The share matters more than the average

On scattered data the average barely means anything — what you want is the share of cases that finished up.

Why the average lies

The average works when the cases resemble each other. On the market they don't: most moves are modest and the rare ones are enormous. A single one of those drags the average so far that it stops describing a typical outcome.

You get an odd result: the average is big and almost every case is down. Or the reverse — the average sits near zero while most cases finished up, and one collapse ate all of it.

What to count instead

First the share: what fraction of cases finished up. That number is steady — no single outlier can move it.

Then the typical case: line every result up in order and take the one in the middle. It answers the question "what usually happens", which the average doesn't answer at all.

And only then the average, as the answer to a different question: how much you end up with if you repeat this many times in a row.

You need all three, because they answer different questions, and swapping one for another is the most common mistake people make reading results.

What usually comes out

That the share and the average pull apart, and pull apart hard. A rule can clear the bar comfortably on the share while looking like noise on the average — and the other way round.

It's also worth looking at the worst cases separately: not "how much on average" but "how bad does it get". That's what sets your position size, and it's what usually wipes an account.

What this does not tell you

A high share of winners does not mean money by itself. A rule that closes at a small profit and sits through the losses gives a beautiful share and ruins you.