Posted on by — Leave a comment

What has to be true before an EA ships

Between “this idea works” and “members can download it” there is a queue of questions, and most ideas do not survive it. Here is the actual list, in order.

1. Can we say why it should work?

Before anything is coded, we have to be able to explain the edge in one sentence that does not contain the word “optimised.” Some participant is doing something for a structural reason — hedging, rebalancing, covering a gap — and that behaviour leaves a footprint we can trade.

If the only justification is that the backtest is green, we stop here. A curve with no explanation is a coincidence with a nice chart.

2. Does it survive data it has never seen?

Parameters get fitted on one period and then tested on a later one, untouched. The out-of-sample result is allowed to be worse — it always is — but it has to stay recognisably the same strategy. A system that returns 70% in-sample and 4% out-of-sample did not degrade; it was never real.

3. Does it survive other people’s brokers?

Every candidate runs against at least three brokers’ spread and commission profiles. This kills more strategies than anything else on the list, and it kills the highest-frequency ones first. An edge measured in fractions of a pip belongs to whoever has the cheapest execution, and that is not going to be a retail account.

4. What does the worst case actually cost?

Then we design the losses. Where is the stop, and why there. What happens on a gap through it. What the daily loss cap is. What the equity breaker does when the account is down more than the tested drawdown — because the honest answer to “what if this stops working” is “stop trading it,” and that should be code, not a hope.

5. Can someone else run it?

The last question is the one that turns a strategy into a product. Defaults that are safe rather than impressive. Set files for each supported instrument. Inputs named in English rather than in the algebra we happened to use. Documentation written for someone who has never read our code, because nobody ever will.

An EA that only works when its author configures it is not finished.

What gets thrown away

Most of it. That is not modesty — it is the arithmetic of the process. Anything that looks too good gets treated as a bug until proven otherwise, and the investigation usually finds one: a look-ahead in the data handling, a filter that memorised a specific year, a fill assumption no real broker would honour.

Finding those is the job. Shipping the survivors is just the last step.

Posted on by — Leave a comment

The trading day that doesn’t revolve around a screen

There is a particular kind of tiredness that comes from watching a chart you have no intention of trading. It is not work. It feels like work, which is what makes it dangerous.

Automation removes the excuse, not the habit

The promise of an EA is that you get your hours back. In practice, most people automate their entries and keep their screen time exactly where it was — now spent supervising a system that does not need supervision, second-guessing a stop that was calculated more carefully than any decision they would make at 2 a.m.

The strategy did not need you at 2 a.m. That was the whole point.

What actually needs your attention

Very little, and on a schedule. A weekly look at whether the account is behaving inside its expected range. A monthly look at whether the market regime the strategy was built for still exists. A quarterly look at position sizing against your current account size, because the size that was right at $5,000 is not right at $12,000.

Everything between those checkpoints is noise you are paying for with attention.

A schedule that survives a bad week

Pick a fixed time — say Sunday evening — and do the review then. Write down what you saw. Close the terminal. The discipline is not in the reviewing; it is in not reviewing at every other moment.

The test of this is a losing week. Anyone can leave a winning system alone. The reason to build the habit while things are calm is that you will not be able to build it while things are not.

The part nobody puts in the marketing

Trading well is mostly boring. If your process is working, most days contain nothing to do. That emptiness is uncomfortable enough that people fill it — with a new indicator, a tweak to a parameter, a second strategy that “diversifies.” Almost all of that activity is a cost.

Go for the walk. Cook something. Learn the thing you keep saying you have no time for. The market will still be there, and your equity curve will be measurably better for your absence.

Posted on by — Leave a comment

Why we publish our drawdowns

Most trading products lead with the biggest number they own. Ours leads with the worst one. That is deliberate, and it costs us sales.

The number that decides everything

A strategy’s annual return tells you what happened. Its maximum drawdown tells you whether you would still have been holding it when it happened. Those are very different questions, and only the second one has anything to do with you.

Here is the uncomfortable arithmetic. A system that returns 60% a year with a 25% maximum drawdown will, at some point, show you a 25% loss. Not might — will. That is what “maximum drawdown” means: it already happened in the test. And when it happens live, it will not arrive labelled “this is the normal 25% drawdown, please hold.” It will arrive as three bad weeks that feel like the strategy is broken.

The traders who lose money on profitable systems are almost never the ones who picked a bad system. They are the ones who picked a system whose drawdown was larger than their patience.

Why the headline number is the wrong one

Return is the number a marketer optimises. Drawdown is the number an engineer optimises. We would rather sell you a 30% system you can hold for five years than a 90% system you will abandon in month four — because your actual return is the strategy’s return multiplied by how long you stayed in it, and most people’s multiplier is close to zero.

This is also why we publish recovery time alongside depth. A 15% drawdown that recovers in three weeks is an inconvenience. A 15% drawdown that grinds sideways for eight months is a test of character. The depth is identical; the experience is not.

What we ask you to do with it

Before you run anything of ours, do this: take the published maximum drawdown, apply it to the account you are about to fund, and write the dollar figure down. Not the percentage — the dollar figure. Then ask whether you would keep the EA running the day you see it.

If the answer is no, you do not need a different strategy. You need a smaller position size. Halve it and do the sum again.

That single exercise prevents more losses than any filter we could code.

The version we do not show

There is one more reason we lead with drawdown. It keeps us honest. It is easy to tune a strategy until the equity curve is beautiful, and every one of those tunings quietly makes the drawdown worse in ways that only appear later, on data the optimiser never saw. Publishing the ugly number first means we have to look at it first — every release, before we decide anything else.

A result that looks too good is not a success. It is a bug we have not found yet.