Why I'm publishing these
What these notes are, who they are for, and what I hope comes back from writing them in public.
I have spent a long time learning how systematic trading works, first professionally and then, especially over the last two years, by building and running my own portfolio and the end-to-end platform that runs it. A good deal of that learning came from making mistakes. Looking back, there were many moments when I would have valued a clear, well-reasoned article on the problem in front of me, and could not find one. These notes are my attempt to write some of them. I am sharing them on the chance that others are working towards similar goals and trying to overcome similar problems, and might find something useful here.
Who they are for
The notes are written for a few kinds of reader. Some may be starting out in systematic trading and trying to avoid costly mistakes. Others may be experienced and looking for a different answer, or a better one, to a problem they already know well. Some may simply want to understand how I work and how I think about trading. I have tried to write for an educated reader rather than an expert one, explaining each term the first time it appears, while keeping the detail precise enough to be useful to a practitioner.
What they cover
Most of what I found when I was looking focused on one part of the problem, usually the strategy itself. These notes try to look at systematic trading from end to end: the research process, the platform that runs it, the data it depends on, and the trading and risk management that turn a tested idea into real positions. In my experience, the problems that matter most tend to sit where those parts meet: where a research assumption about fills meets live execution, or where a revision to historical data quietly changes a result. That is where many of these notes sit.
What I hope comes back
I would welcome conversations with other practitioners, and I would particularly welcome feedback that challenges my thinking, because that is usually where I learn something. If something here is wrong, or you have solved the same problem in a better way, please write to me at hello@determinisk.com.
Where to begin
I will publish as new subjects come up in my own work, rather than to a fixed schedule. The topics are broad and nobody needs to read everything, so the Start here page groups the pieces into series and suggests an order to read them in.
Related
- Method
5 reasons your P&L doesn't match your backtest
Before blaming the market, check whether the backtest and the live system were really doing the same thing. These five are the most common reasons they were not.