Spring — it’s one of my favourite times of the year.
Memories of winter’s gloom are quickly fading. The weather is warming and the days are getting longer…just how I like it.
The climate and extra daylight aren’t the only things I enjoy. It’s also time for the Spring Racing Carnival — an exciting mix of friends, fashion, and fun.
I was trackside for all the action last Saturday. One of my highlights was the feature event. Watching a time-honoured race from a 100-year old grandstand is special. I feel part of history every year.
Another aspect I enjoy is to watch how people bet. Some punters do their own analysis of the form. Others put their faith in a ‘hot tip’ from a mate, while many wager on colours, or a name.
And these strategies can get results…
Champagne corks were popping after each race. Backslapping and smiles were in abundance. Winning tickets were everywhere.
But here’s the thing. Rarely did I see the same people celebrating. Sure, there were lots of winners…although it was usually different people after each race.
I doubt many of Saturday’s betting strategies would stack up week after week.
Luck versus skill
The whole betting culture got me thinking about trading. You see, success needs to be repeatable. Anything less is just a game of chance.
I gave up betting on horses years ago. It turns out I didn’t have an edge. And without an edge, you’re just giving money to those that have one.
It took me a long time to figure this out. A few good wins would trick me into thinking I was on to something. But then I’d lose money for weeks on end.
During my university days, I decided to keep a ledger. For one year, I kept a record of every bet and the outcome. The year-end result was revealing.
A small loss was all I had to show for my efforts. The jury was in…I wasn’t going to be a professional punter. Race day has been a social event ever since.
So what does a day at the track have to do with quant trading?
Well, it made me reflect on my experiences.
It took me a year to confirm I wasn’t a skilful punter. Trial and error is a tedious way to find out. You eventually get an answer…but you can waste a lot of time in the process.
My experience as a trader has been entirely different.
Gone are the days of a time-hungry ledger. I now test my ideas with a computer and lots of share price data. This is a much more efficient way to see if a strategy has merit.
Which signal type is best?
I regularly get questions about Quant Trader’s portfolio. One of the most popular relates to the various entry signals. People want to know which category is best: 1, 2, or 3.
A member recently sent in an email. She had a concern that signals 2s and 3s were not as good as signal 1s.
Here’s what she said:
‘I have noticed a marked difference between the 1st, 2nd and 3rd signals. You do not explain what you believe the long term result of these trades to be, which is my suggestion to you.
‘For me to buy on the 2nd signal, I would obviously want to be confident it can go up from my buy in price. Do you believe a stock can keep rising after signals 2 and 3?’
I can understand where Deborah’s coming from. We often associate a second or third offering as being a lower tier.
Deborah also sent me an analysis of recent signals. It shows that signal 1s are up the most on average. Signals 2 and 3 are a notch behind.
Drilling down on results can really boost your understanding. But it takes a lot of time and data to do this properly — just like my old betting ledger. You also need to consider the full picture. Only looking at one outcome (like percentage gains) can bias your conclusions.
I’m going to show you a few performance figures in a moment. This will give you some specific data on the various signals.
The data you’re about to see isn’t for Quant Trader’s live signals. Many of the most profitable signals are still open. It’s simply too early to fully analyse these. Instead, I’ll use the data from Quant Trader’s extensive back-testing. This will give you two decades of results. I look at this type of data when evaluating a system.
Unlike my year of keeping a ledger, this should bring you up to speed a lot quicker.
OK, here’s the first table. This shows the average trade length for the signals. It covers the period between 1 January 1993 and 31 October 2014.
|Signal||Average Holding Period Winning Trades||Average Holding Period Losing Trades|
As you’d expect, signal 1s have the longest average holding period. These figures will likely vary over time. But they give us a general guide of what’s possible.
The next table shows the average returns from these trades.
|Signal||Average Profit||Average Loss|
Signal 1s appear to be the most profitable. But good analysis goes beyond a headline number. It often pays to go a few steps further.
Let’s annualise the average returns.
This shuffles the deck. Signal 3s now have the top rank. The reason for this is the shorter average holding period.
But we haven’t considered strike rates. This is the historical success rate for the signals.
Here’s how they stack up.
|Signal||Success Rate||Number of Trades|
Notice the high number of trades. This improves the data’s robustness. It means the results are not due to a handful of lucky trades. Repeatable success is a key part of the strategy.
Finally, there is the risk-adjusted return. Let me convert all this into a dollars won/dollars lost ratio. We can then make a comparison.
This is what I do. Multiply the average profit for signal 1s (39.7%) by the percentage of winners (58%). This gives us 0.2305.
I then do the same for losses (19.3% multiplied by 42%). This comes to 0.0808.
The last step is to divide 0.2305 by 0.0808. The result is 2.85.
In other words, for every $2.85 dollars won on signal 1 trades…$1 is lost.
The respective ratios for signals 2 and 3 are $3.40-to-$1 and $3.12-to-$1.
Dollars Lost Ratio
We can now look at the performance from every angle.
Here’s a summary:
- Signal 1s are the choice if your preference is win rate and outright percentage gain;
- Signal 2s have the best dollars won/lost ratio;
- Signal 3s annualised return is the highest.
These figures will vary over time. This is only a general guide to help you understand the possibilities. But it goes to show there’s more to a trade than just the return.
Personally, I don’t worry about signal types. Each of them have historically made a lot more money than they’ve lost. The fact is, a trade can have many profitable entry points.
Knowing your trading strategy’s past performance is vital. Not only can this save a lot of time spent on trial and error, it can also give you the understanding to trade it effectively.
Back-testing is a key part of the Quant Trader process. It can give you an edge over the ‘punters’ who simply jump from trade to trade. I hope you can use this method to your advantage.
Until next week,
For Markets and Money
Editor’s note: Jason’s trading system — Quant Trader — has just identified an opportunity. The company has a market cap of $44 million. So you probably won’t read about it in your broker’s reports. The company is also outside the All Ordinaries. This puts it ‘below the radar’ for many traders.
This developing situation is getting interesting. The stock has been RISING during the recent selloff. This could be just the start of a much bigger move.