Risk Management Basics
Position sizing, fixed-percent risk, R-multiples, expectancy, and the brutal math of drawdown: the framework that decides whether a strategy survives long enough to be tested.
The published statistics are not subtle. Across brokers regulated under the European Securities and Markets Authority (ESMA) framework, the mandatory risk-warning disclosures show that between 74% and 89% of retail CFD accounts lose money (the figure has held in that band since the disclosure rules were introduced in 2018). The UK Financial Conduct Authority’s own analysis sits at roughly 80%. The figures cover all products in the CFD wrapper, of which retail forex is the largest single category.
The interesting question is not whether trading is hard. The interesting question is what separates the small minority of accounts that survive long enough to find out whether they have an edge from the majority that do not. The honest answer, in nearly every careful study, is risk management, and specifically the unglamorous mechanics of position sizing. This article walks the framework. It is the most important piece of practical literacy in trading, and the part the marketing brochures spend the least ink on.
The first decision is sizing, not direction
Beginners almost always frame trading as a sequence of direction calls. Will EUR/USD go up or down? Will the Fed surprise hawkish? The framing feels natural and it is wrong. The first decision a serious trader makes on any position is how much, not which way.
The reason is mechanical. A trader who is wrong half the time can still make money if the wins are larger than the losses. A trader who is right 80% of the time can still lose money if a single bad trade is sized larger than the previous twenty winners. The size of the position controls the relationship between being right and being solvent. Direction is a thesis. Sizing is a hard constraint.
This is also why a strategy that “works” in screenshots is so often unprofitable in practice. The screenshots show the direction calls. The account blowups happen in the sizing.
Position sizing: the formula
Position sizing in retail FX runs on one short equation. It looks arithmetic-heavy at first; it becomes second nature within a few weeks of using it.
position size (units) = (account equity × risk per trade) ÷ (stop distance × pip value per unit)
Worked numerically:
- Account equity: $10,000.
- Risk per trade: 1%, so $100.
- Stop distance: 30 pips (the planned distance from your entry to the level at which the trade thesis would be wrong; see Order Types Explained).
- Pip value per unit on EUR/USD: $0.0001.
Position size = ($10,000 × 0.01) ÷ (30 × $0.0001) = 33,333 units, or just over three mini lots.
If the trade hits the stop, the loss is 30 pips × $0.0001 × 33,333 ≈ $100, which is exactly the 1% risk you set. If the stop has to be wider because of where the relevant level actually sits, the position size automatically shrinks; if it can be tighter, the position size grows. The dollar risk is constant by construction.
The point of the formula is that the trade tells you the size. You do not decide to “go in big” or “go in small”. You decide the maximum loss you are willing to take, you read the stop distance from the chart, and the position size falls out. It is the only step in the process the trader can fully control.
Fixed-percent risk: why a fraction, not a fixed dollar amount
A common alternative is to risk a fixed dollar amount per trade (“I always risk $200”). Fixed-dollar is simpler to compute but it has a well-understood pathology: when the account is winning, you are risking proportionally less; when the account is losing, you are risking proportionally more. The math points exactly the wrong way.
A fixed-percent rule (1%, 2%, sometimes lower) does the opposite. After a loss, the next trade is sized off a slightly smaller account, so the absolute dollar risk shrinks. After a win, it grows. The account de-leverages into losses and re-leverages into wins without any conscious decision. Most professional risk frameworks of any flavour are variants of this.
How much per trade is the right fraction is a long argument with no clean answer. The most-cited theoretical reference, the Kelly criterion, gives the geometric-growth-maximising risk fraction for a known edge, but Kelly assumes you actually know your edge, which retail traders almost never do; using full Kelly with a mis-estimated edge is a fast route to ruin. Most professional risk frameworks use a small fraction of Kelly. The conservative range that nearly every careful practitioner converges on is 0.5% to 2% per trade, with 1% as the common default.
R-multiples and expectancy
Talking about wins and losses in dollars stops scaling once the account size changes. The standard fix is to express every trade as an R-multiple: a unit equal to your planned risk on that trade.
A trade where you risked $100 and made $250 is a +2.5R trade. A trade where you risked $100 and lost $100 is a −1R trade. A trade where you risked $100 and lost $150 because of slippage is a −1.5R trade.
Express a strategy’s record in R, and you can compute its expectancy: the average R per trade across the sample.
expectancy = (win rate × average winning R) − (loss rate × average losing R)
A strategy that wins 50% of the time, with average win 1.5R and average
loss 1R, has an expectancy of 0.5 × 1.5 − 0.5 × 1 = +0.25R per trade.
Over a hundred trades, the expected outcome is +25R. If you risk 1% per
trade, that is +25% on the account, before friction.
Expectancy is the right metric. Win rate alone is not.
