Principles of Risk Management

The principal cause of account failure

A trading method with a modest positive expectancy, applied with consistent and limited risk, produces a gradual accumulation of capital interrupted by periodic drawdowns. The same method applied with excessive risk can exhaust an account during an ordinary sequence of losses, before its positive expectancy has had the opportunity to assert itself over a sufficient number of trades. The quality of analysis is therefore secondary to the control of risk: sound analysis cannot compensate for position sizes that expose the account to ruin.

Leverage magnifies this consideration. Onys Kapital offers leverage of up to 1:1000 under its dynamic leverage schedule, which means that a comparatively small margin deposit can control a large notional position. The availability of high leverage is a facility for capital efficiency; it is not a guide to appropriate exposure. Trading CFDs on leverage carries a high level of risk, and losses can exceed the initial deposit on a position. The rules in this lesson exist to ensure that the exposure taken bears a sensible relationship to account equity.

Fixed fractional risk and the arithmetic of losing sequences

The most widely adopted risk rule limits the loss on any single trade to a fixed small percentage of account equity, commonly one per cent. The justification becomes apparent when the effect of a consecutive sequence of losses is calculated. Ten consecutive losses at one per cent each reduce an account to approximately 90.4 per cent of its starting equity, since each loss is taken on the reduced balance. The account remains intact and the trader retains the capacity to continue applying the method.

The same ten consecutive losses at ten per cent per trade reduce the account to approximately 34.9 per cent of its starting equity, a loss of roughly two thirds. A sequence of ten losses is not an exceptional event: a method with a 50 per cent win rate will produce ten consecutive losses at some point in a sample of a few thousand trades, and sequences of five or six are routine. The risk percentage must be chosen so that a normal losing sequence is survivable in both financial and psychological terms.

• Ten consecutive losses at 1 per cent: approximately 9.6 per cent of equity lost.
• Ten consecutive losses at 2 per cent: approximately 18.3 per cent of equity lost.
• Ten consecutive losses at 5 per cent: approximately 40.1 per cent of equity lost.
• Ten consecutive losses at 10 per cent: approximately 65.1 per cent of equity lost.

The asymmetry of drawdown recovery

The percentage gain required to recover a loss is always greater than the percentage of the loss itself, because the gain must be earned on a smaller base. A loss of 10 per cent requires a subsequent gain of 11.1 per cent to restore the original equity. A loss of 25 per cent requires a gain of 33.3 per cent. A loss of 50 per cent requires a gain of 100 per cent, and a loss of 75 per cent requires a gain of 300 per cent. The general formula is: required gain = loss ÷ (1 − loss), with both expressed as decimals.

The relationship is non-linear and becomes increasingly severe as the drawdown deepens. A trader whose account has fallen by half must double it merely to return to the starting point, a performance that would be regarded as exceptional under any circumstances and that is particularly unlikely under the pressure of recovering a loss. This asymmetry, rather than caution as a general disposition, is the mathematical reason for keeping risk per trade small: shallow drawdowns are recoverable through ordinary performance, whereas deep drawdowns are not.

• Loss of 10 per cent: gain of 11.1 per cent required to recover.
• Loss of 25 per cent: gain of 33.3 per cent required.
• Loss of 50 per cent: gain of 100 per cent required.
• Loss of 75 per cent: gain of 300 per cent required.

The core rules of a risk management framework

A functional risk framework consists of a small number of rules, each addressing a specific way in which accounts are lost. First, every position carries a stop-loss order placed at the price at which the trading idea is demonstrably incorrect, rather than at the point at which the loss becomes uncomfortable. Second, the monetary risk on each trade is a fixed percentage of current account equity, from which the position size is derived. Third, a daily loss limit is defined in advance, and reaching it ends trading for the session without exception.

Fourth, total exposure across correlated positions is capped, so that several positions dependent on the same underlying factor are treated as a single risk. None of these rules requires analytical skill. They require the trader to write them down in a period of calm and to observe them in periods that are not calm. A rule that is amended during a session, in response to the discomfort of an open position, has ceased to function as a rule.

Escalation of risk following losses

The most destructive pattern in retail trading is an increase in position size following a loss, undertaken with the intention of recovering that loss quickly. This behaviour is commonly described as revenge trading. It is experienced by the trader as a means of regaining control, but its effect is the opposite: it compounds a normal loss with an abnormally large exposure at the moment when the trader's judgement is most impaired.

The rule that prevents escalation is structural rather than psychological. Position size is determined by the account equity and the stop-loss distance, using the calculation set out in the next lesson. It is never determined by the result of the previous trade, by the size of a recent loss or by the trader's conviction about the current opportunity. When this rule is observed mechanically, the sequence of losses that precedes most account failures cannot occur.