Ask Colin

If my share price goes through my stop during the week should I exit or should I base my exit on the close of the week?

The Full Question was:

I am a trend trader using weekly charts. If my share price goes through my stop during the week should I exit or should I base my exit on the close of the week? I often find I get stopped out during the week only to find that by the end of the week it closes above my stop. Or do I do whatever my trading plan says?

The quick answer to your question is that you should always follow your trading plan. You should do this through to the end of your trade. If you start trying to re-think your plan during a trade, you will be taking a step on a very slippery slope. It is very easy for you to fall into several psychological traps, such as availability bias, where you make decisions based only on most recent observations, which may represent an atypical sample. What you are obviously asking is whether you ought to change your trading plan. If you think that you want to review your plan, this is best done while you are out of the market. That is why I suggested the quick answer was to follow your plan.


You seem to have observed some situations where, if your trading plan rules had been different, the result may have been better. These ideas should occur to thinking traders all the time. They should then test them. That means going back through past data for the sort of stocks you trade in rising, falling and flat markets. You need to find hundreds of situations that you may have traded on your selection and entry rules and then compare the results from your current stop placement rule versus the new idea. That is the only way you will be able to know whether the new idea would have worked in the past. This gives you some basis then for whether you might want to change that aspect of your trading plan.


This can be painstaking work, but the best traders got to be where they are by doing the hard work, rather than guessing. The alternative to testing on lots of past data is to test it going forward with real money. Most people do it that way and they get an answer one way or the other eventually. However, the market can charge very high tuition fees which it hides from you by labelling them trading losses. It is much smarter to do the testing on paper on past data. You can get part of the way with software that lets you test trading rules. However, it is only a partial solution. Testing in this way gives you some quantifiable data, but it does not deal with the psychological aspect. Ideally, you need to set up the charts and step through them by adding one bar at a time and try to experience the emotional pressures. For example, your testing may show that your new method works well over time. It is easy to gloss over some of the software testing results like the maximum drawdown (paper loss) during the trade and the number of consecutive losing trades. In reality you may not be able to stand these levels of heat or pain in the trade, so no matter how good the system over lots of stocks and lots of time, you would have stopped using the method before it paid off, or maybe skipped some trades that turned out to have been the big ones.


As for my experience of closing prices versus intra-day or intra-week lows, I do not have what you are looking for. I don’t know how you place your stops and I don’t know which kinds of stock you trade in. My own method is based on intra-day/week lows (they are the same thing) because the logic of my method is to trade the trend, which is defined by troughs. However, in thinly traded stocks I will carefully examine very large downward intra-day spikes that occur on low volume because of cascading automatic stop-loss orders. Notice that I said that I will carefully examine them, not that I ignore them. I treat every case on its merits and will be closely monitoring daily and even intra-day data for whether there is follow through on the fall or that it was a system anomaly caused by beginners using automatic stops in an unsuitable stock to apply such a tool.


If you told me what sort of stock you trade and your present and proposed stop rules, I could test them for you. However, I have neither the time nor the motive to do so. Even more importantly, it would be a bad idea because you would not have lived your way through the hundreds of situations that ultimately build up faith in your method in your mind.


A couple of years ago I discussed with a very good trader how he came up with a new signal that he used. His answer was what I have told you above. When I asked him how long it took, he said about 80 hours of intensive work. This is actually a short time, because it was a short term signal that set up in and was resolved in about ten periods. If you are trading a longer time frame, more time will be needed in testing. That is the shape of the territory if you want to be a good trader. These are the people you are competing with to take profits out of the market. If you don’t have the answers on these issues and they do then money will flow from your bank account to theirs. I am sorry to be a bit brutal about this, but it is better I tell you than the market teach you.