Whether you have a reliable and a good trading strategy, you will see after some time. In trading we call this walk forward test. It is the final and most reliable test.
In this blog I share with you a walk forward result of a trading strategy, that I released in April 2016. This is 9 months of unseen data for the strategy and gives a good indication about the reliability of the strategy. I annouced the strategy in the blog “E Mini Futures trading systems“.
It is a day trading strategy with a slightly different approach.
Normally you trade in the daytime together with the crowd and maximum competition. But what happens if you avoid trading with the maximum competition and instead look for your edge in a quiet period but nevertheless with a good potential? So I thought that it might be a good idea to trade at the opposite time, the night time.
During the night you can’t trade equites, but you can trade the big indices, like the Russell 2000, the S&P 500, the Nasdaq 100 or the MidCap 400. All these indices tend to open with a down gap or an up gap the next day at the oping bell. The idea was to profit from this scenario.
I started to develop a fully automated trading strategy, that trades only the nighttime and it should deliver good result for the above mentioned indices. If it delivers good results for all 4 indices it already passed the robustness test, i.e. the strategy is valid for several markets. This increases the likelihood, that the strategy will survive with unseen data, which is the walk forward test.
After some time and several tries and errors I finally got a good automated trading strategy with the following characteristics:
- trading overnight
- valid for different markets
- good backtest results
- valid over 10 years
- works in different market conditions like a bull market , bear market or congestion
The strategy delivered over the last 10 years on average over $5,000 US per year, trading one single contract for each index. This is over $450US each month, and only one year (2010) the strategy was not profitable. Futhermore, it trades only very selective, i.e. on average one or two trades each month in one market. Overall it is a very quiet strategy, but with good performance statistics.
Big question: What happens the rest of 2016 with unseen data?
Lets take a look at the overall behavior of the strategy in all four markets:
As you can see, the strategy drops in a drawdown at first, even though the MidCap had a very good month with a gain of more than $1,300US. But the other markets override the good result and in the end, we suffered a drawdown of $1,550US. This is painful, but in the normal range of the strategy. In summer the strategy was flat and in September it started a recovery and we ended, we a very small profit of $60US. The next month, October was disappointing again, but in November we made over $1,500US and ended with a cumulative gain of $750US. December was nearly flat and the first week in January 2017 delivered more than $1,200US.
For the whole walk forward period of approximately 9 months we are $2,010US up.
This is an impressive result and shows the robustness of the strategy and the good behaviour with unseen data.
If you would like to trade the strategy, you can find it at the TradestationApp Store.
If you would like to see more of my strategies, you can find them here.