chinloong0

chinloong0 / Strategy-Factory

Public

Strategy research pipeline with WFO and permutation testing

11
3
100% credibility
Found Apr 11, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

An open-source toolkit for rigorously validating cryptocurrency trading strategies through optimization, permutation testing, and walk-forward simulations to distinguish real edges from overfitting.

How It Works

1
🔍 Discover Strategy Factory

You stumble upon this open project on GitHub that helps everyday traders test ideas without falling for fake wins.

2
🛠️ Set up your playground

You create a cozy space on your computer where everything runs smoothly with a few simple preparations.

3
📥 Grab price histories

You pull in real past charts for coins like Bitcoin and Ethereum to fuel your tests.

4
🚀 Tune your first idea

You pick a simple trend idea like moving average direction and watch it smartly find the strongest settings.

5
🧪 Stress-test for reality

You shuffle the data randomly and simulate future rolls to prove if your idea beats chance.

6
📊 Rank your winners

You create colorful leaderboards showing which ideas survive every challenge.

🎉 Gain trading confidence

You walk away with battle-tested strategies that shine in real markets, ready to use.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Strategy-Factory?

Strategy-Factory is a Python-based research pipeline for quantitative traders that runs trading strategies through a cascade strategy factory of validation stages: in-sample optimization, permutation testing, signal transfer, walk-forward optimization, and holdout verification. It downloads Binance futures data into fixed vaults, lets you plug in modular strategies like EMA slope or ADX regime, and outputs JSON reports, charts, and leaderboards viewable in a standalone HTML viewer. Developers get a CLI-driven workflow to fail fast on fragile edges, using Optuna for bayesian tuning and Polars for fast data handling.

Why is it gaining traction?

Unlike basic backtesters, it enforces a strategy factory pattern with permutation matrices to test against synthetic noise, plus github strategy matrix parallel jobs for speed. Dynamic fusion combines strategies into ensembles without custom code, and batch scripts like mass_optuna scan combos across timeframes. The hook is rejecting overfit ideas early via WFO and holdouts, with leaderboard generators for quick research analyst strategy ranking.

Who should use this?

Algo traders prototyping crypto perps on 1m-1h data, quant researchers needing permutation-tested WFO pipelines, or teams iterating strategy github workflows. Ideal for those tired of manual backtest scripting and wanting a factory strategy to scale idea validation.

Verdict

Worth forking for its teaching value on robust quant methods (stars: 11), but 1.0% credibility score signals alpha maturity—docs are solid via README quickstarts, though test coverage and private data exclusion limit plug-and-play. Try for methodology, not production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.