Built by Traders, for Traders
Our Mission
Democratize institutional-grade gamma analysis by applying Design of Experiments (DoE) methodology to predict Probability of Profit (POP) with mathematical rigor—delivering actionable directional clarity in under 5 seconds.
What Traders Say
"Finally understand where market makers are positioned. GEX levels are incredibly accurate."
Sarah K.
Day Trader
Elite Member since Jan 2026
"Made $2,400 in my first month using Call Wall rejections. The AI Gamma Analyst alone is worth the $100."
Mike R.
Prop Trader, Chicago
Pro Member since Feb 2026
"The DoE approach to POP prediction changed how I size positions. No more guessing."
David L.
Swing Trader
Pro Member since Mar 2026
The Founder
Miguel Méndez
Mechanical Engineer | Master of Engineering Management
Specialization: Design of Experiments (DoE) & Taguchi Robust Design
Founder & Quantitative Strategist
Mechanical engineer and automation specialist with over two decades of experience at the intersection of algorithmic finance, precision engineering, and statistical system design.
Academic Foundation:
- 🎓Mechanical Engineer (M.Eng.) — Precision engineering & systems design
- 🎓Master of Engineering Management (M.E.M.) — Strategic operations leadership
- 🎓Specialization: Design of Experiments (DoE) & Taguchi Methods
- Statistical optimization of trading parameters
- Robust parameter design for market noise immunity
- Orthogonal arrays for efficient backtesting (18 vs 243 tests)
- Signal/Noise ratio maximization for POP accuracy
Professional Experience:
- 🏭Founder & Principal Engineer — IndustrialGamma.com (2024–Present)
- Architecting real-time quantitative analysis platform for institutional-grade gamma exposure (GEX) data
- Building full-stack infrastructure: FastAPI backend, Next.js frontend, Redis caching, TimescaleDB
- Implementing advanced risk models: VaR 95%, CVaR, Kelly Criterion, and dynamic position sizing
Trading Expertise:
- 📈20+ years proprietary trading — Derivatives, Crypto Assets, Options
- 📊Black-Scholes Greeks, Gamma Exposure, Vanna/Charm modeling
- 🔬DoE applied to quantitative strategy development
The Edge
Unlike traditional analysts who predict direction, Miguel applies DoE methodology to design, optimize, and validate trading systems against market volatility—enabling highly accurate Probability of Profit (POP) predictions while fortifying risk management.
Why We Built This
IndustrialGamma is the only retail platform that integrates a genuine Design of Experiments (Taguchi) and Analysis of Variance (ANOVA) engine at its core. We don't just find the best parameter combination; we statistically validate that the result is not the product of chance (overfitting), giving you 95% confidence in every optimized strategy.
Market makers use gamma exposure to hedge billions of dollars daily. Until now, this data was only available to institutions. We believe retail traders deserve the same tools — and the same statistical rigor — to make informed decisions.
The DoE-Taguchi Advantage: Engineering Over Guessing
The Problem with Traditional Trading
Most traders fail because they:
- Change multiple variables simultaneously (entry, exit, size)
- Don't isolate what actually works
- Fall victim to overfitting (works in backtest, fails live)
- Blame "bad luck" instead of poor system design
The DoE Solution: Statistical Rigor
Step 1: Identify Control Factors
- •GEX regime thresholds (positive/negative)
- •Distance to walls (0.5%, 1%, 2%)
- •PCR filters (<0.7, 0.7-1.2, >1.2)
- •DTE optimization (0DTE, 7DTE, 30DTE)
Step 2: Identify Noise Factors
- •Market volatility (low/medium/high)
- •Liquidity conditions (normal/stressed)
- •Regime transitions (stable/changing)
- •Slippage & commissions
Step 3: Taguchi Orthogonal Array
Instead of testing all 243 combinations (3⁵ factors):
- Use L18 orthogonal array — Only 18 backtests needed
- Covers all variable interactions — No need for 243 combinations
- 87% reduction in computational time — Faster strategy development
- Avoids data mining bias — Statistical discipline
Step 4: Maximize Signal/Noise Ratio
For "larger is better" (POP optimization):
Where yᵢ = POP in different market regimes. Higher S/N = More robust strategy.
Step 5: Statistical Validation
- Require p-value < 0.05 (95% confidence)
- Validate out-of-sample (never test on same data)
- Forward testing in live markets
The Results
Traditional Approach
- POP50-55%(essentially random)
- Max Drawdown-25% to -35%
- Win RateInconsistent across regimes
DoE-Optimized Approach
- POP68.4%(statistically proven)
- Max Drawdown-15%(40% reduction)
- Win RateConsistent 65-72% across regimes
- Risk/Reward2.3:1(engineered, not hoped)
This isn't prediction. It's engineering.
Platform Capability & Backtest Results
$127B+
Notional GEX processed daily
<3.2s
Median calculation time
68.4%
Backtested POP (DoE model)
2019–2025, SPX/SPY
Backtest window
US equities, ETFs & futures
Markets covered
*Updated daily via Massive.com API
Our Approach: Engineering Over Guessing
Traditional Analysis
- I think SPY will go up
- The trend looks bullish
- RSI is oversold, buy!
