What the badges mean
Shorthand so you can scan a post without reading every word. Each one is something I actually saw, and each has a rule written down here in advance so I can't quietly move the goalposts when I like or dislike something.
You'll notice there's no "SCAM" badge. Not because I'm shy — most of what I look at doesn't survive contact with a tester, and I say so. But "the backtest doesn't reproduce and the live account got deleted in March" is something I can show you, and "scam" is just a word. The specific version is more use to you anyway.
The Trust Index
Every post gets a number out of 10. It answers exactly one question:is whoever is selling this being straight with you?
It is not a safety rating. A terrifying martingale sold with complete honesty scores well. A gentle little EA sold on invented numbers scores badly. If you read a 9 as "safe to trade", that is on you — and I put a warning on the score to make that harder to do by accident.
Where I've run a backtest the number is arithmetic, not a mood. It starts at 5.0 — I know nothing yet — and moves only on findings that carry evidence, so I can't nudge it because I liked someone's email. Where there's no backtest I set it by hand, and the post says so and shows my reasoning.
Martingale, grids and eye-watering drawdowns are weighted zero here. They matter enormously to whether you should touch the thing, but they are about risk, not honesty, and this number only tracks honesty. They are stamped all over the review regardless.
Weights are listed against each badge below. Every review shows its own sums, so you can check my maths.
The bad stuff
Only applied when I saw it myself and kept the receipt. Never on a hunch, never because I disliked the sales page.
- Negative finding: Backtest not reproducible-2.0 trust
I could not get their numbers using their own settings. Draw your own conclusions.
Criteria for applying: Our run, using vendor settings/period/symbol on verified tick data, deviates from vendor claims by more than 25% on net profit or max drawdown.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: No verified live account-1.0 trust
No verified live account anywhere. Just backtests and vibes.
Criteria for applying: No public verified live account located; vendor either provided none on request or did not respond.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Demo presented as live-3.0 trust
The impressive track record is a demo account being sold as real money.
Criteria for applying: Linked account is demo per the verification provider, while marketing copy describes it as live or real-money. Archive both.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Track record withdrawn-2.0 trust
The public track record vanished right after it went bad. Funny timing.
Criteria for applying: Archived snapshot shows the account public and in drawdown; account later private or deleted. Both states archived.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Misleading claims-2.5 trust
The sales page contradicts the vendor's own published numbers. Not my numbers — theirs.
Criteria for applying: A quoted claim is directly contradicted by vendor-published figures. Quote the claim verbatim and archive the page.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Blown account-3.0 trust
A live or signal account running this went to zero, or close enough that it makes no difference.
Criteria for applying: Dated screenshot or archived record of the account before and after. Say whose account it was and whether it was mine.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Blown up repeatedly-4.0 trust
Not a one-off. This has gone to zero more than once, across accounts or across versions.
Criteria for applying: At least two separate documented blow-ups, each with its own dated evidence. List the dates.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Deposit topped up-3.0 trust
Money was quietly added to the public signal account, which flatters the percentages and hides how bad the drawdown really got.
Criteria for applying: Deposit events visible in the account history alongside losses, in a way that keeps the headline return looking healthy. Archive the history — this is a favourite and it gets edited.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Rebranded after failure-2.5 trust
Same code, new name, fresh sales page — sold again to people who missed the first round.
Criteria for applying: Demonstrable link between old and new product: shared code, shared settings, same dev, or an admission. Archive both pages.
Cannot be published without an accompanying dated evidence entry.
- Negative finding: Dev has form-2.0 trust
This developer has previous. Earlier products failed, vanished, or were quietly abandoned.
Criteria for applying: At least two prior products with documented problems, tied to the same developer or company.
Cannot be published without an accompanying dated evidence entry.
Risk characteristics
How the thing actually trades. Not accusations — plenty of people run martingales on purpose — but you deserve to know before you find out the hard way.
- Risk characteristic: Martingale0.0 trust
Doubles down after losing. Wins for months, then gives it all back in one afternoon.
Criteria for applying: Observed lot size increasing following a losing trade, in logs or decompiled logic. Neutral as to legality — this is a risk disclosure.
Cannot be published without an accompanying dated evidence entry.
- Risk characteristic: Grid0.0 trust
Stacks more positions as price moves against it. Fine until it is not.
Criteria for applying: Multiple simultaneous same-direction positions opened at regular price intervals, observed in trade logs.
Cannot be published without an accompanying dated evidence entry.
- Risk characteristic: No stop loss0.0 trust
No hard stop on the broker side. Your downside is whatever the market feels like.
Criteria for applying: Trade log shows positions with SL of 0, or SL set only virtually in EA memory.
Cannot be published without an accompanying dated evidence entry.
- Risk characteristic: Curve-fit suspected-1.0 trust
Nudge a setting by 10% and the profits vanish. It learned the past, not the market.
Criteria for applying: Parameter sensitivity test shows profitability disappearing under +/-10% shifts on key inputs.
Cannot be published without an accompanying dated evidence entry.
- Risk characteristic: High drawdown0.0 trust
Lost more than 40% peak-to-trough in my test. Most people bail long before that.
Criteria for applying: Observed relative max drawdown > 40% in our own backtest run.
Cannot be published without an accompanying dated evidence entry.
Credit where it is due
Same evidence standard as the bad stuff. I am not handing these out to be nice.
- Positive finding: Tick-data verified+0.5 trust
Tested on real tick data at 99%+ modelling quality. The boring baseline everyone should meet.
Criteria for applying: Test executed with real tick data from a named source, modelling quality >= 99%, report published.
Cannot be published without an accompanying dated evidence entry.
- Positive finding: Out-of-sample passed+1.0 trust
Still made money on data it was never tuned against. Rarer than you would hope.
Criteria for applying: Out-of-sample period of at least 12 months, never used for parameter selection, ends net profitable.
Cannot be published without an accompanying dated evidence entry.
- Positive finding: Verified live account+2.0 trust
A real, verified, live account exists and it matches what the sales page says.
Criteria for applying: Public MyFxbook/FXBlue/broker-verified account, live (not demo), at least 6 months, consistent with marketing claims.
Cannot be published without an accompanying dated evidence entry.
- Positive finding: Reproducible+2.0 trust
I reran their own backtest and got their own numbers. It happens.
Criteria for applying: Our run matches vendor net profit and max drawdown within 10%, using their stated settings, period and symbol.
Cannot be published without an accompanying dated evidence entry.
- Positive finding: Survived live+2.0 trust
Ran on real money for a decent stretch without falling over. Boring, which is the point.
Criteria for applying: At least 12 months live — mine or a verifiable third party — with no blow-up. Evidence of the period and outcome.
Cannot be published without an accompanying dated evidence entry.
- Positive finding: Dev has a track record+1.0 trust
This developer has shipped before and the earlier products are still standing.
Criteria for applying: At least two prior products still supported and not associated with blow-ups or disappearances.
Cannot be published without an accompanying dated evidence entry.
Notes
How I got hold of it, and what kind of post this is.
- Note: No backtest0.0 trust
I haven't run this through the tester. Everything here is experience and observation.
Criteria for applying: Applied automatically whenever a post carries no test results.
- Note: Purchased copy0.0 trust
I paid full retail for this one. The vendor had no idea I was testing it.
Criteria for applying: Retail purchase receipt retained.