Descriptive only. Shows where each stock ranks against its peers across four foundational factors based on current data. We have not validated that these ranks predict forward returns. We will only upgrade these to "predictive" signals after they pass rigorous out-of-sample holdout validation.
+
+ ๐ก
+
+ This is a research starting point, not a buy list. High rank = "worth a closer look," not "will go up."
+ Think Zillow for stocks: it shows you the facts fast โ you still do the homework and decide.
+ We will only upgrade these to "predictive" signals after they pass rigorous out-of-sample holdout validation.
+
+ Descriptive Only. Options flow imbalance is shown strictly for market context and sentiment analysis.
+ The "Smart Money" thesis has not been validated out-of-sample as a predictive signal by our engine. Do not treat flow as a buy/sell signal.
+
+
+
{/* Header with Date and Quick Actions */}
diff --git a/frontend/src/components/modals/MethodologyModal.jsx b/frontend/src/components/modals/MethodologyModal.jsx
new file mode 100644
index 0000000..97d31b0
--- /dev/null
+++ b/frontend/src/components/modals/MethodologyModal.jsx
@@ -0,0 +1,91 @@
+import { Info, X } from 'lucide-react';
+import { Button } from '@/components/ui/button';
+
+export default function MethodologyModal({ isOpen, onClose }) {
+ if (!isOpen) return null;
+
+ return (
+
+
e.stopPropagation()}
+ >
+
+
+
+
+ Our Methodology: Why We Don't Sell Predictions
+
+
+
+
+
+
+
+
The Out-of-Sample Reality
+
+ Most platforms show you backtests that work perfectly. We're showing you ours that didn't โ because that's how you know our descriptive tools are honest.
+
+
+ We built a highly sophisticated predictive scoring engine based on technical indicators and options flow. We optimized it, tuned the hyperparameters, and achieved a 63% win rate in-sample.
+ Then, we tested it honestly. We ran a strict out-of-sample forward test on unseen data.
+
+
+
+
+
+
+
Test Group
+
Expectancy (R-Multiple)
+
Hit Rate
+
+
+
+
+
Random Control (Coin Flip)
+
+0.027R
+
49.2%
+
+
+
Top-Graded "A" Signals
+
+0.019R
+
48.5%
+
+
+
+
+
+
+ The Result: The edge completely vanished. The highly-engineered signals failed to beat a random coin flip. So we stopped selling them.
+
+
+
+
+
The FactorLab Approach
+
+ Instead of black-box predictions, we built the FactorLab. This is a purely descriptive lens.
+
+
+ We take the Top 50 liquid stocks and ETFs, ingest their raw fundamentals and technicals, and mathematically rank them across four academic factors:
+ Value,
+ Quality,
+ Low-Volatility, and
+ Momentum.
+
+
+ When you see a stock is in the 95th percentile for Quality, it is a mathematical fact about its balance sheet relative to its peers. It is not a buy recommendation. Think of this tool like Zillow for stocks: we show you the facts fast, but you still do the homework and make the decision.
+