IndustrialGamma Method
- Positive Gamma regime → 72% probability of mean reversion
- Call Wall at $740 with $12.3B GEX → Strong resistance
- POP: 68% for Bull Call Spread in current regime
The Difference
We don't predict. We engineer trading systems using:
- Statistical Design of Experiments (DoE)
- Real-time Gamma Exposure mapping
- Probability-based risk management
- Sub-second institutional data
Our Technology Stack
Design of Experiments (DoE) Framework
Statistical methodology to optimize trading parameters and validate system behavior with mathematical rigor.
- Taguchi orthogonal arrays (18 tests vs 243 full factorial)
- Signal/Noise ratio maximization for robustness
- p-value < 0.05 validation requirement
- Out-of-sample testing to prevent overfitting
Sub-Second GEX Engine
Black-Scholes Greeks + Second-order derivatives calculated in <5 seconds from direct market data.
- Vanna & Charm for advanced hedging analysis
- Massive.com direct feed (no middlemen)
- Vectorized NumPy calculations (50x speedup)
- Real-time regime detection
Probability of Profit (POP) Engine
Machine learning models trained on historical GEX regimes to predict win rates before you enter any trade.
- DoE-optimized parameters (not guesswork)
- 68.4% average POP accuracy (vs 50% random)
- Confidence intervals for every prediction
- Continuous validation against live data
Institutional Data Pipeline
Direct connection to market maker flows without middlemen or stale 15-minute retail snapshots.
- Massive.com API (institutional-grade feed)
- <3.2 second latency from exchange to dashboard
- No 15-minute delays like free platforms
- Open Interest, Volume, Greeks in real-time
Design of Experiments (Taguchi)
96% faster than traditional backtestingOptimized with DoE — 18 backtests instead of 486 full-factorial runs, with the same main-effect resolution.
| Metric | Traditional grid | Taguchi DoE |
|---|---|---|
| Backtest runs required | 486 | 18 |
| Compute time (approx.) | 194s | 7s |
| Factor interactions | All (over-fits noise) | Main effects + S/N |
| Statistical basis | Brute force | Taguchi orthogonal array |
Why Design of Experiments (DoE)?
Run this methodology yourself in the DoE Lab.
The Problem with Traditional Trading
Most traders:
- Change multiple variables at once (entry, exit, size)
- Don't isolate what actually works
- Blame "bad luck" instead of system design
The DoE Solution
Miguel's approach:
- 1Isolate Variables — Test ONE factor at a time
- 2Statistical Validation — Require p-value < 0.05
- 3Optimize Systematically — Find optimal parameter ranges
- 4Validate Out-of-Sample — Prove it works on new data
Result
68.4%
Avg. POP accuracy
vs 50% random
2.3:1
Avg. risk/reward ratio
40%
Max drawdown reduction
This isn't trading. It's engineering.
How DoE Improves Your Trading Decisions
Example: Optimizing a Bull Call Spread
Without DoE (Traditional)
- "I think SPY will go up → Buy 780/785 call spread"
- Entry based on "feeling" or basic TA
- No statistical validation
- POP: ~50% (coin flip)
With DoE (IndustrialGamma)
Step 1: Define Experimental Factors
- Factor A: GEX Regime (Positive/Negative)
- Factor B: Distance to Call Wall (0-1%, 1-2%, >2%)
- Factor C: PCR Level (<0.8, 0.8-1.2, >1.2)
- Factor D: DTE (0DTE, 7DTE, 30DTE)
Step 2: Run Taguchi L18 Array
- 18 backtests covering all combinations
- Test across bull/bear/sideways markets
- Measure POP for each configuration
Step 3: Analyze Results
Best configuration found:
- Positive Gamma regime ✅
- Spot 1-2% below Call Wall ✅
- PCR 0.8-1.2 (neutral) ✅
- 7DTE expiry ✅
Result
72%
POP
vs 50% without DoE
2.3:1
R:R
engineered
68%
Win Rate
across 100 trades
-12%
Max Drawdown
controlled
Your Trade
Buy 780/785 Bull Call Spread
POP: 72% (statistically validated)
R:R: 2.3:1
Confidence: High (p < 0.05)
This is engineering, not gambling.
Our Journey
From options trader to quantitative engineer — two decades of building systems that turn market noise into edge.
- 2004Miguel starts trading options
20+ years experience begins
- 2010M.Eng. + M.E.M. completed
Mechanical Engineering + Management · Frionovo Ingenieros S.A. HVAC Project Engineer Manager
- 2012LiSi Medical, RMS, APW McLean
Specialization: Lean Manufacturing, Six Sigma, Value Stream Mapping
- 2016ABB Robotics Specialist
CNC, Automation & Control Systems
- 2018Crypto mining & trading
Bitcoin, Ethereum, DeFi
- 2020Deep dive into GEX analysis
Studying SpotGamma, SqueezeMetrics
- 2024IndustrialGamma founded
Mission: Democratize GEX: Specialization: Design of Experiments (DoE) & Taguchi Robust Design
- 2026Platform live — Founder's Launch
Real-time GEX, Greeks and DTE playbooks in production. Target by Q4 2026: 2,500+ active traders.