Simply, you’ll learn everything that I would seek to learn if I were starting my algo trading journey all over again.

The courses cover a lot of ground. At a high level, you will learn the technical skills that enable algorithmic trading. You will also learn about the various statistical biases and traps that tend to thwart aspiring amateur algo traders. You’ll also learn research processes, workflows, and quantitative tools & techniques that you would find in a professional quantitative trading firm.

For more detail, including descriptions of modules and titles of individual units, check out the curricula for the Fundamentals and  Advanced courses respectively.

To get the most out of the course, you will need as a minimum a beginner’s level of knowledge about the financial markets. In particular, you should know about the mechanics of trading, for example, the different order types, the difference between exchange-traded and over-the-counter markets, and stop losses and position sizing. Ideally, you will have done some manual trading already.

If you don’t have this level of knowledge, you’ll get the best value for your membership if you do some research prior to signing up. There is plenty of material available online for free that will get you up to speed with these prerequisites. Get started with our Back to Basics: Introduction to Algorithmic Trading blog series and one or two of the more introductory texts in our recommended reading list.

Note that programming knowledge is NOT required as a prerequisite – we will get you up to speed with what you need to know. See the next FAQ for more detail.

If you have no prior coding experience, that’s OK! In fact, the first module of Fundamentals of Algorithmic Trading was designed specifically with you in mind! It will get you up to speed with Lite-C – an extremely flexible, powerful, yet simple scripting language – in a surprisingly short amount of time. In fact, you will likely be writing your first code within hours of starting the Fundamentals course.

In many ways, the first part of the Fundamentals course is an ideal introduction to programming. It will take you through all the important basic concepts and will leave you with the ability to turn simple trading ideas into code. As you progress through the remainder of the Fundamentals course and the Advanced courses, your programming skills will continue to grow. Both courses contain dozens of fully explained code examples and you will be supported via the member forums.

If you already know coding, that’s great! That means you already have one of the key skills for successful algo trading and can dive into the others, like statistically sound backtesting, risk management, portfolio construction, and the rest. These other skills, particularly their practical application, are the subject of the latter part of Fundamentals and the entire Advanced course.  Fundamentals will help you leverage your existing coding skills by showing you how to apply them wisely and effectively in a research process for algo trading. Advanced will show you how to take that process further by incorporating quantitative, statistical and machine learning tools into your arsenal.

If you are already a coder, you can probably skip the first part of Fundamentals, however it will provide a handy reference for Lite-C syntax and have you up and running with the language in a matter of hours.

This is something I feel really strongly about, so brace yourself for a long answer.

Algo trading is first and foremost a practical endeavour. Sure, there is probably a minimum level of cross-disciplinary knowledge that is needed, but contrary to popular belief you most definitely don’t need to be a mathematician or physicist to succeed at algo trading. In fact, the theoretical grounding that comes with such a background might even hinder progress rather than help it.

My personal approach to learning algo trading was empirical and experimental in nature. I tried everything I could think of in order to discover what was worth keeping and what to throw away. This was a fascinating path of discovery and really ignited my passion for algo trading. Later, I learned a bunch of theory about why things worked and others didn’t, but if I’d started with the theory I’d have probably lost interest before I even got started.

My point is that you really shouldn’t let your lack of theoretical background stand in your way. Algo trading is all about solving practical problems in the real world, and theory might not even be the best place to start looking for solutions. As you make real progress, you’ll probably become naturally interested in the theory, as I did.

When you’re starting out, rather than having knowledge of advanced mathematics, far more important is having a good, robust research process and the ability to apply the tools that help support it. This is exactly what the Robot Wealth courses provide.  We’ll teach you how to operate a tool that enables efficient, useful and above all practical research. We’ll share with you a process for using that tool wisely. We’ll provide you with dozens of coded examples so that you can implement things straight away that are beyond your current coding abilities. If you start out with these, and work hard to understand the examples line by line, you’ll be well on your way in no time.

