Methodology

How The Signal System Works — Full Parameter Disclosure

Every data source, every decision tree threshold, every position-sizing multiplier, every backtest configuration parameter, and every known limitation of the Market Signal Dashboard. This page exists so anyone — signed in or not — can audit the system before trusting it.

View Live Dashboard

Data Flow

Data Sources (Live APIs)
FRED
Federal Reserve
Credit OAS (daily)
Corp Profits (quarterly)
Free / Government
FMP
Financial Modeling Prep
S&P 500 EPS (7d cache)
SPY price (5m cache)
Historical bars (1h cache)
Paid API
Unusual Whales
Options Intelligence
Option chains (5m)
IV rank (5m)
Flow alerts (1m)
Dark pool (1m)
GEX, max pain
Paid API
Perplexity
AI Search
Live news (5m cache)
Paid API
Signal Engine (Deterministic Decision Tree)
1Earnings Growth > -2% YoY?YES / NO
2Credit Spread Quartile ≤ Q2?YES / NO
3Valuation Spread in Q4?YES / NO
Output: BULLISH / CONTRARIAN_BUY / HOLD / CAUTION
Base probabilities: 0.78 / 0.72 / 0.55 / 0.50 + live calibration bonuses
Hypothetical Play Generator
Strike = signal template x VIX adjustment x DTE multiplier
Size = bankroll x 5% x confidence x half-Kelly fraction
Price = UW real bid/ask when available, B-S estimate otherwise
Adjustments: IV rank (>80 = 0.6x size), GEX (volatile = 0.8x), backwardation (+10% confidence)
AI Chat (Claude Opus 4.6)
22 on-demand tools — calls data sources as needed
Extended thinking (32k token budget)
Mandatory disclaimer on every market response
AI responses are generated, not deterministic. May contain inaccuracies.

All Model Parameters

Decision Tree

ParameterValueSource / Rationale
Earnings threshold> -2% YoYWalk-forward backtest (226 signals, 1998-2025)
CONTRARIAN_BUY base probability0.78Same dataset
BULLISH base probability0.72Same dataset
HOLD base probability0.50Same dataset
Credit Q1-Q4 bonuses-0.03 / +0.02 / -0.02 / +0.05Bayesian-blended from quartile accuracy
Valuation Q1-Q4 bonuses-0.03 / +0.03 / -0.01 / +0.05Bayesian-blended from quartile accuracy
Earnings coefficient0.004 per 1% growth (r²=0.01)Very weak predictor, capped at ±0.03
Probability clamp0.10 to 0.95Prevents extreme confidence

Position Sizing

ParameterValueSource / Rationale
Base risk per trade5% of stated bankrollConservative starting point
Kelly fractionHalf-Kelly (0.5x)Standard ruin-avoidance practice
CONTRARIAN_BUY multiplier1.0xFrom 81% backtest win rate, 5.1% avg return
BULLISH multiplier0.58xFrom 79% win rate, 2.9% avg return
CAUTION multiplier0.06x~52% accuracy (near coin flip)
IV Rank > 80 adjustment0.6x size (expensive options)Reduces exposure when IV is elevated
IV Rank < 20 adjustment1.2x size (cheap options)Increases exposure when IV is low
Negative GEX adjustment0.8x sizeReduces exposure in volatile regime

Pricing Model

ParameterValueSource / Rationale
ModelBlack-ScholesStandard options pricing
Risk-free rate4.5% annualApproximate current T-bill rate
IV proxyVIX (market-wide)NOT strike-specific — ignores smile/skew
Real pricingUW bid/ask when availableFalls back to B-S estimate

Backtest Configuration

ParameterValueSource / Rationale
Period1998-2025 (226 monthly signals)Walk-forward, no lookahead bias
Holdout (out-of-sample)2021-presentSeparate from training period
Transaction cost4% per tradeConservative estimate
Sharpe method30-day non-overlapping, √12 annualizedAvoids autocorrelation inflation
Earnings data lag60 days (FRED) / 90 days (FMP)Simulates publication delay

Known Limitations

ParameterValueSource / Rationale
Earnings signal strengthr² = 0.01 (1% explanatory power)Very weak predictor
CAUTION accuracy~52%Near coin flip — sized at 0.06x Kelly accordingly
Sharpe ratioBelow 1.5 institutional thresholdModest edge, not fund-grade
CSCV overfitting testHigh PBO detectedIn-sample may overfit out-of-sample
VIX as IV proxyMarket-wide, not strike-specificIgnores volatility smile/skew

All parameters above are derived from historical backtested data and may not reflect future market conditions. This system is an educational demonstration built by a software developer — not a registered investment advisor. See the full risk disclosure for details.

Educational Purpose Only

This methodology describes the Market Signal Dashboard — an educational research tool, not a trading product. Every parameter listed above is derived from historical backtested data and may not reflect future market conditions. The creator is a software developer, not a registered investment advisor, broker-dealer, or financial planner. Nothing here constitutes investment advice. Full Risk Disclosure

Want to understand this better? Book a call.

Happy to walk through the methodology, discuss the limitations, or answer questions about how this applies to your own data.

Book a Call