What Are Forex Signals (and Why This Site Doesn't Sell Them)

A forex signal is a recommendation to enter a specific trade, usually specifying the pair, the direction (buy or sell), an entry price or zone, a stop-loss, and a take-profit. Signals are sold by individuals and companies on a subscription basis, distributed through Telegram channels, Discord servers, paid forums, mobile apps, and “VIP rooms” on broker platforms. The industry around them is large, opaque, and the single biggest source of avoidable losses among retail traders.

This site does not sell signals, and the reasons we do not are exactly the reasons most readers should not buy them. This article walks through what signals actually are, how the business model really works, what the regulator-collected data shows about outcomes, and the simple arithmetic trap that makes a confident-sounding “70% accuracy” claim mean almost nothing. It is, on purpose, the longest piece of brand position on this site.

What a signal actually contains

The format is more or less standardised across the industry:

EUR/USD · BUY Entry: 1.0850 Stop-loss: 1.0820 Take-profit 1: 1.0880 Take-profit 2: 1.0910

Followed in better-quality services by a short justification, and in worse ones by an emoji or a screenshot of a chart with an arrow on it.

The implicit promise is: we have done the analysis; you take the trade. The customer is asked to skip the cognitive work of forming a view and to substitute the seller’s judgement. This is the part of the proposition that is appealing, and also the part that does most of the damage.

The business model is the subscription, not the trade

Pause on the economics for a moment. A signal-seller who genuinely believed they had a tradable edge has two ways to capitalise on it. They can trade the edge themselves, scaling capital up to the point the edge starts to decay. Or they can sell the edge to a few hundred subscribers at $99 a month.

The first path scales with the edge: a strong edge becomes serious money. The second path scales with the audience: a strong edge and a weak edge both pay roughly the same as long as people keep subscribing. For a seller with a real edge, the first path is overwhelmingly more profitable; for a seller with no edge, only the second path pays anything at all. The presence of a paid signal business is therefore mild evidence against the seller having an edge, not for it.

This is not a strict proof. There are edge cases (the seller is capital-constrained, the seller is genuinely altruistic, the service is sold to other professionals as a research feed). But the base-rate logic is real, and it explains why the most prominent voices in retail signal selling have always seemed disproportionately busy on social media and disproportionately quiet about audited trading records.

The published-track-record problem

A reputable systematic fund publishes a long, audited, third-party verified track record covering all trades, with the methodology and fees clearly stated. A typical retail signal service publishes screenshots of winning trades.

The asymmetry matters. Two patterns dominate the “track records” that do appear:

Selection. Only the winning trades are highlighted; the losers exist but do not feature in the marketing graphic. Even an unedited random sequence of trades will contain enough wins to fill a highlight reel.
Reinterpretation. Trades are reported as “closed in profit” if any part of the position reached any partial profit target before the stop was hit. Trades that ran straight to the stop are reported as “closed at stop” without their proportional weight in the overall performance summary.

The defensive practice is to ignore any track record that is not every trade in chronological order, with timestamps, including the ones that were closed manually, and ideally verified by a third-party platform that tracks the broker account directly rather than scrapes a Telegram channel. Anything short of that is the financial equivalent of an unverified before-and-after photo.

What the regulator-collected data shows

This is the part that does not require trust. Brokers regulated under the European Securities and Markets Authority (ESMA) framework are required by law to publish the percentage of their retail accounts that lose money on CFDs, of which forex is the largest category. The warning text is mandated:

“[insert percentage per provider] % of retail CFD accounts lose money when trading CFDs with this provider.”

Across the regulated brokerage population, the disclosed figures have held in the 74% to 89% range since the disclosure regime began in 2018. The UK Financial Conduct Authority’s analyses produce broadly similar numbers (around 80% of UK retail CFD traders unprofitable). These are not survey results or industry estimates: they are the broker-by-broker, regulator-mandated, observable disclosures sitting under the risk-warning banner of every CFD brokerage homepage in the EU and UK.

There is no parallel published figure for “the percentage of signal service subscribers that lose money”, because signal services are neither regulated venues nor required to disclose anything. But the relevant universe of customers is essentially the same population (retail CFD traders) trading with overlapping brokers. If the population-level figure is 74-89% losing, the signal-subscriber subset inherits that baseline as a floor, then adds the cost of the subscription on top.

The 70%-accuracy trap

The signal services that do quote a numerical claim almost always quote a win rate: “73% accurate”, “8 out of 10 winners”. This is the single most exploited number in the industry, because a high win rate is not the same thing as a profitable strategy, and most subscribers do not know it.

Consider a service that genuinely does win 70% of its calls. The account follows the calls faithfully, risks 1% per trade, takes the typical signal-service risk-reward ratio (average win $50 on a 0.5R target hit early, average loss $200 on the full 1R stop). The trade distribution looks reasonable. The bottom line does not.

Fig. 1 Equity curve of a hypothetical $10,000 account that follows a 70%-accurate signal service for 60 trades, with average win $50 and average loss $200. The account is right far more often than it is wrong. It still ends below where it started, because the losers are four times larger than the winners. This is the structural trap behind almost every 'high accuracy' signal marketing claim. Illustrative data: a synthetic series generated for teaching, not a real market quote.

