Why Most Traders Lose Money — and What the Numbers Actually Say

Somewhere on this website — under the sign-up button, beside every advert we run — sits a sentence regulators require brokers to publish: the majority of retail investor accounts lose money when trading CFDs. This report does something unusual with that sentence. It takes it seriously — not as legal wallpaper and not as a dare, but as a measured fact about people in exactly your position. The fact has documented causes. None of them is a mystery, none of them is rigging, and every one of them is visible before your first deposit.

The number on every website

Open the homepage of any regulated CFD broker in Europe and read the small print. You will find a percentage: the share of that broker's own retail accounts that lost money over the recent measurement period. This is not a survey, an estimate, or an opinion. Firms are required to measure their clients' outcomes and print the result next to their own advertising — one of the stranger and more useful pieces of consumer protection in modern finance.

The numbers are remarkably consistent. The figures national regulators supplied to ESMA in 2018 ran from 74% to 89% of retail accounts losing money. Disclosures published since — broker by broker, year by year, jurisdiction by jurisdiction — cluster in the same region. Whatever explains the number, it is not one bad broker, one bad market, or one bad year. It is structural, which is precisely why it is worth studying rather than flinching at.

Read literally, the disclosure says: most people who attempt what you are about to attempt lose money at it. This report treats that as a research question, and the answer assembled from regulator studies and published research on retail accounts is uncomfortable precisely because it is mundane. Accounts do not fail because the market is rigged against small traders. They fail for three documented, mechanical reasons — and because the reasons are mechanical, each one can be inspected, measured, and partly defused in advance.

Two honest footnotes before the causes. First, the percentage measures accounts over a period, not people over a lifetime — some losing accounts are small, deliberate tuition paid on the way to competence. Second, it hides churn: the population is continuously refreshed by newcomers repeating the same first-year mistakes, which is part of why the figure refuses to improve. Neither footnote softens the headline. Both sharpen the question this report exists to answer: what, mechanically, do most of those accounts do wrong?

Cause one: leverage makes ordinary streaks terminal

Losing streaks are not a malfunction; they are arithmetic. A method that wins half the time will produce five consecutive losses roughly once every 32 five-trade sequences — trade for a few months and you will meet one. What ▲leverage changes is not the probability of the streak. It changes the price of it.

Work the numbers once. A trader with a $1,000 account uses leverage to open 0.5-lot positions on EUR/USD — about $5 per pip. A 40-pip move against the position costs $200, a fifth of the account in one trade; the same 40 pips in the trader's favour pays the same $200, which is exactly the attraction. The mechanism is perfectly symmetrical. The streak is not survivable, though: five such losses in a row — the ordinary, statistically routine kind of streak — and the account is gone. In practice the platform's stop-out closes everything before the fifth loss even completes.

Now run the same streak through a smaller position. A trader risking 1% per trade meets the identical five losses and is down about 5% — annoying, recoverable, survivable. Neither trader predicted the streak; streaks cannot be predicted. The entire difference between a bruise and a funeral was position size, chosen before any forecast was made. The reason size dominates is that losses and recoveries are not symmetric:

That ladder is why overleverage is terminal rather than merely painful. Each step deeper into ▲drawdown raises the climb back at an accelerating rate, while the skill of the trader has not changed at all. The full mechanics — and the sizing rule that contains them — have their own sheets in the risk region of this survey.

Cause two: overtrading, and the bill nobody computes

Every trade starts slightly behind. The ▲spread — the gap between the buying and selling price — is collected the moment you enter, and it is collected whether the trade goes on to win or lose. On a quiet major pair it is small enough to ignore once. It is not small multiplied.

Suppose the EUR/USD spread is one pip and a trader works in 0.2-lot positions, so each round trip costs about $2. Three trades a day is $6; fifteen trades a day is $30. Over a 20-day trading month the selective trader pays roughly $120 in spread and the busy one roughly $600 — on a $2,000 account, a 30% monthly hurdle to clear before a single pip of actual profit counts. The market did not take that money. The fee schedule did, invisibly, two dollars at a time.

▲Overtrading is what turns a small toll into a structural cause of failure, and it compounds twice. The first compounding is the bill above. The second is quality: the second-best opportunity of the day is, by definition, worse than the best one, and the fifteenth is worse still — so the busy trader pays five times the costs to hold a weaker set of decisions. Published studies of retail brokerage accounts point the same direction: trading activity correlates with worse outcomes, not better ones. The traders who survive tend to be the ones who treat frequency itself as a cost.

Cause three: no plan, no sizing rule, no journal

The third cause is the absence of anything that could honestly be called a method. A trader with no written plan decides position size by feel, entry by mood, and exit by stress level. Each decision feels reasoned in the moment. Across fifty trades, the pattern those decisions produce is statistically hard to distinguish from coin-flipping — except that flipping coins is free, and trading carries the bill from the previous section. Random entries plus fees is a system with a negative expected outcome by construction. No market opinion is required to lose this way.

What makes this cause durable is that it hides from its owner. Trading results are noisy, so a planless trader wins often enough to believe a method exists — and memory does the rest, keeping the wins vivid and quietly editing the losses. Without a written record there is nothing to argue back. The journal, covered in its own sheet, is the instrument that breaks this loop: not because writing is virtuous, but because it is the only way to discover which of these three causes is currently charging your account.

What the surviving minority does differently

Flip the disclosure around: if 74–89% of accounts lose money, then somewhere between roughly one in ten and one in four does not. Honesty about that minority first: over any single year, some of it is simply lucky, and the membership rotates. But the studies and the structure of the causes above point at what the durable members share — and it is a short, unglamorous list.

• A sizing rule that runs before any prediction: a fixed, small fraction of the account at risk per trade, so that no single idea — however convincing it feels — can do structural damage.
• A written plan with rules specific enough to be broken knowingly, rather than intentions vague enough to be reinterpreted mid-trade.
• A journal, because the only causes of loss you can fix are the ones you can see in your own record.
• A cost bill they have actually computed — which usually means trading less, in fewer markets, at deliberately chosen hours.

Notice what is missing from the list: a forecasting gift. Nothing above predicts the market. All of it is process — control over size, frequency, cost, and evidence — which is exactly why it is learnable, and exactly why the rest of this survey spends more pages on risk and process than on patterns. Prediction is the part of trading you control least. Process is the part you control completely.

Be equally clear about what process does not do. It does not convert trading into a reliable income, and it does not move you out of the disclosure's denominator by itself — a disciplined trader with a small edge still has losing months, and a disciplined trader with no edge loses slowly instead of quickly. What process buys is time and information: an account that survives long enough to find out, from its own journal, whether an edge exists at all. That is the only honest promise this page will make.

Why a broker publishes this

It is fair to ask why a broker would write this page, so here is the honest accounting. We earn from spreads when you trade, which means we earn more when you trade more — the incentive behind cause two points partly at us, and pretending otherwise would make every other paragraph here less believable. We also keep a client for years only if their account survives its first one. An account that arrives uninformed, oversizes, and is stopped out within three months is a bad outcome for its owner first and for us second. Teaching the causes is not charity; it is the version of our interest that happens to align with yours.

Every broker says trust us. We'd rather you check.

That sentence is the working policy of this whole education, and this report is its test case. Every load-bearing claim above is checkable from your chair: the loss percentages are printed on regulated brokers' websites, the ESMA range sits in a public regulatory decision, the streak arithmetic and the recovery ladder can be re-derived on paper, and the cost bill can be computed from any platform's live spread. Check us. Then check whoever tells you it is easy.