QuantAlgo PropFirms

An Opening-Range-Breakout (ORB) trading engine engineered for prop-firm funded-account challenges.

---

Why this exists

Prop-firm challenges (Topstep, Apex, etc.) have a convex payoff: your downside is capped at the challenge fee, but a pass unlocks a funded account with uncapped upside. That structure rewards a strategy with a high win-rate and bounded daily risk — exactly what a disciplined Opening Range Breakout delivers. This project turns that idea into a tested, reproducible engine you can backtest, stress-test with Monte-Carlo, and run on a paper account.

⚠️ Disclaimer. This is research and educational software, not financial advice. Trading futures involves substantial risk of loss. Past or simulated performance does not guarantee future results. Use the paper/sandbox mode until you understand the risks.

Features

• 📈 ORB strategy engine — opening-range detection, breakout entries, fixed-multiple TP/SL, and a volatility filter (quantalgo.strategy).
• 🧪 Event-driven backtester — correct position sizing with Topstep-style daily loss limit, trailing drawdown, and profit target enforcement (quantalgo.backtest).
• 🎲 Monte-Carlo pass-rate simulator — bootstrap thousands of challenge attempts to estimate pass rate, expected value, and P&L percentiles (quantalgo.montecarlo).
• 🔌 Pluggable execution — in-memory PaperBroker for testing plus an async TradovateBroker wrapper for Topstep's venue (quantalgo.execution).
• 🛰️ Offline-first data — Yahoo Finance loader with Parquet caching and a deterministic synthetic generator, so everything runs without a network (quantalgo.data).
• 🧰 Batteries included — typed config, a quantalgo CLI, text reports & charts, a pytest suite, and CI across Python 3.10–3.12.

Architecture

                ┌──────────────┐      ┌──────────────┐
   data ───────▶│   strategy   │─────▶│   backtest   │─────▶ metrics / equity curve
 (yfinance /    │  ORB signals │      │ sizing+risk  │             │
  synthetic)    └──────────────┘      └──────┬───────┘             ▼
                                             │              ┌──────────────┐
                                             └─ trade P&L ─▶│ montecarlo   │─▶ pass rate / EV
                                                            └──────────────┘
   execution (PaperBroker / TradovateBroker)  ◀─── live & paper sessions
   reporting (tables + matplotlib charts) · config (typed, env-overridable)

Installation

git clone https://github.com/tylrcc/quantalgo-propfirms.git
cd quantalgo-propfirms

python -m venv .venv && source .venv/bin/activate
pip install -e ".[all]"     # runtime + data + plots + live + env extras

Minimal install (pip install -e .) only needs pandas and numpy; live data, charts and broker connectivity are optional extras (data, plots, live, env).

Quick start

quantalgo info                                   # config + supported symbols
quantalgo backtest   --symbol MES --days 180 -p  # backtest, save equity_curve.png
quantalgo montecarlo --symbol MES --sims 10000   # estimate challenge pass-rate

Example output:

──────────────────────────────────────────
 Backtest — MES (2025-12-06 → 2026-06-04)
──────────────────────────────────────────
 Total Trades            42
 Win Rate            64.29%
 Profit Factor         1.83
 Expectancy R        0.2611
 Total Return Pct     5.74%
 Final Equity    $52,870.00
 Max Drawdown     $1,420.00
──────────────────────────────────────────

Challenge outcome: PASSED  •  42 trades

Library usage

from quantalgo.config import get_settings
from quantalgo.data import DataLoader
from quantalgo.strategy import ORBStrategy
from quantalgo.backtest import Backtester
from quantalgo.montecarlo import MonteCarloSimulator

settings = get_settings().with_symbol("MES")
df = DataLoader("MES").load("2025-12-01", "2026-06-01", interval="5m")

trades = ORBStrategy(settings.orb).generate_trades(df)
result = Backtester(settings.challenge, settings.symbol_spec).run(trades)
print(result.metrics, result.outcome)

mc = MonteCarloSimulator(settings.challenge, settings.montecarlo)
print(mc.run([t.pnl_dollars for t in result.trades]).as_dict())

Configuration

All defaults work out of the box; override via environment variables or a .env file (see .env.example). Strategy, risk and simulation parameters are typed dataclasses in quantalgo.config:

Group	Examples
ORBParams	opening_range_minutes, take_profit_mult, stop_loss_mult
ChallengeRules	daily_loss_limit, trailing_max_drawdown, profit_target
MonteCarloParams	n_simulations, max_trading_days, seed

Development

make dev      # install with dev + all extras
make test     # pytest
make cover    # pytest with coverage
make lint     # ruff

CI runs lint, the full test-suite, and CLI smoke-tests on Python 3.10/3.11/3.12.

Project layout

src/quantalgo/   config · data · strategy · backtest · montecarlo · execution · reporting · cli
tests/           pytest suite (runs fully offline)
.github/         CI workflow

License

MIT © tylrcc