Many who trade Forex claim that a bottom of a short trend is associated with a candle where the distance between the high of the day to the low of the day is very long, as well as we can observe a long wick/shadow down:
Of course, this is an example of a hammer or alternatively doji if the body of the candle is very “thin” (Japanese candlesticks). The same can be said for uptrends, i.e. the wick is up and the long candle occurs at the top of a trend. Then this may signal a reversal so that traders may begin to short the pair.
In my backtesting, I limited myself just to buying a Forex pair, not shorting (because the conclusions would be the same).
At first, we say that today the low of the last candle should be the lowest of the past 5 days (in order to indicate that the trend is down). Also, the difference between the high and the low of this day should be the highest of the past 3 days. Then we say that the lower wick today should be at least twice longer than the body of the candle and the upper wick must be rather short, less than half of the body. We buy on open on the next day.
Here we can see an example:
We set a stop loss 15 pips below the today’s low and calculate the take profit level in a bit artificial way: we add the total length of today’s candle (i.e. the high minus the low) to today’s high. This should make the risk/reward ratio to be at least 1.0 and also will indicate the market direction. As such, the strategy is extremely simple and perhaps I would not use it in practice but still it would teach us if the observations done by Forex traders are really true.
We test all the main Forex pairs in the period 1 January 2000 – 31 May 2018.
Here are some selected results (using NinjaTrader software using data from Kinetick):
|Number of trades||215|
|Average time in market||11 days|
So, the results are positive at least (the profit factor is greater than 1.0). This shows that the strategy should really work. The pairs that led to positive income were: USDJPY, USDCHF, NZDCAD, NZDCHF, EURAUD, GBPCHF, EURCAD, GBPJPY, CADJPY, EURCHF, EURNZD, AUDCAD, AUDUSD, AUDJPY, EURUSD.
We can try to improve the results by making the lower wick slightly longer. Let us demand that it should be at least 2.5 times longer than the body. Here are the results:
|Number of trades||138|
|Average time in market||12 days|
Most parameters have improved. On the other hand, the number of trades is lower so that this set-up occurs not very often.
And now, let us change the scale from daily to weekly! Here are the results, if we return to the first strategy where the lower wick was at least twice longer than the body:
|Number of trades||54|
|Average time in market||42 days|
The average time in market is certainly longer, as it should. The profit factor and the profit from a single trade have also slightly improved. This may confirm that weekly signals are more powerful (as we would expect considering a higher volume traded on a whole week in comparison with just one day). On the other hand, the improvement is not so significant.
How to use this strategy in practice: you can consider entering if this type of candle occurs. In addition, you can take into account other factors like touching a trendline etc.