Market State Lab
SPY internal divergence
SPY versus RSP, QQQ, breadth, sectors, VIX, cross-assets and one- through ten-session lead-lag tests.
Shift Forge · Build in public
The current edge is not prediction.
It is honest validation.
The current fitted signal is not authorized as a trading edge. The public research feed records the gap between in-sample fit and out-of-sample usefulness.
Model ruling—
In-sample correlation—
Out-of-sample correlation—
Current mode—
Activation gate—
Updated—
Signal scoreboard
Every theory receives a ruling
| Signal | Status | OOS result | Baseline improvement | Decision |
|---|---|---|---|---|
| Loading… | ||||
Pre-registration
Acceptance before results
Target, holding period, costs, split dates, primary metric and rejection threshold are committed before untouched testing.
Ablation
Simplify first
Regime, persistence, dominance, contradictions, event direction and liquidity normalization are tested alone and in pairs.
A failed theory is a published result. A beautiful in-sample chart is not a production signal.