Of course, there is a place for theory, and you can always take a top-down approach and tackle it first. After all, that’s the way most things are taught at university. But following the empirical path, you tend to learn whatever theory you need to at the right time – and you get to simultaneously test it out in real life. Further, the theory tends to stick a lot better when you are actually using it in context. If you didn’t become proficient at mathematics through formal study, I’d be willing to bet that the empirical approach leads you to a level of proficiency that you never dreamed possible. I’m absolutely serious about this – you will surprise yourself.

In my experience, the bottom up, empirical approach is a lot more fun and yields faster results, so long as you’ve got the right systems and guidance in place. Being first and foremost a practical problem, algo trading really lends itself to this approach, so don’t let your perceived lack of mathematical ability stand in your way.

Unlike a background in advanced mathematics, a scientific approach to algo trading is critical. But a scientific approach is a skill that can be learned. It is also highly amenable to being encapsulated in a process or a workflow. We will teach you how to approach the markets from a scientific perspective and you’ll see the processes and workflows that worked for me.

The scientific approach is actually a very practical one. After all, science is all about experimentation, observation and inference. If you don’t come from this world, you just need to be shown how it’s done. We’ll introduce you to these concepts in the Fundamentals course, and they’ll be a constant theme throughout the Advanced course as well.

I wish I could tell you that the path to mastery is one that is quickly and easily navigated. But that’s just not true. The reality is that algo trading is HARD. It is difficult because it requires both a broad and detailed knowledge base: ideally you would possess skills in programming, risk management, statistics, finance, markets, computer simulation and data management.

Learning enough about these different fields is not something that happens overnight. If you’re new to both programming and algo trading, you are looking at a minimum of six months of consistent effort (a couple of hours a day, most days) to become competent with all our course material. It may take significantly longer, depending on how much time you can devote to it. If you already have experience with algo trading, you can probably take a couple of months off that figure, again depending on how much time you can devote to the courses.

Also remember that getting through the courses will provide you with the knowledge and tools to develop your own trading algorithms in a scientifically robust manner. You still have to do the actual research and development to get your algorithms into production.

That sounds like a lot of time and effort, but consider that it took me several years to work this stuff out on my own. Here you have a repository of everything I would learn if I were starting out again.

Just to be clear, algo trading is not a get rich quick scheme. Yes the potential rewards are significant, but you will have to work very hard over a long period of time to attain them. As one old pro once told me, “trading is the hardest way to to make easy money.” You’ll need to put in a lot of work over at least a six month period (and probably longer) to attain the skills and knowledge required of successful algo traders.

I don’t know. It really depends on you and how much work you put in. As mentioned in the previous question, algo trading is difficult and it takes time to accumulate the multi-disciplinary skills that are needed for success. Of course, the potential rewards are enormous, but they are not easily attained.

Having said that, if you want to get up and running with an algorithmic trading system quickly, you can always take something from our members’ library and start using it immediately. We provide code for a variant of Dual Momentum, a long-term trend following strategy requiring only monthly ETF rebalancing. It tends to be a good place to start for many (but I don’t know your circumstances and this is not advice) since it is simple to implement and manage and is backed by an extensive body of research. While such a slow-moving strategy is generally not what people look for when they come to algorithmic trading, starting off slow and simple is usually a good idea.

Dual Momentum won’t make you rich overnight, and who knows, it might stop working at some point in the future. In any event, you’ll give yourself the best chance of making consistent money when you have a process for developing algorithms and adding them to a well-diversified portfolio managed with sensible oversight. That’s what you’ll be able to do having completed the Robot Wealth algorithmic trading courses; how quickly you get there is completely up to you.

You will be supported via the Members’ Forum, which is currently moderated by Kris. He strives to reply to questions directed to him within 48 hours. We also offer email support, but ask that this be reserved for non-technical queries. For everything else, the strong preference is for support to be via the forums or the comments section on individual course units. Someone else might have a similar question later, and answers preserved in the forum are beneficial for the entire community. This approach also saves Kris some time, which he can better spend contributing to the community’s collaborative research and development efforts.

You will need to download the Zorro automated trading platform. You can complete the course using the free version, which enables nearly all the functionality of the commercial version, but in live trading has a cap on account size and annual profit. This is a nice aspect of the Zorro platform: the developers ask for nothing until you start making some money with it. You will also need a computer that runs Windows in order to use Zorro.

