Napoleon Hill had “Think and Grow Rich.” For algorithmic traders, the equivalent is simpler: automate to grow rich.
That’s Tomas Nesnidal’s philosophy, and it’s not just a catchy phrase. When you can automate strategy development, robustness testing, pre-selection, and portfolio drafting, you free up the mental bandwidth to actually trade – and to keep improving. Tomas spent four and a half months travelling across Asia – Malaysia, Myanmar, Vietnam, Cambodia – while his hedge fund kept running. Not because he was on holiday, but because the majority of the development work was automated. He still worked, sometimes eight hours a day. But he had flexibility.
That flexibility is what automation buys you. In this episode, Tomas shares five practical tips for getting there – and the biggest traps to avoid along the way.
Why Automation Is Non-Negotiable for Serious Algo Traders
A few years ago, it could take Tomas two or three months to develop a single trading strategy from scratch. Today, he can produce 100 strategy candidates in 48 hours using computer automation. That’s not a marginal improvement – it’s a different business entirely.
Strategy development, robustness testing, pre-selection of candidates, design and prototyping, even portfolio drafting – all of it can be automated. The question isn’t whether to automate. It’s how to do it without wasting years on the wrong things. This is where most traders go wrong.
Tip 1: Don’t Build What Already Exists
The most common mistake Tomas sees – especially among traders who can code – is spending years building a proprietary backtesting platform or trading engine from scratch. It sounds appealing. You get exactly what you want. No compromises.
In practice, it’s a disaster. Three to four years of development. Zero trading during that time. And when you finally finish, your custom solution will have bugs too – just like the commercial platforms you were trying to avoid.
Andrew’s entire automation stack is built around TradeStation. It has bugs. It has quirks. But he’s actually trading, not writing software. The point of automation is to free up time, not to create a new engineering project that consumes all of it.
Take what already exists. Think about how to put these pieces together smartly. You don’t need to build a spaceship to automate your trading process.
There are plenty of powerful, proven platforms available. Use them. Your competitive edge comes from your models and your process – not from your code infrastructure.
Tip 2: Always Think in Terms of Models
This is probably the most important point, and the one Tomas comes back to at the end of the episode. Automation without a conceptual foundation is just random noise generation at speed.
It’s very tempting – once you have the tools – to start throwing random indicators together, sampling different parameter combinations, and seeing what sticks. That produces strategies that look good on paper but have no underlying rationale. No sound model. No reason to believe the edge will persist.
The right approach is to start with a model that is powerful and sound on its own merits. In Tomas’s work, that’s the breakout model – the same foundation he uses in his hedge fund and shares in the Breakout Strategies Masterclass. Automation then runs on top of that proven foundation.
Think about it this way: if traders who spend years building custom platforms instead spent that same time developing and deeply understanding a solid model, they’d get dramatically better results. The model is the edge. The automation is just leverage.
Tip 3: After You Have a Model, Automate It as Fast as Possible
Once you have a sound conceptual model, the goal shifts to implementing it in automated form as quickly as possible. Tomas calls this “Smart Code” – code built within an existing platform (TradeStation, in their case) that automatically generates strategy candidates based on the model’s rules.
No fancy programming. No high-tech proprietary infrastructure. Just an existing platform, a solid model, and code that operationalises that model to generate candidates at scale.
This is the combination that enables the 100-strategies-in-48-hours output. The model does the heavy lifting conceptually. The automation does the heavy lifting computationally. You provide the judgment at selection and validation stages.
Tip 4: Extend the Automation Mindset to Every Part of the Process
Strategy design and prototyping is just one step. The full development process includes:
- Strategy design and prototyping
- Pre-selection of candidates
- Robustness testing (multiple levels)
- Deeper robustness testing
- Strategy verification and completion
- Portfolio construction and drafting
Each of these steps can be automated. Each needs its own conceptual foundation – a fundamental model for robustness testing, a model for pre-selection criteria, and so on. The approach is the same for every step: identify the conceptual model first, then automate on top of it.
In Tomas’s hedge fund, everything is automated through strategy completion. Portfolio construction is still a work in progress – they’ve been experimenting but it doesn’t yet work perfectly. That’s fine. You don’t need to automate everything at once.
Where to Start
If you’re building from scratch, start with strategy design and prototyping. This is where the most time savings are available, and it’s the foundation everything else builds on. Andrew’s approach was linear – automate Step 1, then Step 2, working through the Masterclass framework in sequence. That methodical approach works.
Expect it to take a couple of months to automate the full pipeline from strategy design through robustness testing. That’s a reasonable investment given how much time it saves thereafter.
Tip 5: Think in Terms of Iterations
Your first automation won’t be perfect. That’s not the goal. The goal is to get something running that saves you meaningful time, then improve it continuously.
Tomas’s hedge fund operates on constant iteration – portfolios, strategies, development processes, everything. There’s always something to go back to, learn from, tweak, and test in a new version. That’s not a sign that the process is broken. It’s how the process works.
The lean startup principle applies here: build a minimum viable version, get real feedback from actually running it, then iterate. Trying to build the perfect automation system from the start is exactly the trap that leads to three-year delays. The thing you need to build the perfect version is the feedback you can only get from running an imperfect version.
We always need a margin for learning. Have it imperfect, see what happens, learn from that, and iterate. And iterate. And iterate.
Andrew started building his automations toward the end of 2017. Years later, he was still iterating – converting to TradeStation APIs, automating false breakout and market internals testing. It’s a continuous process. Accept that and build the habit of steady improvement rather than chasing a finished state that doesn’t exist.
The Practical Payoff
What does a well-automated trading operation actually look like day to day? Not total freedom – Tomas is clear about that. He still works hours every day on iterations and improvements. But the flexibility is fundamentally different.
When the foundation is automated, you can take a meditation retreat in Myanmar without internet access for a week and the system keeps running. You can have breakfast, think about new ideas, come back to the desk and review overnight results. The bulk of mechanical work – generating candidates, running robustness tests, screening outputs – happens without you.
That’s the real payoff: not elimination of work, but transformation of what kind of work you’re doing. Less repetitive mechanical execution. More judgment, improvement, and strategic thinking.
Summary: The 5 Tips
| Tip | Principle |
|---|---|
| 1 | Don’t build something that already exists – use existing platforms |
| 2 | Always think in terms of models – automate on top of sound foundations |
| 3 | After you have a model, automate it as fast as possible |
| 4 | Extend the automation mindset to every part of the development process |
| 5 | Think in terms of iterations – imperfect and running beats perfect and theoretical |
The most important of these is Tip 2. There are plenty of automation tools available. What most traders lack is the conceptual foundation to automate on top of. Before you think about how to automate, think about the model. The automation follows naturally from that. Without it, you’re just generating random results faster.
For a deeper look at the breakout model that underpins Tomas’s automation framework, see the Masterclass Part 1, Part 2, and Part 3 series. If you’re new to algo trading without a programming background, this guide covers how to get started.
Ready to automate your trading and stop being glued to screens? Check out Breakout Trading PRO. It’s a structured, visual strategy builder inside BreakoutOS, powered by the proven Mr. Breakouts Formula used by thousands of traders.