentum · Value · Quality · Low-Vol Factor Sleeves
→ p.8
ARCHETYPE 1 OF 5PAGE 4 OF 10
Rolling Puts · Var Swaps · Vol Overlays
Stop bleeding premium during STABLE.
Programs that buy protection pay premium every day whether or not the protection is doing work. MRC tells you which days are STABLE, the regime where that premium has historically decayed without paying for itself. Size the protection book down on STABLE days; ramp it back up the moment MRC flags any non-STABLE state. Same protection during stress. Less premium spent during quiet.
The one rule
During STABLEHedge notional = 50% of baselineOtherwiseHedge notional = 150% of baselineRe-enter STABLEOnly when MRC reclassifies STABLE
Historical effect, 1990-2025
~87%
of trading days are STABLE
The regime where hedge premium has historically decayed without paying for itself. 1990-2025 walk-forward.
Half-size during STABLE. Full-size everywhere else.
- ›Since 1990, VIX spiked 25% or more while the classification was STABLE 123 times. The system escalated on the 40 that proved real, held through the 77 that faded, and was wrong 6 times.
- ›STABLE runs are long. Median STABLE stretch in the 1990-2025 backtest lasted 35 trading days; the longest ran 432 days.
- ›SHIFTING carries a 92.8% precision record (64 of 69). That is the trigger to restore full hedge size.
- ›Crisis-class days were 2.7% of the 1990-2025 window. CRISIS hit 12 of 13 events (92.3% precision); SHOCK hit 57 of 67 (85.1%).
ARCHETYPE 2 OF 5PAGE 5 OF 10
Trend · Systematic Macro · CTA
Cut peak drawdown in half.
Trend strategies make money when regimes hold and lose money when regimes break. The pain concentrates in a small number of regime-change days. MRC flags those days as SHIFTING, ELEVATED, SHOCK, or CRISIS, naming which severity tier is arriving. Hold full size during STABLE (where trends actually work). Scale down or go flat the moment MRC flags a regime change. Same trend signal. Same execution. Smaller worst-case path.
The one rule
STABLEPosition size = 1.0x baselineSHIFTING0.5xELEVATED0.5xSHOCK0.25xCRISIS0.0x (flat)
Historical effect, 1990-2025
-52%
peak-drawdown reduction
COVID: -33.92% to -16.33%. SPX 50/200 SMA, 1990-2025 walk-forward.
Sustained trends require sustained regimes. The overlay scales to the regime.
- ›Episode-by-episode drawdowns shrunk: COVID -33.92% → -16.33%. Q4 2018 -10.01% → -7.09%. Aug 2024 -8.49% → -6.87%. Aug 2015 -11.24% → -7.57%.
- ›Annualized return: 8.82% static vs 6.17% with the overlay over 1990-2025 walk-forward. Sharpe 0.63 → 0.62.
- ›STABLE covers ~71% of days, so the overlay holds full size where trend P&L is generated. The overlay is a scaling factor on the position size you were already going to take.
- ›Vol-targeting, multi-asset risk balance, and signal selection stay exactly as they are today.
ARCHETYPE 3 OF 5PAGE 6 OF 10
Short Straddles · Short Variance · Dispersion · Vol Carry
Close before the regime breaks.
Selling vol earns carry when markets are quiet and breaks expensively when the regime shifts. SHIFTING puts a name on the regime shift the morning it is classified, with a scored record behind it. Desks have used the morning SHIFTING classification to lighten short-vol exposure, restoring it once STABLE is classified again. The historical re-entry signal is consistent: 83% of crisis episodes ended with a SHIFTING classification within 10 trading days, which is when carry conditions historically returned.
The one rule
STABLEFull short-vol carry onSHIFTINGClose short vol; go flatCRISIS classStay flat or long convexityRe-entryFirst STABLE after crisis ends
Historical effect, 1990-2025
83%
of crises end with SHIFTING within 10 days
24 of 29 crisis endings, 1990-2025. The historical re-entry signal.
Carry on during STABLE. Flat through SHIFTING. Re-enter on the next STABLE.
- ›STABLE-day VIX distribution: median 14.3 (1990-2025). Forward 5-day VIX change essentially zero in median. The carry environment short-vol books are designed for.
- ›On SHIFTING days, the median forward 5-day VIX move is -3.3%, but the tails widen sharply (10th percentile -16.4%, 90th percentile +19.8%) compared to STABLE. Closing on SHIFTING compresses exposure to the painful tail.
- ›Late short-vol entry on SHOCK has historically traded into mean reversion. SHOCK classifications happen at already-elevated VIX (median 25.89, 1990-2025); median forward 5-day VIX move is -19.6%. Wait for the STABLE or SHIFTING that follows.