 If you decide to purchase the licensed version of Zorro, Robot Wealth members get 15% of the purchase price refunded. Get in touch to find out more.

Towards the end of the Advanced course, we also use the R programming language for statistical computing. R is free and open source. We use and recommend the R Studio integrated development environment.

• With some guidance, it is extremely learnable for new coders thanks to the simple syntax of Lite-C and the huge library of pre-built functionality.
• Zorro makes doing serious algo research and trading about as simple as it can possibly get. Note the word serious – forget drag and drop software or any tool that promises you can do algo trading without coding. The fact is that you simply can’t, at least if you don’t want to waste your time. If you’re serious about algo trading, you simply have to bite the bullet and learn to code. Zorro makes that about as painless as it could possibly be, without sacrificing performance, power and flexibility.
• Zorro’s simple scripting language combined with the abstraction of the vast majority of the minutiae associated with algo trading code (MQL coders – you know what I’m talking about) results in a platform that truly facilitates fast and efficient research and prototyping of trading strategies. Once proficient, you will spend 90% of your time testing ideas for trading strategies, rather than writing and debugging code. For beginners, this may not sound all that exciting, but I can assure you that it is a huge win. It wouldn’t be unreasonable to say that developing in MQL means you spend 90% of your time writing and debugging code and only 10% of your time testing ideas. Guess which option leads to faster, better results.
• Zorro compiles your scripts to machine code and uses a flat file structure for historical data stored in binary format. This makes Zorro backtests super fast – 10x faster than MQL, 100x faster than Python and 400x faster than R on benchmark code.
• The combination of fast prototyping and lightning backtests leads to more efficient research. Most individual traders are forced to use what little spare time they have for trading, therefore research efficiency is crucial for sticking with it long term, and ultimately for success.
• Zorro backtests are accurate. This is a big deal, as the markets are notoriously difficult to simulate with high fidelity since you have variable transaction costs and execution dynamics to contend with. Most commercial backtesters I’ve used tend to deliver optimistic backtests, which is great for getting you excited about a trading idea, but really poor for hanging onto your money.
• Zorro goes from research environment to execution engine with the flick of a switch. This is incredibly handy because it means that your research code can be used directly for execution. This saves a lot of time and potential errors. Out of the box, Zorro has support for trading with Interactive Brokers, Oanda, Dukascopy, FXCM and any broker offering the MetaTrader 4 platform. Plugins can be written to trade with any broker.
• Zorro is highly extensible and customizable. While it is a very advanced piece of software out of the box, it is easily extended to access the library of R packages (over 13,000 at last count) and control other windows processes and leverage external DLLs.
• Zorro can download data from various sources and store them as efficient binary files in your own data library.

I do! Check out this video of me coding a simple, momentum-based trading strategy in Zorro. The video demonstrates Zorro’s optimization, walk-forward analysis, and portfolio trading tools – which are really just scratching the surface of what Zorro is capable of, but provides a nice introduction.

At the time of writing, Zorro’s user base was relatively small, but growing at an accelerating rate.

I am a massive fan of what the developers of Zorro have built (and continue to build – they roll out new features every few months), but it is pretty clear to me that they are genius software engineers first and foremost. With all the respect in the world, I speculate that judging by their website, they aren’t the world’s best marketers or salespeople (says the guy with the electric green website). A lot of people would never get past that first page and download the software, although Johan’s (the lead developer) blog is certainly getting the word out that the software is the real deal.

The other obstacle is that the software isn’t particularly well documented. Sure, the manual describes all the functionality, but important things are hidden away in remarks or comments sections and are almost sure to be overlooked. This is not meant to be critical – I’m actually really glad that the documentation is what it is as I’d rather the developers spend their time working on new features than filling the manual with perfectly written technical prose.

This is not really an issue for verteran programmers looking to pick up Zorro, but the downside however is that there is a siginificant barrier to proficiency for non-coders. My belief is that it is not feasible for non-programmers to pick up the software, learn to write code, learn how to apply the advanced functionlaity of Zorro using that code, as simple as that code is, in a reasonable amount of time. I think many non-coders would walk away in frustration.

