Stop Losses

This article was first published on the Motley Fool discussions boards

There has been a good discussion on the Value Shares board as to whether stop losses are appropriate or not (see this thread). My take on this is that stop losses should only be considered as part of an overall trading strategy – and that strategy should be based on your own personal investing objectives. As such, there is no 'right' answer. Stop losses are appropriate for some people and strategies, not for others.

The thread did indicate that there is a certain amount of confusion about what a stop is. The following gives some brief pointers of various types of stops and when they might be useful (*). Hopefully it will raise the level of awareness – and if not then simply writing them down helps increase my own understanding :). Remember that these are not necessarily mutually exclusive – a strategy can employ several different types of stops as long as each one fits in with the objectives behind the strategy.

Time Stops – sell if the share price hasn't risen by a suitable amount after a certain time. The rationale for this is that if a share you currently own is not going anywhere then your money might be better off elsewhere. An example might be buying a share that displays PYAD like value characteristics but has not had any of the value outed after a couple of years.

Psychological Stops. There is no law that states that you must always be in the market. If you know that your life style is likely to be changing significantly (e.g. new job, baby, house) and you will not have the same amount of time available to follow your current strategy then it might be sensible to reduce your positions or switch to something that will require less of your time.

Noise – not a type of stop but important when deciding how to set a stop level. Benjamin Graham's famous quote “In the short-term the stock market is a voting machine. In the long-term it is a weighing machine” applies. Markets are 'noisy'. The price moves up and down based on sentiment and a certain amount of pure randomness. Eventually it will move towards some form of 'fair' price (complicated though since 'fair' is itself a moving target!). The aim of a price based stop is to try and indicate that the current fair value is worse than the one you had expected it to become, whilst not getting triggered by the noise.

Pound Stops – decide what is the most you are prepared to lose on a single investment and sell if losses reach that level. Has the psychological advantage that, barring the price gapping through the stop, you can be comfortable that you will not lose more than you can bear. If the amount you have invested is large compared to the amount you can tolerate losing then you risk being caught up in the noise. For example, say you can bear to lose £1,000 on a purchase. If you buy £1,000 worth of stock then you don't need any form of stop loss. If you bought £10,000 then your limit would be reached after a 10% fall. For £100,000 it would only require a 1% fall. Hence a pound stop also requires some knowledge of the amount of noise likely to be encountered.

Percentage Stops – sell if the share price drops more than a certain percentage amount. Simple to calculate and used by many. Small percentage amounts are likely to fall victim to the noise, whilst large ones will signal too late. Hence you also need to understand the amount of noise likely to be encountered to set a suitable percentage.

***Warning*** - the remaining techniques are based on aspects of TA. If this is anathema to you then probably best to skip to the last paragraph :)

Volatility Stops – these are an attempt to take account of the noise factor. For example, if you calculate the average weekly movement of the share and set the stop a multiple of this beyond the share price then you can be reasonably certain that it is beyond any 'normal' noise. This can work very well for short-term trades (perhaps using the average daily movement instead of the weekly). Not so appropriate for longer term ones though – if the anticipated holding period is more than a year then there is a good chance that a total adverse movement of several times the average weekly amount could occur over a month or two. Using a multiple of the average monthly move might work well in this case but is likely to lead to a rather wide stop.

Dev-Stops – similar to volatility stops, but based on a measure of the standard deviation of average weekly (or daily / monthly) movements. Two standard deviations should encompass about 97% of the price moves. In practice this will be slightly different, as share price moves do not form an evenly balanced bell curve so some adjustment needs to be applied to take this into account. The period over which the standard deviation is calculated needs to be consistent with the intended holding period. As with volatility stops, these are probably not appropriate for longer holding periods.

Support / Trend Line Stops – sell if the share price chart falls below a significant support level. Clearly you must believe that such technical analysis concepts have some validity before considering such an approach. In the hands of an expert TA practitioner these can be very successful. One point worth bearing in mind though is that market professionals are also likely to take note of obvious support levels. Especially with illiquid stocks it is quite easy for them to cause the price to drop just below the support with the aim of buying the shares that will be sold as a result of the support being breached!

Channel Breakout Stops – sell if the share price falls below the lowest value of the previous x days. This is appropriate if you are buying simply because the price has set a new high for the period considered. It is probably not so appropriate for other situations.

Moving Average Stops – sell if the share price falls below the average price of the previous x days. Some people consider that a share is in 'bull territory' if it is, say, above the 200 day moving average and 'bear' if below. On that basis, sell if it drops below the moving average. Again, probably not appropriate unless you are using moving averages as part of your reason for buying the stock in the first place.

One key point to remember about stops is that they will always result in a greater number of losses than if you do not have a stop. The aim though is to avoid the very large losses that can occasionally occur. If you are having too many small losses (and subsequently missing out on large gains) or still having the occasional large loss then your stop loss strategy is probably not appropriate for you.

(*) with acknowledgement to Van Tharp for his book “Trade Your Way to Financial Freedom” from which the list of stop types is taken.