Visit the homepage of any European, UK, or Australian retail CFD broker and the first piece of regulated content you see, usually in 9-point grey type at the bottom of the screen, is the loss-rate disclosure. It looks something like this:
"76% of retail investor accounts lose money when trading CFDs with this provider. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money."
The number changes by broker and updates every quarter. It is a useful piece of information for the regulator, who designed the disclosure, and a less useful one for the customer reading it. Most retail customers see the number, do an internal calculation that goes "but I will be in the 24%," and click "Open Account."
This is not the most informative line on the page. The disclosures that come next, written in dense PDF documents linked from the footer, contain four specific data points that are far more useful for deciding whether to deposit money. They are rarely read.
One: The actual entity that holds your account
The brand on the homepage is almost never the legal entity onboarding you. A broker called Generic Brokers Limited might operate three regulated subsidiaries (Generic UK Ltd, Generic EU Ltd, Generic AU Pty Ltd) and one offshore unregulated subsidiary that takes overflow from countries the regulated entities cannot serve.
The disclosure document will name the entity assigned to your jurisdiction. Find that name. Then go to the public register of the regulator named on the page (FCA in the UK, CySEC in the EU, ASIC in Australia, etc.) and search the entity name there. Read the licence scope. Confirm that the activities the entity is licensed to provide actually include the products you intend to trade.
This sounds paranoid. It catches a meaningful number of issues. The most common one we see is a broker advertising stock trading on the homepage when the entity onboarding non-EEA customers is licensed only for CFDs. The customer thinks they are buying shares. They are buying contracts that reference the share price. The economics are similar in normal markets and very different in a corporate action or a delisting.
Two: The negative balance protection clause
Negative balance protection means that if the market gaps against you while you are leveraged and the broker cannot close your position before it goes deeply negative, the broker eats the loss rather than billing you for the difference. It is mandatory for retail clients in the EU and UK. It is not mandatory in many other jurisdictions.
The clause should appear in the risk warning document with the precise wording "negative balance protection applies to retail clients" or similar. If it is missing, ambiguous, or qualified ("up to a limit of"), the customer can in principle owe the broker more money than they deposited. This has happened in real flash-crash events, including the SNB removing the EUR/CHF floor in January 2015 and oil futures going negative in April 2020.
Three: The order execution policy
Brokers regulated under MiFID II in Europe and the equivalent regimes elsewhere are required to publish an order execution policy. It is usually a separate PDF, often hosted at /legal/best-execution-policy or similar, and it tells you something the marketing pages will not: who actually fills your trades.
You are looking for two things in this document. First, whether the broker fills your trades against an external venue (a regulated exchange, a multilateral trading facility, or another bank's market) or against its own book. The latter, called "principal" or "B-book" execution, is not inherently bad, but it changes the broker's incentive structure in ways the customer should be aware of. When you lose money on a B-book trade, the broker keeps the loss. This is a legal arrangement the regulators have signed off on, but it puts the broker on the other side of your trade.
Second, look for the broker's own published execution quality statistics. Under MiFID II these are mandatory annual disclosures, called RTS 27 reports historically and sometimes appearing under different names now. They show, for the broker's most-traded venues, the average price improvement, the slippage on stop orders, and the percentage of orders that were filled within a defined band of the quoted price.
Four: The withdrawal policy
This is the most overlooked section in any account agreement and the one we recommend reading most carefully. Specifically, find the paragraphs that cover:
- Whether the broker can refuse a withdrawal if the receiving account is in a different name from the funding account.
- The maximum number of days the broker has to process a withdrawal under normal market conditions.
- Any conditions under which the broker can convert your withdrawal currency or apply additional fees.
- Whether the broker requires re-verification of identity before a withdrawal of a certain size.
None of these terms are unusual or sinister on their own. The reason to read them before depositing rather than after the first profit is that the entire customer experience pivots on this section. A broker with low fees, a fast platform, and a slow withdrawal process is a worse broker than the spreadsheet suggests.
What the loss-rate number is actually for
The "76% of retail accounts lose money" headline is a disclosure, not a forecast. It is the regulator's way of forcing the broker to acknowledge, on every customer-facing page, that the product is genuinely high-risk for retail customers. It is not a number the customer should treat as a probability that applies to them personally.
The right reaction to the disclosure is not "I will beat the odds." The right reaction is to read the four documents above, decide whether the broker is operating the way you want a broker to operate, and then size your position so that being in the 76% does not change your life. That is the entire point of the warning.