I wrote a little last Thursday about how most Forex traders won’t consider trading the higher time frames. I pointed out there are a few good reasons to take that approach but noted that the monthly charts (for example) are extremely statistically reliable in producing a good trading “edge”, so here are some further details as promised.
I built a trading system that just looks at the price at the start of a new month and asks two questions:
Is the price higher or lower than it was 3 months ago?
If higher, did the monthly candlestick close in the top decile (10%) of its range? If lower, did the monthly candlestick close in the bottom decile of its range?
If the answer to the second question is yes, then I entered a long trade (if the price was higher than it was 3 months ago) or a short trade (if the price was lower than it was 3 months ago). Exit one month later.
Surprisingly, when this strategy is back tested over the past 10 years on a few important assets (a major U.S. stock index and the two major Forex currency pairs), the strategy is extremely profitable, even without using a stop loss or anything but a time-based exit. In fact, the average expectancy per trade was as much as a positive 40% of the 4-period monthly average true range (ATR) if using a 25% stop loss. This is a very high profit, and you will never be able to duplicate it with a strategy of similar simplicity on a shorter time frame, even on something like the daily time frame.
Wait a minute though, what about overnight fees? Won’t that reduce the profitability severely? Well, let’s take a closer look at that. Almost all retail Forex brokers will charge for most positions held for more than 24 hours, and the net result is firmly on the debit side even if you occasionally get paid a little when the currencies’ respective interest rates are firmly in your favor. Now let’s say that on average you must pay 1 pip per day, and as we are holding the back tested trades for 1 month, that is usually going to be a loss of 30 pips. The back test is measured by ATR (4) on the monthly chart which typically is something like between 250 to 500 pips on a major currency pair. The back test showed an average profit of 40% of that per trade. So even if you must lose approximately 30 pips off that on average, it is only going to reduce the profit from 40% to about 30% per trade. Not a bad result!
More details soon.