Skip to content

README.md

Latest commit

 

History

History
91 lines (66 loc) · 3.27 KB

README.md

File metadata and controls

91 lines (66 loc) · 3.27 KB

Backtest Strategy Examples

Historical trading simulations with forward PnL analysis. Each script simulates a multi-leg trade at a specific historical date and shows what the PnL would have been at +1, +2, +4, and +7 days.

Available Scenarios

Script Date Strategy Assets Outcome
covid_crash_hedge.py 2020-02-21 Flight-to-safety Long GOLD + Short SP500 SP500 fell 12% in 7 days; short leg dominated
ftx_crypto_contagion.py 2022-11-08 Crypto contagion hedge Short BTC + Short ETH + Long GOLD BTC -15%, ETH -18%; crypto shorts drove profit
nvda_earnings_alpha.py 2023-05-25 Sector alpha pair Long NVDA + Short SP500 NVDA +5% while SP500 flat; pure alpha
ukraine_oil_shock.py 2022-02-24 Commodity supply shock Long OIL + Long GOLD + Short SP500 Oil surged 16% in 7 days on sanctions

Setup

No API keys or wallet configuration needed. Just build the backtest binary:

cargo build --release

Usage

# Run any scenario
python3 examples/backtest/covid_crash_hedge.py
python3 examples/backtest/ftx_crypto_contagion.py
python3 examples/backtest/nvda_earnings_alpha.py
python3 examples/backtest/ukraine_oil_shock.py

# Override binary path
python3 examples/backtest/covid_crash_hedge.py --backtest /path/to/backtest

Each script:

  1. Resets the portfolio to a clean state
  2. Fetches historical prices at the scenario date
  3. Executes simulated trades (spot buy/sell)
  4. Displays a PnL table at +1, +2, +4, +7 days for each leg
  5. Shows the final portfolio summary (cash balance, positions)
  6. Cleans up (resets portfolio)

How It Works

The backtest CLI fetches historical OHLCV data from Yahoo Finance (with CoinGecko fallback for crypto) and computes forward PnL by looking up actual prices at future dates.

  • Auto-pricing: If --price is omitted, the historical close price is used
  • Portfolio state: Trades are tracked in ~/.fintool/backtest_portfolio.json
  • Cash balance: Buying subtracts cost, selling adds proceeds (spot trades only)
  • Perp trades: Use perp_buy/perp_sell with leverage for leveraged PnL

Dependencies

  • Python 3.10+ (no third-party packages — uses only stdlib)
  • backtest CLI binary (compiled from this repo)

Writing Your Own

Create a new Python script following this pattern:

#!/usr/bin/env python3
import json, os, subprocess
from pathlib import Path

REPO_DIR = Path(__file__).resolve().parent.parent.parent
BT = os.environ.get("BACKTEST", str(REPO_DIR / "target" / "release" / "backtest"))
DATE = "2024-01-15"

def cli(cmd, date=DATE):
    r = subprocess.run([BT, "--at", date, "--json", json.dumps(cmd)],
                       capture_output=True, text=True, timeout=30)
    return json.loads(r.stdout)

# Reset
cli({"command": "reset"})

# Quote
btc = cli({"command": "quote", "symbol": "BTC"})
print(f"BTC on {DATE}: ${btc['price']}")

# Trade
result = cli({"command": "buy", "symbol": "BTC", "amount": 0.01})
for p in result["pnl"]:
    print(f"  {p['offset']}: ${p['price']} (PnL: {p['pnl']}, {p['pnlPct']}%)")

# Cleanup
cli({"command": "reset"})

Supported symbols include all major crypto (BTC, ETH, SOL, ...), stocks (AAPL, NVDA, TSLA, ...), commodities (GOLD, SILVER, OIL), and indices (SP500, NASDAQ, DOW).