This is the second of a four part series. You can read the other parts of the Forex Review 1H2017 here: Part 1: Determinants of a winning t...

FX Trades Review Part 2: New Setups

This is the second of a four part series. You can read the other parts of the Forex Review 1H2017 here:
Part 1: Determinants of a winning trade
Part 3: Stop Losses
Part 4: Tying up the Loose Ends

In Part 1 of the FX trades review, I discovered that the counter-trend pinbar and outside bar setups are responsible for my losing half of year. My goal in this post is to fix them by modifying those setups.

Because of the small sample size of the double inside bar setup or trades in the direction of the trend, I will keep those setup as status quo and only focus on counter trend pinbar and outside bar trades.

These trades have very poor performance and intuitively, they are not the setups that I can ride on the trend for a long time. One thing that I would like to test is if I were to take profit quickly, will I able to have a positive edge on these seemingly unprofitable trades?

Since my definition of trend is based on the 10 day moving average, a logical place to take profit will be at the 10 MA. In the following table, I examine what proportion of counter trend and whipsaw trades will hit the 10 MA.

Breaking down by setups, the proportions are different across setup. For the analysis, I will be looking at "Counter Trend and Price on Trend Side" and "Whipsaw" pinbar and outside bar setups. Excluding setups where the entry price is almost at the 10MA, that leaves us with 25 of these trades - 13 pinbar trades and 12 outside bar trades.

100% of the counter-trend outside bar trades managed to reach the 10MA and only 4 out of 13 pinbar trades reached the 10MA. 

Using the findings above, I came up with alternative strategies for counter-trend pinbar and outside bar:
  • Counter-trend pinbar (bullish bar): short at top of bar, stop loss at 10MA and take profit at bottom of bar.
  • Counter-trend outside bar (bullish bar): buy at top of bar, stop loss at bottom of bar and take profit at 10MA

The counter trend and whipsaw pinbar setup has a less than even chance of profiting, and based on the past trades, I am risking more than a dollar for every dollar I aim to earn. This spurred me to take a more drastic measure to reverse my setups.

Before I proceed to show the results, I have to strongly caveat what I am doing will incur the wrath of the Statistics god. The sample size is far too small and I am obviously over-fitting the model. This exercise is meant for me to get a sense of how changing my strategy may affect my trading performance.

Counter-trend Pinbar

By reversing my setup, no surprise that the 9 trades that produced a loss now produced a profit and the 4 profitable trades turned into a loss. More importantly, the average profit of the 9 winning trades is above $1239 (my risk amount) vis-a-vis an average loss of $728, giving a good risk-reward ratio. Even with a 50% win rate, these trades would have a positive return.

With the flip of the strategy, the 6.8k loss would have turned into 8.2k of profits.

Counter-trend Outside Bar

Prior to the test, I have no idea if this modification will help to make the outside bar trades profitable. Losing trades will become profitable, but the gains are offset by reduced profit from losing trade. The test shows that the effect of the modification is highly positive.

100% of the outside bar trades reached the 10MA within 2 days, before they hit the stop loss of course. There are a few small wins of less than $100 which I may not take in actual trading, but even the largest win is less than 1k (my risk).

But the small wins add up fast with the 100% win rate. With the change in strategy, the $500 gain from these 12 trades turned into a 3.5k profit.


To recap, I had a S$6.2k loss in the first half of 2017. By implementing the 2 modifications above, I would have profited S$14.6k instead. What a difference some small changes can make. I do not know how to emphasise it enough but the above exercise is an unfair comparison which over-fits the model. The results are not robust at all until I do an out-of-sample testing with a bigger sample.

I can only say that this is a good start and my existing strategies definitely have potential although some modifications may be required. In the next post, I will look at protecting the downside of a trade to look for ways to further improve the risk to reward ratio.

In the next part of the review, I will examine if my exit strategies can be improved to give me a bigger edge.

You may also like

No comments:

Powered by Blogger.


All information in The Trader Diaries (TTD) does not constitute to investment advice or recommendation to buy, sell or hold. Readers are advised to consult their financial advisors prior to making any investment or pursuing any investment strategy. TTD will not be liable for any losses resulted from information published or shared from the blog.