The arithmetic is the same as in Risk Management Basics: expectancy is (win rate × average win) − (loss rate × average loss). In this example, 0.7 × $50 − 0.3 × $200 = −$25 per trade. The system loses money on every trade on average, even though it wins three times out of every four. is what most subscriber equity curves quietly do once trading costs are factored in.

This is not a hypothetical pattern. It is the shape almost every “high-accuracy, low-risk-reward” signal service produces. The reason the marketing leans on the win rate is precisely because the win rate is the only number in the equation that flatters the product.

The affiliate-kickback layer

A great many signal services do not sell only the subscription. They also direct subscribers to a specific broker, from whom they receive a revenue-share on the spread the subscriber pays on every trade. This is called an introducing-broker or affiliate relationship, and on average it pays the signal seller in the range of $5 to $15 per standard lot the subscriber trades.

The incentive distortion is direct. The signal seller’s revenue grows with subscriber trading volume, which means more signals, more frequent signals, larger position sizes. A signal service whose average subscriber trades twenty times a week generates more affiliate revenue than one whose subscribers trade twice. The seller’s optimal behaviour, independent of any edge, is to keep the subscriber trading actively.

This is the second business-model leg, on top of the subscription, that keeps signal businesses viable even when the underlying calls do not make money. It also explains why so many signal sellers are paired with brokers that few regulator-conscious traders would otherwise choose: the affiliate payouts at offshore venues are typically much larger than at fully-regulated ones.

The regulators’ view

Authorities have been active on this market for years. A non-exhaustive sample:

The UK Financial Conduct Authority has issued repeated warnings about unauthorised signal providers and the use of social media to promote leveraged products, and has taken action against firms presenting signal services as if they were investment advice.
The US Commodity Futures Trading Commission has prosecuted multiple “guru” cases for fraudulent signal and copy-trade promotion, including via Telegram channels and YouTube.
The European Securities and Markets Authority introduced the CFD intervention package partly in response to the harm caused by the promotion ecosystem (of which signals are a major component) to retail traders.

The pattern of enforcement is consistent: the services that are prosecuted are not edge cases. They include some of the most prominent “trading gurus” of the past decade. That should inform how a reader weighs the rest of the field.

Where signals do make sense

It would be dishonest to claim that every signal-style product is bad. A few cases where an external trading recommendation can be a legitimate, sometimes useful, instrument:

Institutional research feeds. Banks and brokerages publish trade ideas as part of their research output for institutional clients. The recipients are professionals who incorporate the ideas into a broader view; nobody is acting on the recommendations blindly.
Regulated copy-trading on transparent platforms where the lead trader’s full track record (every trade, position sizes, fees) is publicly verifiable on the platform’s books, and the platform takes a regulated cut rather than running a private affiliate arrangement.
Quantitative subscription services with published, audited, long-window track records that include methodology, risk parameters, and downside disclosure, sold to subscribers who size positions against a documented expectancy curve.

The shared feature in all three is verifiable evidence and a properly-aligned incentive. The shared feature in the typical Telegram signal service is the opposite.

Why this site doesn’t sell signals

We are an education site, openly published by TradingFuse, and one of the principles every site in the portfolio runs on is that we do not sell signals, run paid groups, or wrap a product in “education”. The economic case for selling signals is exactly the one above: subscriptions scale with audience, not with edge, and we prefer the long, slower business of being a reference readers actually trust.

The editorial case is the more important one. The marketing flywheel that signal-selling runs on (urgency, win-rate claims, affiliate broker funnels, emphasis on direction calls instead of risk frameworks) is the same flywheel that explains the published 74-89% loss-rate disclosures. We would be participating in the harm rather than explaining it.

What to do instead

If you wanted to subscribe to a signal service for the convenience of not having to form your own view, the productive substitution is learning the underlying analysis. That is a slower process and a less satisfying one in the short term. It compounds. Specifically:

Understand what moves exchange rates so that a trade idea is grounded in a thesis, not a screenshot.
Build a position-sizing rule and stick to it, regardless of how “high-conviction” any individual idea feels. (See Risk Management Basics.)
Track every trade in R-multiples to build a real, evidence-based view of whether your own decisions have any edge.
Treat any external recommendation, paid or free, as raw input to your own analysis, never as a substitute for it.

If, after all of that, you still want to follow an outside view, the filter is the one above: verifiable, complete track record; clear, proper alignment of the seller’s incentives with the subscriber’s outcomes; full disclosure of fees and the realistic downside. Anything short of that is paying to be a customer rather than to be a trader.

The takeaway

A forex signal is a trade recommendation. The signal-seller business model is structured around subscriptions and affiliate-broker kickbacks, both of which scale with subscriber activity rather than with the underlying quality of the calls. Published track records almost never withstand scrutiny; published win-rate claims hide the risk-reward arithmetic that determines whether following them makes money. The mandatory regulator-disclosed loss rate for the broader retail CFD population, of which signal subscribers are a subset, sits at 74-89%.

This site does not sell signals because the economics, the incentives, and the evidence all point the same way. We would rather spend the time explaining the mechanics that let a reader form their own view, with the trading costs, order types, and risk framework they need to test it carefully.