Waiting seems so unnecessary these days. We live in a digital world with on-demand access to just about everything. There’s an expectation for things to move quickly.
The financial world is no different — our tolerance for waiting is rapidly shrinking.
Take the holding period of shares, for instance.
The average trade length is a fraction of what it once was. Research shows it’s gone from seven years in the 1940s to just seven months in 2007. Some estimate the figure is even lower today.
There are a number of factors behind this shift. But I believe one of the biggest is the chase for performance. Many traders expect almost instant returns.
I see this as a double-edged sword.
Impatience for performance can go one of two ways. It can get you out of losing trades early, which is good. But it can also see potentially good trades cut short.
The Goldilocks principle says the best result is between two extremes. The trick is to strike a balance. Don’t stay in an unprofitable trade too long…but don’t jump out too soon, either.
Well, that’s the theory. How should we manage this in real-time?
I’ll answer this in a minute. First, let me share an email with you.
‘Congratulations on your service Jason — it allows me to rest easy. Setting stops has taught me a great lesson.
‘I’ve been investing in shares for a long time, but your advice on letting winners run and cutting losers makes great sense. I enjoy placing my stops because it feels more disciplined and takes a worry off my mind.
‘I had been wondering about the apparent lack of signal 1s. Can I suggest using some sort of time limit — for example, 3months? If a signal 1 has failed to perform, then Quant Trader sells the stock to make way for an overflow signal.
‘Would there be some way of back testing my idea? It would be very interesting to see if this works.
‘I especially look for to your Friday general educational email. I have become a believer in the idea that it takes 10,000 hours to master a skill. Thankfully I’m getting better at sorting the wheat from the chaff when it comes to advice. Your email helps me in that process.’
I’m grateful for messages like this. Hopefully you find them as interesting as I do. There’s always something to learn from another person’s experience.
Many people find trading stressful. But it doesn’t have to be this way. Strategies like spreading risk and exit stops can help you sleep at night. I’m sure David would agree.
But there’s more to this message than a few kind words.
David also has an interesting idea. He suggests giving stocks a time limit to perform. If they don’t, replace them sooner. This is the sort of logic that makes a good trader.
Time exits are an interesting area. I’ve done quite a bit of work on this over the years.
I did some testing a while back with time-based profit taking. I wanted to see if there was an ideal timeframe to hold a winning trade.
It didn’t take long to see this type of time exit was a flop. Profitability went backwards. Letting winners run was by far the better approach.
I came back to time exits when I was designing Quant Trader.
Part of my development process is to analyse lots of trades. And not just numbers. I also look over the charts for recurring situations. These often lead to system improvements.
I noticed some of Quant Trader’s signals didn’t do much. Upward momentum would stall soon after the entry. The shares would then drift sideways before eventually hitting their stop-loss.
This looked promising. Perhaps exiting these trades early would boost performance. A time-based exit could be the perfect solution. There was only one thing to do — put it to the test.
Here’s what happened.
A number of underperformers left the portfolio early. That’s what I was expecting.
But there were two unintended consequence:
- The success rate fell as the number of early losses rose; and
- There was an increase in the number of trades.
This sort of thing happens all the time with system design. One change has a knock-on effect. You need to balance the benefits with the side effects.
Getting back to David’s email. He asks a specific question:
‘I had been wondering about the apparent lack of signal 1s. Can I suggest using some sort of time limit — for example, 3months? If a signal 1 has failed to perform, then Quant Trader sells the stock to make way for an overflow signal.’
OK, let’s find out.
To do this, I’m going to compare two versions of Quant Trader — one with the standard exit…the other with an early exit for slow starters.
The time limit I’ve set is 60 trading days (or about three months). Any stock not in profit after that time exits the portfolio. This will free a spot for an Overflow Signal.
The start date for the tests is 1 January 2009. To simplify the results, I’m only tracking signal 1s.
Have a look at what happens…
This chart shows the hypothetical profit from the two strategies. It assumes placing $1,000 on each long signal. And it doesn’t take into account costs or dividends.
The blue line represents the standard Quant Trader exit. This is where a stock stays in the portfolio until it hits the exit stop. The red line shows the results of exiting slow movers early.
The difference is small — a profit of $159,183 versus $152,913. But it’s the regular exit in front. The time-based strategy didn’t produce a benefit for this particular period.
So what about the two consequences I mentioned earlier?
Have a look at this table…
|Number of trades||748||1,392|
This is interesting. You can see what’s happening behind the scenes. A more actively managed portfolio results in more trades and fewer winners.
Now, this isn’t necessarily a bad thing. But you need to be aware of it. Some people find lower strike rate systems emotionally more difficult to trade.
Let me show you two examples. This will help you see why the early exit strategy performs the way it does.
Here’s our first chart…
This is a hypothetical trade in Mirvac [ASX:MGR]. It combines Quant Trader’s entry method with a time-based exit strategy.
You’ll notice there are three trades. The first two don’t hit their stop-loss. They exit with a small loss after the first 60 days. It’s only on the third attempt that we see a winning trade.
Now have a look at this chart.
It’s easy to spot the difference. Three trades become one. This method avoids two exits by not limiting time. You can see why this approach generates fewer trades.
I chose this particular example to make a point. There’ll be times when an early exit helps. It can get you out of a stock that never gets going.
But there’s a cost — more trades, and a lower win rate.
So does that mean you should forget about time limits? Not necessarily.
The testing you’ve seen today is for a 100-stock portfolio. It really doesn’t matter if a few stocks are slow off the mark. They only have a small overall impact.
But what if you don’t own 100 stocks?
Well, that’s the topic for next week. I’m going to show the effect time exits can have on a smaller portfolio. I think you’ll find it interesting.
Until next week,
Editor, Quant Trader
Editor’s note: Have any of your stocks hit an all-time high this month? Chances are the answer is no. And that’s understandable…the All Ordinaries is well off its peak. But some stocks are surging. They could make a big difference to your portfolio.
Take Vita Group [ASX:VTG] for instance. You’ve probably never heard of this stock, but it’s trading at record levels. And that’s good for Quant Trader’s members. You see, Quant Trader has signalled this stock three times. It’s now 206% above the initial entry — check it out on a chart.
Anyone can get gains like these. It’s all about having the right strategies. You can learn more about these here.
PS: Quant Trader sources all images in the article above.