XAUZenbotTreeBot · Tree-based ML — strategy & live paper-trading performance | Research

Q: "Is XAUZenbotTreeBot profitable live?"

"Since deployment, XAUZenbotTreeBot has produced an annualised return (CAGR) of 5.03% on a $10,000 paper seed (2.63% cumulative), with a Sharpe ratio of 0.55 and a max drawdown of -4.52% over 126 trading days. Numbers refresh daily on the [live page](/strategies/xauzenbottreebot/)."

Q: "Can I run XAUZenbotTreeBot myself?"

"Yes. XAUZenbotTreeBot is part of the open-source [python_tradingbot_framework](https://github.com/JustinGuese/python_tradingbot_framework) — you can fork the repo, inspect the strategy, and run the code locally or on your own Kubernetes cluster. The exact source file is [here](https://github.com/JustinGuese/python_tradingbot_framework/blob/main/tradingbot/xauzenbot.py)."

Q: "How does XAUZenbotTreeBot compare to other quantitative strategies?"

"XAUZenbotTreeBot belongs to the **Tree-based ML** family. Gradient-boosted and random-forest models trading FX and metals on engineered features. Browse the [full leaderboard](/) to compare it against strategies from other families."

XAUZenbotTreeBot · Tree-based ML — strategy & live paper-trading performance

23 June 2026

AIB Strategies and DataFortress.cloud are not giving any trading advice. Statistics and opinions stated on this website are purely statistical, and should never be used for real life trading. Form your own decisions and speak with a registered investment advisor before investing.

XAUZenbotTreeBot is a production algorithmic trading strategy paper-traded live from a $10,000 seed. Live paper-trading performance: 5.03% annualised return (2.63% total), Sharpe 0.55 over 126 days.

TL;DR

Strategy family: Tree-based ML — Gradient-boosted and random-forest models trading FX and metals on engineered features.
Live track record: 5.03% annualised return (CAGR), Sharpe 0.55, max drawdown -4.52% over 126 trading days.
Code: open-source on python_tradingbot_framework — tradingbot/xauzenbot.py.

Live performance snapshot

Annualised return (CAGR): 5.03%
Current paper-portfolio worth: $10262.96 (seeded at $10,000)
Total return since seed: 2.63%
Sharpe: 0.55
Max drawdown: -4.52%
Days live: 126

See the full chart, current holdings, and historical backtest on the strategy’s live page →.

How XAUZenbotTreeBot works

A full technical writeup for XAUZenbotTreeBot is in progress. In the meantime:

Source code: tradingbot/xauzenbot.py.
Open-source framework: python_tradingbot_framework.

Run it yourself

XAUZenbotTreeBot is part of the open-source python_tradingbot_framework — fork the repo, run the code locally, or deploy your own variant. The source lives at tradingbot/xauzenbot.py. The live page shows today’s portfolio state alongside the full historical backtest.

EURUSDTreeBot — also in the Tree-based ML family. See live performance →
XAUSyntheticMetalTreeBot — also in the Tree-based ML family. See live performance →

FAQ

Is XAUZenbotTreeBot profitable live?

Since deployment, XAUZenbotTreeBot has produced an annualised return (CAGR) of 5.03% on a $10,000 paper seed (2.63% cumulative), with a Sharpe ratio of 0.55 and a max drawdown of -4.52% over 126 trading days. Numbers refresh daily on the live page.

Can I run XAUZenbotTreeBot myself?

Yes. XAUZenbotTreeBot is part of the open-source python_tradingbot_framework — you can fork the repo, inspect the strategy, and run the code locally or on your own Kubernetes cluster. The exact source file is here.

How does XAUZenbotTreeBot compare to other quantitative strategies?

XAUZenbotTreeBot belongs to the Tree-based ML family. Gradient-boosted and random-forest models trading FX and metals on engineered features. Browse the full leaderboard to compare it against strategies from other families.