Introducing Tobacco Plant: An Automated Portfolio Manager (APM) That Learns How To Allocate Capital
How a Desert Shrub Solved Portfolio Theory Before Markowitz Was Born
Important: There is no guarantee that ATS strategies will have the same performance in the future. I use backtests and forward tests to compare historical strategy performance. Backtests are based on historical data, not real-time data so the results shared are hypothetical, not real. Forward tests are based on live data, however, they use a simulated account. Any success I have with live trading is untypical. Trading futures is extremely risky. You should only use risk capital to fund live futures accounts and if you do trade live, be prepared to lose your entire account. There are no guarantees that any performance you see here will continue in the future. I recommend using ATS strategies in simulated trading until you/we find the holy grail of trading strategy. This is strictly for learning purposes.
I’m on a quest for the holy grail of automated trading. Questions? Check the FAQs or feel free to reach out directly: AutomatedTradingStrategies@protonmail.com.
A Parable of Risk Management
The wild tobacco plant—Nicotiana attenuata—has been running the same algorithmic trading strategy for roughly 7 million years. Like any veteran trader, she’s learned that the market giveth, and the market taketh away.
Every evening around dusk, somewhere in the Utah desert, a tobacco plant starts pumping out something called benzyl acetone. This is the botanical equivalent of ringing the opening bell at the NYSE. It’s a chemical perfume so intoxicating to hawkmoths that they’ll fly miles through pitch-black desert to find the source.
The hawkmoths arrive like traders flooding into the pit. They’re big. They’re fast. They can carry pollen across vast distances, introducing genetic diversity that keeps the tobacco plant’s lineage robust for generations. They are, in a word, alpha.
But here’s what the tobacco plant learned a few million years ago (probably right around the time the first one got eaten alive): hawkmoths come with strings attached.
While the male hawkmoth is busy doing his pollination work—converting nectar into genetic diversification—the female hawkmoth is looking at those nice, broad tobacco leaves and thinking one thing: baby food.
She lands. She deposits eggs. She leaves.
Within days those eggs hatch into Manduca sexta larvae, aka the hornworm caterpillar seen below.
They are, pound for pound, among the most efficient defoliating machines ever produced by natural selection. We’re talking about a creature that exists solely to eat. It doesn’t sleep. It doesn’t mate. It just converts leaf matter into caterpillar and it does so at an exponential rate. For the first two weeks, the damage is manageable, but once it hits the final growth stage, it consumes 90% of its total diet in a final binge. At that stage, a single hornworm can strip a tobacco plant to the stem in under 72 hours.
So let’s review the trade position:
Trade Entered: Premium pollination services (genetic alpha)
Counterparty Risk: Total account liquidation.
Risk Profile: Exponential. Manageable drawdown for 2 weeks, followed by 100% loss in the final 72 hours.
The tobacco plant is running the oldest trade in nature. Another organism —your fern, simple grass or average retail trader—would simply accept this fate. It’s the circle of life or the cost of doing business. Ashes to ashes, and all that jazz. Not the tobacco plant. When those caterpillar mandibles first break the leaf surface and that distinctive hornworm saliva hits the plant tissue, a chemical cascade begins. Within hours of detecting hornworm damage, the wild tobacco plant executes a complete strategic pivot.
The perfume that summoned moths from miles away? Gone.
Flowers that once opened at dusk now open at dawn.
Nectar composition changes.
The flower tubes reshape. Become narrower.
Same flowers. Same plant. Completely different operating hours and optimized for a completely different kind of visitor.
All of this happens autonomously. No committee meetings. No board approval. The plant reads the signal and reacts.
Morning comes. The newly restructured flowers open into the desert sunlight, and a different kind of trader appears. Hummingbirds.
Hummingbirds are the Treasury bonds of the pollination world. They’re safe and reliable. They never—EVER—lay eggs on your leaves.
There is a trade-off for this risk-free pollination however…
Hummingbirds are locals. They tend to visit multiple flowers on the same plant before moving on. This leads to self-pollination—which works, technically, but produces offspring with less genetic diversity. It’s the reproductive equivalent of index funds.
The Best Risk Manager Alive
Here’s what makes the tobacco plant the best risk manager alive: she doesn’t commit. When the hornworm threat passes and the caterpillars have pupated or been picked off by predators, the plant can start pumping benzyl acetone again.
Court those moths baby!
The tobacco plant has been running this adaptive algorithm through ice ages and droughts, through extinction events and continental shifts. She is, in every way that matters, an algorithmic portfolio manager.
Seven million years of backtesting. Exposed to every market condition evolution could throw at her. The tobacco plant arrived at the same conclusion that keeps showing up in every serious study of systematic trading: The best strategy is the one that doesn’t get you killed. It’s the one that can adapt based on performance. It welcomes managed loss in exchange for unlimited alpha.
The question isn’t which strategy is best?
The question is: How do I automate the selection so that capital is allocated to strategies that are currently performing, while protecting myself from strategies that are bleeding out?
That is the goal of the Tobacco Plant APM.
How The Tobacco Plant APM Works
Before I explain what the Tobacco Plant APM does, let me explain what an APM actually is because this is the first one I’m releasing to subscribers.
An Automated Portfolio Manager sits above your strategies. It monitors their performance in real-time and makes decisions about capital allocation based on objective metrics.
Think of it as a manager who watches your team of traders and decides who gets more capital and who gets benched. Except it’s completely automated. The APM maintains its own account where it executes trades. Your individual strategies run separately, and the APM decides which ones have earned the right to influence your portfolio.
I spent years searching for something like this on the web, but I couldn’t find a live, real-time portfolio manager for NinjaTrader that automatically allocates capital based on rolling performance. When I reached out to NT8 they said it could not be done with code. At first I thought they might be right, but then I had an architectural epiphany. Four months later, here we are. My apologies for all the delays, but I think you’ll agree it was worth it. The feedback I’ve received so far is excellent, especially for an APM called The Ultimate Bootstrap, but I’m getting ahead of myself.
It’s important to note that the Tobacco Plant APM is strategy-agnostic. It doesn’t care what logic your strategies use— trend following, mean reversion, scalping, breakouts, machine learning, whatever. If your strategy runs on a SimAccount and generates trades, Tobacco Plant can trade it.
This means you can:
Use your own custom strategies
Use strategies from other vendors
Use ATS Strategies (I provide a list of suggestions below)
Mix and match all of the above
The APM only cares about one thing: Is this strategy following the PM logic?