That’s where Robot Wealth can really help – our courses start off by approaching Zorro form the absolute beginner’s perspective, using it as a vehicle to teach programming rather than assuming that programming is already known. With a focus on the practical, we teach you firstly how to operate the software and then how to use it wisely through extensive code examples that are broken down and fully explained.

The data that we use in the course is all available for free with the Zorro platform. We also show you how to use Zorro and other software to build your own library of data from freely available and paid sources. We will be rolling out a shared data platform for members, subject to licensing conditions and constraints imposed by the providers.

The courses focus on Lite-C and R.

Lite-C is a powerful, simple, C-based language that enables rapid development, efficient research and fast backtesting.

R is a hugely popular and widely used language for statistical computing, data science and visualization. While the syntax of R certainly has it’s own unique nuances, its library of over 13,000 packages give it an enormous amount of power. Thanks to these packages, we can readily use advanced tools from the fields of statistics, econometrics, data science, artificial intelligence and many others without having to code them from scratch.

Lite-C and R make a formidable partnership: we can rapidly prototype and debug trading algorithms using the former, then call advanced R functions directly in our Lite-C script.

Absolutely nothing! I love Python. I use it on an almost daily basis, especially for standalone machine learning or artificial intelligence work. Amongst general purpose programming languages, Python almost has to be the first choice.

The Lite-C + R combination we use in the courses is just a really simple and easy way to get almost a complete algo trading technology stack out of the box: an accurate and fast backtester, an efficient research environment, advanced statistical tools and a trade execution engine. We could get all of this and more with Python, but it would take a lot longer to set up and ongoing effort to maintain.

When it comes to programming languages, I am agnostic for the most part. I take a pragmatic approach by using the best tool for the given task. In this case, the task is to set up an algo trading technology stack that is accessible to beginners and experts alike and which allows the trader to focus on strategy research and development by taking care of as much of the painful detail as possible. The Lite-C and R combination accomplishes this with minimal fuss, but a similar stack could be set up using Python as well.

While Lite-C and R are the programming languages of our courses, the courses are really about an approach to algo trading rather than using a specific programming tool. Everything in the courses can be implemented in Python, and other programming languages as well for that matter.

Robot Wealth provides individual DIY traders with the tools and resources that professionals have access to. One of those resources is a group of other traders, developers, and risk managers of diverse backgrounds and complementary skillsets. Professional traders have the luxury of being able to be a specialist in one or two of the skills needed for successful algo trading, since they can work in teams that together cover all the bases.  Individual traders don’t really have that luxury – unless they are part of a wider community.

The power of the Robot Wealth community is that through knowledge sharing and the combination of various skills and experience, we can achieve a lot more together than any of us could individually. We all come from different backgrounds, and at the time of writing, we counted everyone from complete beginners to ex bank and hedge fund traders as our members. Imagine the power of such a diverse community of motivated and driven individuals, all armed with a baseline level of knowledge from the Robot Wealth courses, working together to build trading systems.

Pretty exciting, isn’t it?

In practical terms, we communicate via an online forum and the messaging application Slack. The forums are repositories of knowledge that we wish to preserve for the benefit of the entire community. Slack is more used for informal discussion and trading system collaboration. We are always on the lookout for new members who are keen and willing to either simply be involved or to even lead the development of a particular trading strategy or idea.

As part of the Advanced Algorithmic Trading course, students have the opportunity to put what they’ve learned into practice. Specifically, they can choose any trading idea or approach to the markets and, following a sound methodology like we teach in the course, undertake a research project that explores the potential to turn that idea into a trading strategy.

We examine each research project and provide detailed, personalized feedback. This is a great way to really round out the Advanced course and to help the student put the skills they’ve learned into practice. I don’t know of any other service that provides personalized, tailored feedback to individuals that doesn’t charge thousands of dollars for the privilege.

Once a student’s submission has been deemed to be of a professional level (sometimes it can take one or two iterations based on the feedback we provide, but this is a great opportunity to learn), the submission is added to the repository of research projects.

