Profit per trade is not a business

A rule can win on every single trade and still lose to just buying and holding — because the money has to be split.

Where the gap comes from

Profit per trade is counted like this: take every signal, count what each one made, average them. In that calculation every trade gets the whole account.

In real life you have the money you have. If the rule fires ten signals on the same day, each one gets a tenth of the capital. And in a week with no signals the money sits idle.

So a rule with a healthy profit per trade can end up making less than simply buying — and with a deeper hole along the way.

How to count it properly

Count the account, not the trade. That means modelling it the way it will actually happen: one pot of capital, several signals at once sometimes, money divided between them, part of the time nothing open at all.

Two things show up immediately that per-trade numbers hide. First, how much of the time your money is working at all. Second, how the account behaves when several trades go against you at the same time.

And compare it to the simplest thing there is: buy and hold. Over the same stretch of time, because over a different stretch you can twist any comparison whichever way you like.

What usually comes out

That the rule and plain buying differ not in the money but in the depth of the holes. The rule lands in roughly the same place, but with bigger drops along the way — which means living with it is harder than the table suggests.

The second common discovery: there are more simultaneous signals than you expected. If the rule likes to fire in clusters, splitting the capital becomes the main constraint, and position size decides more than the idea does.

What this does not tell you

Even a properly modelled account ignores the fact that some trades won't open: not enough liquidity, no leverage available, the instrument isn't there. That's the next check.