What is Goosos?
Goosos is a quantitative trading lab. We write about systematic trading the way it should be taught: with code you can run, backtests you can reproduce, and methods that survive contact with real markets.
Too much trading content online is either vague theory or cherry-picked backtests with no code attached. We do the opposite. Every strategy article on Goosos ships with runnable Python code, and every performance claim comes with the assumptions spelled out — data source, costs, slippage, and all.
What we cover
- Tutorials — the modern Python backtesting stack: vectorbt, backtrader, NautilusTrader. Runnable code, zero fluff.
- Methodology — walk-forward validation, probability of backtest overfitting (PBO), slippage modeling. The anti-overfitting handbook.
- Tools — position sizers, Kelly calculators, broker API and market data comparisons. Built for practitioners.
Our approach
Three rules guide everything we publish:
- Code first. If we can’t reproduce it, we don’t publish it.
- No hype. We don’t promise returns, and we don’t sell dreams. Markets are hard; anyone telling you otherwise is selling something.
- Show the ugly parts. Transaction costs, survivorship bias, lookahead bias — the things that kill strategies in production get covered, not hidden.
Freshness promise
Quantitative libraries change fast, and stale tutorials are worse than no tutorials. Here is our commitment:
- Every tutorial carries a tested badge — the exact library versions we verified, and the date we last ran the code.
- When a library ships a breaking change, we re-test the affected tutorials and update the verification date.
- If a tutorial can no longer be made to work, we say so on the page — we do not leave rotting code up silently.
The goal is simple: every article you read on Goosos has code that runs today, not three years ago.
Who runs Goosos?
Goosos is an independent project run by a developer and systematic trading practitioner who builds mid-frequency Python trading systems. The site exists because the tutorials we wished existed didn’t — so we wrote them.