GovGreed Home Bill Pass Index Methodology For Institutions Signals
Home / How it works

The GovGreed Intelligence Stack

Three layers: approved direct connections to federal data systems, licensed private market feeds, and proprietary machine-learning enrichment. Here's how the model scores bills and trades, and what a signal does and does not tell you.

See live predictions → Read the methodology
The Intelligence System

Seven sources. One score.
Scored as filings land.

How we turn public congressional data into a score for every disclosed trade. Data as of 2026-09-25.

99K
Deduplicated trades
45d
STOCK Act filing window
353
Members trading
42K
Bills scored
01
The Model
Seven layers. One prediction.
Each layer adds signal weight. When 3+ layers align on the same ticker, the score multiplies — not adds. Convergence is everything.
Signal Architecture — Four Signal Categories
Congressional Behavior40%
Politician Track Record
Win rate, committee tenure, trade timing history
20%
Herd Detection
3+ politicians converging on the same ticker
20%
Legislative Timing16%
Bill Timing + Committee
Markup proximity, committee seat alignment
16%
Market Signals24%
Technical Indicators
RSI, volume anomalies, momentum
12%
Sector Momentum
Industry flow, sector rotation signals
12%
Influence Patterns20%
Campaign Contributions
Industry PAC money → trade correlation
10%
Lobbying Patterns
Filing cadence before vote windows
10%
Tier Thresholds
SSupreme≥ 85
A+High Conviction≥ 72
AStrong≥ 60
BModerate≥ 48
Convergence Multipliers
3 layers aligned1.3×
4 layers aligned1.5×
5+ layers aligned2.0×
Each layer alone is circumstantial. All together create an exponentially stronger signal — the pattern that preceded every major congressional trading scandal we've tracked.
01b
In Practice
How a real signal scores.
A+ TIER 71.2
DLHC · DLH Holdings Corp · McCaul (Armed Services) · $250K+ purchase
20%
Politician — McCaul: repeat buyer in the sector
+16.4
20%
Herd — No convergence signal yet (0 other politicians)
+0
16%
Bill Timing — DHP markup 12 days out · committee seat confirmed
+13.1
12%
Technical — Volume +180% · RSI 68 · defense sector breakout
+9.6
12%
Sector — Defense momentum +2.3 std deviations above 30d avg
+10.1
10%
Contributions — $40K defense PAC money from DHP contractors
+8.2
10%
Lobbying — 2 defense contractors filed in the past 30 days
+7.8
Raw score
54.8
× 1.3× convergence
politician + bill + sector + contrib
Final score
71.2 → A+
Key Insights — Why DLHC Scored A+
🔁
Repeat Behavior Pattern
Illustrative example: a repeat buyer in a sector their committee work touches scores higher on the politician layer than a first-time buyer.
📋
Legislative Catalyst Imminent
DHP markup scheduled 12 days out. McCaul sits on the committee — buys before markups are his signature pattern.
💰
Follow-the-Money Confirmation
$40K PAC money from defense contractors + 2 lobbying filings in 30 days. Money trail confirms the legislative angle.
⚡
4-Layer Convergence
Politician + Bill + Sector + Contributions all firing simultaneously. 4 layers = 1.3x multiplier, pushing raw 54.8 to final 71.2.
Validation
02
Checking the model
What a signal can and can't tell you.
Every signal is timestamped when it is scored, so it can be compared with what members later disclose.
A signal is a score, not a forecast of returns. It says where a disclosed trade lines up with committee seats, herd activity, bill timing, lobbying, and donor money. Congress discloses trades in dollar bands, up to 45 days late, so no model can say what a member paid or whether a position is still held.

We publish tier-level backtest results, with sample sizes and limitations, on the methodology page rather than headline hit rates here. Past results do not predict future returns. Not financial advice.
03
Signal Examples
What a live signal looks like.
Full reasoning chain — tier, trader, which layers fired, and the dark window countdown before the filing window closes.
A GovGreed signal is not just a trade notification. It's a reasoning chain — connecting who traded, what legislation they oversee, who's lobbying them, and whether the market technicals confirm the thesis. Here's how a signal moves from raw data to actionable alert.
Signal Generation Pipeline
01 — Ingest
8 Federal Data Feeds
STOCK Act · Bills · Committees · Lobbying (LDA) · FEC · SEC Form 4 · Congress.gov · FMP
→
02 — Score
7-Layer Engine
Every trade scored across politician track record, herd detection, bill timing, technicals, sectors, PAC money, and lobbying cadence
→
03 — Converge
Convergence Multiplier
3 layers = 1.3× · 4 = 1.5× · 5+ = 2.0×. Raw scores multiply when signals agree. One convergence beats scattered noise
→
04 — Alert
Signal Fires
S/A+ signals delivered before the STOCK Act window closes. Ticker, thesis, entry context, committee timing included
Live signals carry the member, the disclosed band, the layers that fired, and the committee calendar behind them. The dashboard shows today's real signals, each tied to a specific STOCK Act disclosure.
See live signals with full reasoning chains
Updated daily as congressional trades are filed. Free account, no card required.

Want to see it in action?

The model is live and rescores bills daily as they move through Congress.

Live Whale Table → Founder Access