Gamblers bet a portion of their total capital on every wager. It is indispensable for a gambler to know how many successive losses they can sustain before being wiped out. This helps them work out the optimal size of each bet.
This is also a very useful calculation for a trader. In a trade, the potential loss depends on the leverage and volatility of contract. To a certain extent, losses may be controlled by stop-losses. But in a market running against you, the stops may be broken with gaps.
Most experienced traders are uncomfortable with trades risking more than about one per cent potential loss of capital. A highly aggressive style may risk, say, two per cent of capital per trade. It's reasonable to expect slippages in exiting losing positions.
Also, most traders have several trades out at the same time. A stop-loss can be missed. On a bad day, several stop losses may be missed at the same time. So, capital can evaporate very quickly if there is a succession of losses.
This is why position size is important and so are multiple methods of potential loss calculations.
The simplest calculation of maximum potential loss is the widest daily high-low range of a given period. This is the maximum volatility experienced. If a trader suffers one terrible session, or perhaps, two bad sessions, losses may hit this level.
This is a conservative calculation but it's not very practical for setting stop losses because the widest high-low range may be an outlier. Where, and with what logic, do you set stop losses? Some traders set stops purely on support-resistance levels. This doesn't work well in trending markets where an adverse move may smash through the stop loss.
Other traders use average high-low range calculations, or the average true range (ATR) to set stop-losses. An average range is an average of the daily high-low range over a given period. The trader may set that as a stop-loss. Or he may set stops at a multiple of average range (x2), or a sub-multiple (x 0.5), if he's a day-trader.
ATR is a more complex calculation, though it's directly available from many trading packages and easy to program. The “true range” (TR) of volatility in a given session is taken as the largest absolute difference of the following three calculations: A) high minus low of the given session B) high of session minus previous close C) Low of session minus previous close. This covers everything from an extremely volatile session with a wide high-low difference to a jump, or a drop, or a flat session. The ATR is an average of the TR over the given period (the initial session's TR is taken as high minus low). Traders use multiples and sub-multiples of ATR for setting stops.
There are a couple of other scenarios, which experienced traders try to assess for potential loss. One is correlated losses. If you're holding positions highly correlated to each other, the chances of big losses rise. All the positions could go bust together.
For example, if you're trading only banking stocks and there's a change in interest rates, all your positions will move in the same direction. On the other hand, if you're holding non-correlated and inverse-correlated positions, this is less likely. It's worth keeping an eye on correlation though this will have large error factors.
Another situation, which some traders try to assess is “black swan” impact. Some unforeseen event like an earthquake or terrorist strike may cause huge volatility. How much are your favourite stocks likely to swing on such a day? Again, large error factors are involved but you can guesstimate from previous Black Swan days.
Set stop-losses with reasonable assumptions and keep an error margin for the unusual. If your calculations imply a given contract carries unacceptably high risk of loss, it's better to avoid it. It all sounds easy and common-sense. But most traders only learn from experience how difficult it is to put risk-management into practice.