Any student who has contributed a research project gets immediate access to the entire repository of research projects. 

That’s totally fine! The point of the research project is not necessarily to create a profitable strategy; it’s to help you put the skills you learned into practice, and refine them even further.

Some research projects will turn into viable trading strategies, but many won’t. That’s a fact of life when it comes to systematic trading research.

But here’s the thing. Even the research projects that didn’t work out are still an incredibly powerful repository of knowledge, and really highlight the power of working with a community. That’s because normally independent traders need to investigate dozens of trading ideas themselves before they find success. Traders the world over are researching and discarding similar ideas. That’s a huge waste of resources. Thanks to the library of research projects, we leverage the power of community to reduce the amount of work each individual has to undertake.

Think of it like amplifying your research effort: submit one research project, get access to dozens.

The other nice by-product of the repository is that it provides a library of code that you can use to find solutions to coding problems that have you stumped. You also might take someone’s research and put a slightly different spin on it to generate a profitable strategy. It’s amazing how much inspiration you can find by looking at the work of others in a slightly different way.

We are unlikely to significantly degrade a strategy given the volumes that we would be trading as a group of individuals and the liquidity of the markets we would be focusing on. If this ever looked like becoming a problem, we would be able to finesse the execution of each individual’s copy of the algorithm, or target more liquid markets. I don’t see this becoming a problem for a while though.

Not really. I personally trade with Interactive Brokers and Oanda, both of whom I am happy with and maintain good relationships with.

Honestly, I would love to work one on one with you, show you the ropes and help you get to where you want to be. The reality is however, that I just don’t have the time. If my story resonates with you, the best way to work with me is as a part of the Robot Wealth community. That way, I can share what I’ve learned with as many people as care to learn about it.

I would love to work with you, and it pains me to turn down the opportunity to work with the driven and intelligent people that make up the DIY trading community, but I have had to stop consulting to individuals due to time constraints.  If you’re an individual, the best input you can get into your project is via our community. If you post there, I’ll end up helping out out anyway, without charging you a hefty consulting fee. You will get a lot from the community in return for sharing your project, far beyond assistance with your project.

If you’re a trading firm or a financial institution, I can best help you out via my consulting firm Quantify Partners.

Not on a one-on-one basis, but certainly via our community.

Unfortunately I’ve had to start saying no to collaboration offers. For a long time I tried to say yes to everyone and ended up just spreading myself so thinly that I wasn’t helping anyone, including myself. If you’d like to collaborate, consider joining our community where apart from me, you’ll find a driven, intelligent and passionate group of likeminded individuals who would love to welcome you into their ranks.

I’ve got a couple of engineering degrees and I worked in that profession for over a decade. I spent most of my time in the resources industry where I worked on computer simulations of environmental processes, particularly hydrology. The highlight was working on a whole-of-life data project that involved constructing networks of environmental monitoring stations in the Australian outback, and then using the collected data to empirically model groundwater dynamics – which doesn’t sound all that interesting, but was actually quite a departure from the generally accepted contemporary methods. It was a lot of fun and I learned plenty.

I enjoyed the resources industry because it really epitomizes the concepts of practical design: you end up engineering things on the run to be ‘fit for purpose’, making do with potentially less equipment and materials than you’d prefer in some fairly remote and challenging environments. A lot of problems are solved through gathering data, experimentation and testing – not unlike algo trading.

Towards the end of my engineering career, I became really interested in artificial intelligence. I was lucky enough to work on problems like building machine learning algorithms capable of identifying endangered species habitat from high-resolution LiDAR data.

In parallel with my engineering career, I was also fascinated with the markets, particularly algo trading. For several years, I spent most of my spare time learning all I could about algo trading, conducting research and writing trading algorithms. This was a long process and even now I look back and shudder at the amount of time I spent pursuing dead ends and false leads (helping you avoid this is one reason for Robot Wealth’s existence, by the way).

