What we do not claim about Expert Advisor performance

Search for "EA builder" and you will find win rates, monthly return figures, five-star testimonials from traders with stock-photo faces, and counters claiming tens of thousands of users. Almost none of it is verifiable, and some of it is invented outright.

We do not publish any of it. This post explains what we say instead, and why, because "we make no performance claims" is easy to write and easy to quietly walk back later.

What is not on this site

• No win rates, profit factors or return figures. Not for templates, not for marketplace listings, not for anything.
• No testimonials with numbers attached. No "made 40% in three months" quotes.
• No user counts. We are not going to tell you how many traders use this as an argument for whether it works.
• No aggregate rating markup. Search engines will show a star rating in results if a page declares one. Declaring one without real reviews behind it is a policy violation as well as a lie.

Why not — beyond the obvious

The honest reason is that we do not have the information.

A backtest figure describes a strategy on one symbol, one broker's data, one spread model and one date range. It says very little about the same strategy on your account. Presenting one as a property of the tool is a category error: TradeSmithy builds what you describe. If the strategy is bad, it will faithfully build a bad strategy, and it will do so quickly and correctly.

There is also a rule we apply to ourselves and would rather state publicly: a claim on a paid surface is a promise the code has to keep. This is not abstract. "Priority AI throughput" once sat on our $49 plan card while the rate limits were flat constants with no per-plan branch anywhere in the codebase — a purchase inducement for a feature that did not exist. It was removed. Anything unfinished now carries an explicit "in development" label driven from a single constant, so every surface that mentions it flips together and none of them can drift into overstating.

What is in development, said plainly

At the time of writing:

• Building and exporting MQL4 and MQL5 source: live.
• The AI assistant that edits the node graph: live.
• The marketplace: live. Earnings land in your wallet the moment a sale settles, and they are spendable immediately on your plan and on other traders' strategies.
• Withdrawing sale earnings: live, with conditions worth stating. Earnings mature 14 days after the sale, the minimum withdrawal is $50, and only money you earned can be withdrawn — never funds you deposited yourself. You request a payout in the app and a person reviews and sends it; there is no automated payout button, deliberately.
• Historical backtesting: live. You configure and queue a run, it executes on our test servers, and you get back an equity curve, the full trade list and the run log.
• Parameter optimization: in development.

Backtesting going live does not change anything in the section above. A result the tester produced is a measurement of one strategy against one broker's historical data, and we will publish no aggregate built out of those numbers — not a leaderboard, not a "best performing template", not a win rate on a listing.

If you are reading this well after publication, the plan cards and the landing page are the current answer — they read from the same constants the enforcement does.

What we do claim

Concrete, checkable things:

• The export is real MQL4 or MQL5 source that compiles in MetaEditor. You can read every line of it.
• The graph is validated before an export is generated — an indicator your target platform does not ship, an assignment with no variable connected, a parameter name that is not a legal identifier — each is refused up front and names the node responsible.
• A marketplace buyer receives an editable graph, not a locked binary.
• A strategy you build is yours. There is no runtime library, no licence server, and nothing that stops working if you stop paying us.

What you still have to do

Decide whether the strategy is any good. Test it on data you trust. Run it on a demo account long enough to be bored by it. Understand the risk of the instrument, the leverage and the position sizing you have chosen.

Trading Expert Advisors carry real risk of loss. Automation removes the hesitation, not the risk — a bad strategy executed flawlessly and without emotion loses money faster and more consistently than a bad strategy traded by hand. Nothing on this site is financial advice, and none of it should be read as a suggestion that any particular strategy will be profitable.

That is a duller pitch than a win rate. It is the one we can actually stand behind.