SystematicStrategies

Status: Active | Last updated: 2026-06 | Supersedes Crypto_Quant_Trading (deleted)

Crypto strategy research and paper trading on Binance spot/perp data.

Setup

python -m venv .venv
source .venv/bin/activate
pip install -e ".[research]"

Optional ML stack (frozen module): pip install -e ".[prediction]"

Data

Market data lives in data/ (gitignored, ~500 MB+). Download after clone:

# Spot klines
.venv/bin/python scripts/run_download_and_save_data.py

# Perp klines + funding rates
.venv/bin/python scripts/run_download_perp_data.py

Load in code: from src.data.data_load import load_data

Quick start

# Classical walk-forward (gate-passing strategies only)
.venv/bin/python scripts/run_classical_backtest.py --gate-passing-only

# Stat-arb OOS validation
.venv/bin/python scripts/run_stat_arb.py walkforward --pair BTCUSDT/ETHUSDT --interval 15m

# Paper trade signal (no orders)
.venv/bin/python scripts/run_paper_trading.py --dry-run

# Stat-arb paper trade (BTC/SOL + XRP/ADA @ 4h) — see .env.example for demo API keys
cp .env.example .env   # keys via www.binance.com → Futures → Demo/Mock Trading → API Management
.venv/bin/python scripts/run_stat_arb_paper_trading.py --dry-run
.venv/bin/python scripts/run_stat_arb_paper_trading.py

Outputs go to results/*.csv.

Documentation

File	Purpose
ARCHITECTURE.md	Folder layout, data flow, scripts, cost model
STRATEGY_STATUS.md	Active vs frozen strategies, latest backtest gates
docs/PREDICTIVE_ML.md	Frozen ML direction-forecast module (reference only)
docs/MIGRATION.md	Consolidated legacy repos map
docs/REPO_INVENTORY.md	Local/GitHub repo audit and tags
docs/DATA_PATHS.md	External dataset locations (~/Data/)
legacy/	Migrated forecast code (PriceForecast, orderbook, etc.)

Current focus

Live candidate: dual_ma on BTCUSDT 1h (OOS Sharpe 1.07, stress 0.41). See STRATEGY_STATUS.md.