Here is every signal we got wrong, with the levels we published

Our track record has counted the losses since the day it existed. What it never did was show you the trade.

You could see that a signal stopped out. You could not see what we had actually told subscribers to do — where the entry was, where the stop sat, what we were aiming at. So the record proved we were willing to count a loss, and nothing more. It did not let you judge whether the setup was any good.

That changes today, in two ways.

A closed signal now publishes its setup

Until now, entry, stop and targets were stripped from every public page. The reasoning was sound and it still is — while a signal is open, those levels are the thing subscribers get and nobody else does.

But that argument stops applying the moment a trade closes. A signal that resolved three weeks ago is worth nothing to someone trying to get the product for free. Withholding its levels was not protecting anything. It was just hiding the evidence.

So resolved signals now carry the setup exactly as it went out, before anyone knew how it would end:

Open signals still show no levels. That part has not changed and will not.

Every loss, on one page

The second change is a page: /wrong.

It lists every signal that stopped out — 96 of our 266 resolved signals — newest first, each with its market, its dates, the move it made against us, and the setup we published. Nothing is selected, reworded, or quietly left off. A signal enters that list by hitting its stop, and nothing takes it off.

The worst single one is sitting there at −32.5%.

We built it because of a gap between what we say and what a reader can check. Every channel we have says some version of we publish our losses. That sentence is easy to write and impossible to verify if the losses are scattered across a paginated list. Now it is a link.

One number that needs saying plainly

A page of losses invites an obvious question, so here is the answer before it is asked: 96 losses out of 266 resolved signals is a 64% win rate, and a win rate on its own tells you almost nothing.

It is one of two numbers. The other is what a win pays relative to what a loss costs, and no win rate is meaningful without it. We have written about this before and we will write about it again with our own figures. Treat any service quoting you a win rate and nothing else — including this one, when we do it carelessly — as having told you half a fact.

Also new: a page per market

We scan 3,949 instruments across 26 markets, and most of that is not the US — Japan, Korea, Taiwan and Hong Kong together are more than half the universe. Until now that was one long list on the coverage page.

Each market now has its own page: what we scan there, every signal we have published in it, that market's own resolved count and win rate, and the median move. Türkiye, Japan, Germany, Crypto, and twenty-two more. Some of them are honest and empty — we scan Austria and nothing has resolved there yet, and the page says exactly that rather than showing you a hopeful zero.

Why bother

Because a track record you cannot check is a claim, and a claim is worth nothing in this category. There is no shortage of accounts posting green screenshots. The only thing that distinguishes a real record from a curated one is whether the losing half is as easy to find as the winning half.

Now it is easier.

⚠️ Not financial advice.

⚠️ Not financial advice.