Daily macro signals generated from US Treasury fiscal flow data, institutional positioning, and technical regime analysis. Every signal is logged, every call is tracked.
Our signals are built from six independent data layers. Each measures a different dimension of liquidity, positioning, or market structure. The composite score combines them into a single daily reading.
Net government spending vs. receipts. When the Treasury spends more than it collects, reserves flow into the private sector (supportive). When it collects more, reserves drain (restrictive).
The Treasury General Account at the Fed. A rising TGA drains reserves from the banking system; a falling TGA adds them back. Direction and rate of change matter more than level.
Commitment of Traders data across 5 contracts: S&P 500, 10-Year, 2-Year, USD, and VIX. Z-scores flag extreme positioning that historically precedes reversals.
SPX trend, breadth, and momentum indicators combined into a bullish/bearish regime classification. Provides the market structure context that fiscal flows operate within.
FINRA margin debt levels, momentum, and net balance. Extreme leverage readings historically cluster before drawdowns. Monthly data, z-scored for signal generation.
Weighted combination of all layers into a single score from -100 (maximum caution) to +100 (maximum opportunity). Includes ETF sector tilt recommendations based on the regime.
The pipeline runs automatically every evening after the Treasury publishes the Daily Treasury Statement. No human discretion in signal generation; the rules are fixed.
Treasury FiscalData API (3 endpoints), CFTC Socrata API (weekly), FINRA margin statistics (monthly), FRED market data (daily). All fetched automatically.
1,800+ lines of deterministic Python code. Computes net fiscal injection, regime classification, z-scores, momentum, and cross-asset transmission maps. No ML, no black boxes.
Each day's reading is appended to an immutable log. Regime, momentum, TGA trend, COT z-scores, tech regime, margin signal, composite score. All timestamped.
Forward SPX returns (5, 10, 20 day) are computed against each logged signal. Hit rates are updated nightly. No cherry-picking; every signal is scored.
Forward SPX returns measured at 5, 10, and 20 trading days after each signal. Hit rate = percentage of signals where the market moved in the predicted direction.
The gauge reads the macro environment, not market direction. Green = conditions favor risk. Yellow = neutral/mixed. Red = conditions favor caution. When the gauge turns red before a drawdown, the system is working.
The last 60 trading days of logged signals. Every day the pipeline runs, one row is added. Full history available in the daily Substack briefing.
| Date | Regime | Momentum | TGA ($bn) | Tech | Composite | Top Overweight |
|---|---|---|---|---|---|---|
| Loading signals... | ||||||
Disclaimer: This is an analytical research tool, not investment advice. Past signal performance does not guarantee future results. The fiscal flow framework provides a lens for understanding reserve mechanics and their transmission to financial markets. All signals are generated algorithmically from public data sources (US Treasury, CFTC, FINRA, FRED). No proprietary market data is used. Position sizing, entry timing, and risk management are your responsibility.