Through sheer perseverance, I eventually started getting somewhere with algo trading. I wrote a couple of algos that did really well and started quietly sharing the results with friends of friends and other tenuous connections in the finance industry. This attracted the attention of some high net worth investors, who backed my approach and funded me. This was my first real break, and as things progressed, I wound up being offered a fairly senior role within a hedge fund. The fund was fairly traditional in its approach, but recognized the benefit of automation and more advanced analytics. I was brought in to automate the fund’s existing strategies and to find new sources of alpha through machine learning and big data.

In the beginning, working at the fund was something of a dream come true. I couldn’t believe that I was really here in this world of high finance, being paid to research and trade – the very things I’d freely given all my spare time to for years! Upon entering that world, I assumed that everyone who inhabits it would know the things I know and would be able to do the things that I could do. But it quickly became apparent that this is simply not the case. Sure, there are some firms out there that are literally on the bleeding edge of the latest technologies, like artificial intelligence and even quantum computing. But there are also a lot of firms out there who don’t have a handle on the brave new world of big data and machine learning, and who are at grave risk of becoming irrelevant as a result. It seemed to me that the best informed and brightest fund managers were actively seeking to learn about this new world.

The entrepreneur in me recognized a huge opportunity to provide consulting services to these firms, and working long hours and commuting to the office every day was starting to wear thin (I’ve never been very good at spending time in offices) so I set up a consulting company, Quantify Partners. We helped out trading firms and fund managers who either have a unique and complex data set they want to better understand or even monetize, or who simply want help building capacity in artificial intelligence, quantitative analytics and big data.

I was lucky enough to consult to some of the bigger fund managers in the Asia-Pacific region, but I was even more fortunate to be engaged by a small, tech-focused proprietary trading firm to get them up and running with AI. We hit it off so well that I accepted an offer to join them on a full-time basis as a partner and shareholder to drive the adoption of machine learning and AI across the business. The rest, as they say, is history.

You can read more about my journey from engineer to hedge fund quant to industry consultant and back to professional trader on Robot Wealth’s About page. People tell me its an interesting story, so check it out if you’re curious.

I love this question.

There are so many reasons why algo trading doesn’t work this way that it is hard to know where to start. Ignoring the fact that you don’t just ‘sit back’ and let the algos do their thing (proper oversight can be like a full time job, depending on your trade frequency, and there will always be bugs and infrastructure issues to stay on top of), I don’t want trading to be the sum total of what I do. I love teaching, which provides an enormous amount of personal satisfaction that I simply don’t get from trading alone. Even more than that, my early experiences of learning algo trading in a vacuum really instilled in me a deep belief in the power of community and knowledge sharing.

 Community and knowledge sharing is ultimately about empowerment, and financial empowerment is a big deal in my home country, Australia, and most places around the world. We have a superannuation (retirement fund) industry that is worth billions in annual management fees, yet many funds consistently return less than the local benchmark. We pay billions in fees for this sort of performance and retirees depend on the returns for their livelihoods. A huge part of Australia’s wealth is tied up in this industry and it really disturbs me to see the retirement funds of people who worked hard all their lives needlessly dwindle away. Middle class wage growth has been stagnant for decades while the cost of living continues to go up, so in the future our standard of living as a society is going to be tied more than ever to the performance of retirement funds.

I would love to empower people to take responsibility for their own participation in the markets. After all, no one is going to look after your money as diligently as you!

Further, whenever I share some part of my approach, I invariably get something valuable in return. Sometimes it is as simple as understanding on a deeper level through articulating the concepts. Sometimes it is refinement through feedback, suggestions and ideas that would never have occurred to me. Other times it is simply a new connection with a like-minded person. Very rarely have I regretted sharing what I’ve learned with another person.

I don’t want to give the impression that I want to be the Robin Hood of the finance world. I don’t. I absolutely enjoy making money as much as the next person. I also enjoy the challenges of working in the institutional space, which are different to the DIY space. But the money is only one part of the whole story and would be kind of boring and unfulfilling on its own. I’d also love to see more DIY traders meeting their objectives, as well as a more efficient superannuation industry that returned more money to investors’ pockets, leading to a little more freedom, a little more comfort and a higher standard of living. Those are things that make for a nicer society. And of course I’d like to have some fun and meet some interesting people along the way.