Win rate is not the metric
This is the trap signal-selling and “high-accuracy” systems exploit. Consider two strategies on the same account:
| Strategy | Win rate | Avg win | Avg loss | Expectancy per trade |
|---|---|---|---|---|
| A | 70% | 1R | 3R | 0.7 × 1 − 0.3 × 3 = −0.2R |
| B | 40% | 3R | 1R | 0.4 × 3 − 0.6 × 1 = +0.6R |
Strategy A wins more than two-thirds of the time and loses money. Strategy B is wrong on most of its calls and is the better strategy by a wide margin. The intuitive thrill of “being right” is unhooked from the arithmetic that pays the bills. Any source advertising a strategy by its win rate, without R-multiples, is either confused or selling something. (See What Are Forex Signals for the extended version of that argument.)
A losing streak is normal: the distribution
Even a strategy with strong positive expectancy spends meaningful time underwater. The relevant maths is the binomial distribution. For a strategy with a 50% win rate, the probability of seeing a streak of at least five losses in a row over the course of 100 trades is above 80%. A streak of six is still better than 50/50. A streak of seven, roughly 30%.
If five consecutive losses is enough to make you abandon the strategy, the strategy was abandoned by definition before its edge had time to manifest. Sizing exists partly so that the streaks the math guarantees are survivable.
This is also why systems with very low win rates and very high average-R-per-win (trend-following is the classic example) feel psychologically punishing to trade even when they make money. You can spend months stacking small losses and one big win and end up substantially ahead, but the experience of the losses is not in proportion to their financial damage.
The asymmetry of drawdown
The single most underweighted piece of math in retail trading is the asymmetry between a loss and the gain required to recover from it. If your account is down 10%, you need 11.1% back on the remaining capital to break even. Down 20%, you need 25%. Down 50%, you need 100%. Down 80%, you need 400%.
The function is y = 1/(1 - x) - 1. It is convex, which means it goes
non-linear fast.
The implication for sizing is direct. If your maximum acceptable drawdown is around 25%, you cannot afford to be losing 5% on a single trade, because five bad trades in a row (well within the binomial range above) take you there in one sitting. A 1% per-trade rule means the same five losses cost you a recoverable 5%, and you still have the account, the strategy, and the time to find out whether the strategy works.
This is the deeper reason for the small-percentage convention. It is not a confidence statement about being right. It is a defence against the part of the curve where the math stops being friendly.
Stop placement determines sizing, never the other way around
The mistake the formula above is built to prevent is sizing first and placing the stop afterwards. If you decide to buy three standard lots of EUR/USD and then look at where the stop should go, you have already made the most important decision blind. The stop ends up wherever your dollar tolerance permits, which may have nothing to do with where the trade thesis would be wrong.
The disciplined order of operations is:
- Trade thesis. Why am I taking this position?
- Invalidation. What price would make this thesis wrong? (That is where the stop goes.)
- Risk per trade. What dollar amount am I willing to lose if the stop is hit? (Usually a fixed percent of equity.)
- Position size. What size makes the stop’s distance equal to my dollar risk?
If step 4 produces a position size you find uncomfortably small, the problem is not the size. The problem is that the trade requires a wider stop than your sizing rule supports. The right reaction is to skip the trade, not to override the rule.
What risk management does not fix
It is worth being honest about the limits of the framework, because the trading-education industry has a long tradition of presenting position sizing as the universal solvent.
Risk management does not create an edge that is not there. A strategy with negative expectancy, well-sized, loses money slowly instead of fast. That is a real improvement (it buys time to find out the strategy does not work without ruining the account), but it does not make the strategy profitable.
Risk management does not protect against ruin from a single catastrophic event: a brokerage failure, a sovereign devaluation, a position held over a weekend that gaps through both your stop and the broker’s negative-balance protections. The published loss-rate data above is overwhelmingly about the slow drip of bad trades and friction, but the rare event matters too, and sizing alone does not address it.
Risk management does not override psychology. The trader who follows a 1% rule for ten weeks and then doubles the size on a “high-conviction” trade because the thesis “feels right” has reverted to discretionary sizing, which is the situation the rule was meant to prevent.
What risk management does is keep you in the game long enough for any real edge you have to express itself, and small enough that any edge you don’t have shows up in a recoverable form rather than a terminal one. That is more than enough work for one piece of arithmetic.
The takeaway
Position sizing is the part of trading the trader fully controls. The fixed-percent rule (typically 1%, with 0.5% to 2% as the practical range) keeps risk constant in proportional terms, deleverages automatically into losses, and survives the streaks the binomial distribution guarantees. R-multiples and expectancy are the right way to measure a strategy; win rate alone is the wrong way and is the trap that signal-sellers exploit. The asymmetry of drawdown makes preventing deep losses far more important than planning to recover from them.
The framework is not a shortcut to profitability. It is a shortcut to not getting forced out of the game before you can find out whether profitability is on the table. Combined with a clear-eyed understanding of trading costs and leverage, it is what distinguishes the small percentage of accounts that survive from the large majority that the published disclosures count.