Exits

Two of the cardinal sins of trading - giving losses too much rope and taking profits prematurely - are both attempts to make current positions more likely to succeed, to the severe detriment of long-term performance - William Eckhardt.

I can relate somewhat to what Eckhardt is saying in the above quote because I tried profit targets for my system sometime ago. They didn't work very well. What was initially supposed to increase win % in fact did not, but instead, reduced the size of average winners substantially. I was effectively cutting winners short.

My whole thinking behind it was because I don't like seeing winners becoming losses, as this (sometimes) happens in long-term trend following systems, that by definition give back alot of open profits. But clearly, the flip side of the coin of cutting winners short made the trade-off rather unfavourable.

Thinking about it now, the idea did not make sense at all, as I'm cutting (potentially) big winners in an attempt stop small losses. Losses are always small because of the my stop loss. Big winners, however, can often be the difference between a great system and an average one. If I were to remove the best 10 trades from the trade database, returns fall by almost 10%p.a.

What could work is a profit exit based on rate of change (ROC), designed for stocks that rise very rapidly. If a stock rises X% in a N number of bars, then the exit is taken. But more work had to be done here before I can make any conclusions.

Howard Bandy discusses exits in a section of his book. He mentions there are 5 types;

1. By the action of an indicator or recognition of a pattern, similar to what caused the entry.

a. The parameters can be the same as those that caused the entry, but in the other direction.

b. The parameters can be different for exit than for entry.

c. Some other indicator can be used.

2. By the price reaching a profit target.

3. By the time in the trade reaching a maximum holding period.

4. By the price falling back to the level of a trailing stop.

5. By the price falling back to the level of a maximum-loss stop.

He goes onto mention that the fifth type is the worst i.e. the maximum stop loss level. Those that have designed and traded mechanical systems would agree with this. During testing, we try and aim for the highest percent possible of trades exited in profit. Exits taken with a loss, and these are mostly those that hit the protective stop, should ideally be no more than 10-15% of the total exits.

Initial stops don't always help the system. For example, for my system, that I've just began trading live on this blog, I tried several but couldn't find an initial stop that improved the bottom line (doesn't mean there isn't one). I wanted an initial stop to calculate position sizing so ultimately I decided to place the initial stop at the trailing exit.

Maximum holding periods I must admit I haven't put enough research into this area to have an informed opinion but it just doesn't make sense to me. I believe the time in a stock should be determined by its trend. You want to spend the most time in the winners because profit increases as time in trade increases. At the same time, you want to spend as little time as you can in the losers, so you can take the loss and use the money for another trade (an opportunity cost issue). The problem is that you never know beforehand whether the trade will be a win or a loss, so it would be difficult (impossible?) to deduce the optimal holding time which would outperform other types of exits.