SCALP Quant Engine — Decision System (DEC)

This repository contains the decision engine (dec_writer) of the SCALP quantitative trading system.

The DEC layer aggregates dozens of market signals and produces a final normalized meta-score used by the execution engine to trigger trades.

The system is designed to detect micro-breakouts, liquidity events, volatility shocks, and smart money activity across the crypto market.

ARCHITECTURE OVERVIEW

Market data flows through several layers of analysis before producing a final trade signal.

Market Data ↓ Signal Engines ↓ Breakout / Liquidity / Volatility Detection ↓ Cluster & Regime Analysis ↓ Meta Signal Engine ↓ Meta Score Normalization ↓ Execution Engine

The system runs continuously inside the dec_writer loop.

CORE COMPONENT

dec_writer.py

The dec_writer is the central orchestration engine.

Responsibilities:

Load market context
Refresh ATR and volatility
Update range structure
Run signal engines
Compute breakout energy
Generate cross-market rankings
Produce final meta signals
Store signals for historical analysis

BREAKOUT DETECTION

Micro Breakout Engine

Detects short-term breakouts relative to ATR and range structure.

Signals:

MICRO_BREAKOUT_UP
MICRO_BREAKOUT_DOWN

BREAKOUT ENERGY ENGINE

Combines multiple signals to estimate breakout strength.

Inputs:

volatility
compression
orderflow
liquidity distance
ATR expansion

Output:

breakout_energy

CROSS MARKET RANKING

Ranks assets based on breakout strength across the entire market.

View:

v_cross_section_rank

Used to identify market leaders.

LIQUIDITY MAP ENGINE

Detects proximity to liquidity clusters.

Distances calculated against:

high_20
high_50
high_100
low_20
low_50
low_100

View:

v_liquidity_map

LIQUIDITY HEATMAP

Measures proximity to liquidation zones.

Signals:

UPSIDE_NEAR
DOWNSIDE_NEAR
FAR

LIQUIDITY VACUUM ENGINE

Detects zones where liquidity is thin.

Signals:

UPSIDE_VACUUM
DOWNSIDE_VACUUM

These zones often produce rapid price movement.

LIQUIDITY CASCADE DETECTOR

Detects chain reactions of liquidations.

Signals:

CASCADE_UP
CASCADE_DOWN

LIQUIDITY MAGNET ENGINE

Detects attraction toward major moving averages.

Uses:

MA50
MA100
MA200

Signal example:

MAGNET_200

VOLATILITY ENGINES

Compression Engine

Measures volatility contraction before breakouts.

Metrics:

compression
volatility

VOLATILITY SHOCK DETECTOR

Detects sudden volatility expansion.

Signals:

VOL_SHOCK
STRONG_EXPANSION

WHALE FOOTPRINT DETECTOR

Detects abnormal volatility suggesting large player activity.

Signal:

WHALE_ACTIVITY

ORDERFLOW ENGINE

Simplified orderflow scoring.

Signals:

BUY_PRESSURE
SELL_PRESSURE
BALANCED

Quantized version:

0
0.5
1.0

SMART MONEY TRAP DETECTOR

Detects classic market maker traps.

Signals:

BULL_TRAP
BEAR_TRAP

MARKET MAKER FOOTPRINT

Detects accumulation or distribution phases.

Signals:

ACCUMULATION
DISTRIBUTION

PREDICTIVE BREAKOUT ENGINE

Predicts imminent breakouts using structural pressure.

Signals:

BREAKOUT_IMMINENT
BUILDUP
NONE

SIGNAL PERSISTENCE ENGINE

Stores signals over time to detect persistent pressure.

Table:

signal_history

View:

v_signal_persistence

SIGNAL HALF LIFE ENGINE

Measures signal freshness.

View:

v_signal_half_life

Used to prevent acting on stale signals.

CLUSTER ENGINE

Measures the number of simultaneous breakouts across the market.

Table:

cluster_history

Signals:

MARKET_EXPLOSION
MOMENTUM_BUILD
STABLE

SECTOR ROTATION ENGINE

Detects capital rotation across sectors.

Sectors:

DEFI
L1
MEME
MAJORS
AI

MARKET LEADER ENGINE

Detects assets leading the market.

Signals:

STRONG_LEADERS
WEAK_LEADERS

META SIGNAL ENGINE

All signals are aggregated into a meta score.

meta_score

Final normalized score:

meta_score_norm

Range:

0 → 1

EXECUTION TRIGGER

Execution engine reads:

v_meta_rank_norm

Example:

instId | meta_score_norm

ONDO | 1.00
HBAR | 0.85
PNUT | 0.77

Suggested thresholds:

0.80 strong trade
0.70 normal trade
0.60 scalp

DATABASE

All decision data stored in:

/project/data/dec.db

Key tables:

snap_ctx
snap_range
snap_range_ext
snap_orderflow
signal_history
cluster_history

SYSTEMD SERVICE

Engine runs continuously via systemd.

Service:

dec_writer.service

Update frequency:

2 seconds

LOGGING

Logs stored in:

/project/logs/dec_writer.log

Example log:

[SIGNAL_HISTORY] rows=19
[UPDATE] ctx=66 snap=3 veto=63

FINAL OUTPUT

Final ranking used by execution engine:

v_meta_rank_norm

Example:

1 ONDO 1.00
2 HBAR 0.85
3 PNUT 0.78

SYSTEM PHILOSOPHY

The system uses a multi-signal architecture combining:

liquidity
volatility
orderflow
cross-market behavior
sector rotation
structural breakouts

No single indicator drives the system.

The objective is to detect high-probability market dislocations.

STATUS

Current system includes:

20+ signal engines
cross-market ranking
liquidity heatmap
sector rotation detection
predictive breakout engine
adaptive risk engine

This forms a fully autonomous signal generation system.

NEXT STEP

Execution engine integration:

signal → trade execution → recorder → performance analytics