- ›Re-entry trigger: 24 of 29 crisis episodes (83%) ended with a SHIFTING classification within 10 trading days.
ARCHETYPE 4 OF 5PAGE 7 OF 10
Tail Funds · Long Vol · Crash Protection
Reallocate the premium budget.
Long-volatility books bleed premium during STABLE and earn convexity during crises. The same annual budget gets a much better payoff if you spend less during STABLE and more during MRC-flagged crisis states. CRISIS flagged 12 of 13 events in the 1990-2025 backtest, so the scaled-up notional has historically landed where the convexity actually lives. Same dollar budget. Deeper protection where it matters.
The one rule
STABLENotional = 0.25x baselineSHIFTING1.0xELEVATED1.0xSHOCK1.5xCRISIS2.0x
Historical effect, 1990-2025
~0.5x
average notional held
vs constant 1.0x static baseline. Carry drag halved on the same dollar budget.
Premium reallocated, not added. Same budget, deeper protection where it matters.
- ›Average notional held in the 1990-2025 daily stream was 0.49 versus a constant 1.0 under the static policy. Carry drag (stylized proxy) was cut roughly in half.
- ›Convexity scales up on SHOCK (1.5x) and CRISIS (2.0x). MRC’s CRISIS classifications are 12 of 13 (92.3% precision); SHOCK is 57 of 67 (85.1%).
- ›Across the 13 CRISIS episodes in the 1990-2025 backtest, the scaled-up notional landed on the days the system flagged systemic stress, the regime where long-volatility convexity has historically paid off.
- ›COVID dominates expected payoff. 1 of 9 CRISIS episodes produced a -30.64% SPX drawdown to trough; the other 8 ranged from -0.15% to -6.23%. The modal CRISIS event is much smaller than COVID. Size accordingly.
ARCHETYPE 5 OF 5PAGE 8 OF 10
Momentum · Value · Quality · Low-Vol Factor Sleeves
Tilt factor weights by regime.
Different factors win in different regimes. Momentum crashes hardest during regime breaks; quality and low-vol earn during stress; value rotates with macro. MRC provides an exogenous regime input to drive factor tilts, instead of inferring regime from the same prices the factors already use (which is circular). The framework here is a methodology spec; the historical backtest is reproducible on your firm’s own factor return series.
The one rule
STABLEOverweight momentum; underweight low-volSHIFTINGReduce momentum; lift quality + low-volELEVATEDQuality + low-vol overweightSHOCKMaximum defensiveCRISISDefensive; momentum at floor
Historical effect, 1990-2025
Spec
methodology framework
Reproducible on your factor data. 60-day pilot includes full 1990-2025 backtest data for independent verification.
Exogenous regime input. Not inferred from the prices the factors already use.
- ›Factor returns are regime-dependent. The standard quant portfolio infers regime from the same price data the factors themselves use, which is a circular dependency. MRC provides a regime signal that does not come from the same price stream.
- ›The momentum-crash episodes in 1990-2025 (Aug 2015, Q4 2018, COVID) all coincided with MRC classifying non-STABLE. The underweight-momentum tilt would have been active across each of those windows.
- ›No published quant backtest in this brief because factor-return series are firm-specific. The framework is documented for reader-side reproduction on Kenneth French Data Library factors or proprietary series.
- ›The 60-day pilot delivers daily MRC classifications with the full 1990-2025 backtest dataset so your team can merge them with internal factor backtests directly.
INTEGRATIONWHY YOUR TEAM CAN SHIP THIS FASTThree Reasons
The signal sits upstream of your stack.
1
Your alpha stays yours.
The regime signal scales the trades you already make. Vol-targeting, signal selection, instrument choice, and risk framework all stay exactly as they are today.
2
One signal at 07:30 AM ET.
Delivered before the U.S. open via API or webhook. No intraday updates. Your risk system reads it once, looks up the sizing rule, applies it before you trade.
3
Quants reproduce every number.
The 60-day pilot includes the full 1990-2025 backtest dataset. Your team merges it with internal backtest infrastructure and reproduces every figure on these pages on your own book.
ACCESSPILOT
RUN IT ON YOUR BOOK
60-Day Institutional Pilot
INCLUDES
- • Daily classification via email or API
- • 5+ years historical classifications (full 1990-2025 backfill under pilot)
- • Methodology and integration specs
- • Weekly research briefs
QUALIFICATION
Institutional 60-day pilot for qualified research teams. Reproduce every figure in this brief on your own factor data, hedge book, or strategy series.
RESTRICTED TO INSTITUTIONAL USE
OVERLAY + CRISIS
This brief: how MRC plugs into five strategy archetypes as a sizing overlay.For the regime-classification accuracy that powers the overlay, see
The Escalation Record.