Strategy development
Take an idea you already have and run it through the loop properly, until you know whether it is real. Most of the value is in finding out quickly that it is not.
Systematic trading · Mark
I do not look for setups.
I build systems that can be measured.
Track record
321 months, every one of them published.
Returns are percent of a fixed account size, which means they add rather than compound. A year is the sum of its months and the headline figure is the sum of every month traded. That is the conservative way to state a fixed size record, and it is why the number below is not a compounded one.
Cumulative return
Sum of monthly returns on a fixed position size. Drag across the chart to read any month.
Monthly returns
The framework
The iterative loop.
Strategies expire. Every one of them, eventually. What does not expire is the process that produces them, which is why I teach the loop instead of selling the output of it. Most ideas die inside this cycle, and that is the point: the loop is a machine for killing bad ideas cheaply, before the market does it expensively.
- 01
Hypothesis
A claim about how price behaves that can be true or false. Not a setting on an indicator.
- 02
Encode
Entry, exit, stop, size, written as rules a machine executes. If it needs you to feel something, it is not a rule yet.
- 03
Test
Twenty six years, multiple instruments, every regime. Not the stretch where it happens to look good.
- 04
Measure
Change one variable, measure, record. This is the step everybody skips and the reason their results do not survive.
- 05
Kill or keep
Most ideas die here and go back to step one. The ones that survive earn the right to be validated.
Survives the loop
ValidateOut of sample, then walk forward
SizeFor the drawdown, not the return
LiveAnd published from day one
What I actually do
I help people build their own edge.
Validation and testing
Out of sample design, walk forward, parameter sensitivity, cost stress. The tests that separate an edge from a backtest that memorised the data.
Risk and sizing
Most systems do not fail because the edge died. They fail because the position was too large to sit through a drawdown that was always coming.
Prop firm execution
Passing an evaluation is a different problem from making money, and it has its own maths. I have solved it repeatedly and I show you exactly how.
Free, no card, no pitch
Come and argue with the numbers.
The Discord is free and it stays free. Bring a strategy, post a backtest, ask why something failed. I publish trades and answer questions in the open. That is the whole thing.
Join the free Discord