Recently I've been working on a strategy (insipired by la poule aux oefs d'or, goosey and a strategy provided by Almo on FF (http://www.forexfactory.com/showthread.php?t=565127)). I found it remarkably effective once my eye caught the pattern and have been trading it successfully for a few days now. The actual method is very straightforward and mechanical. Therefore it is perfectly suited for automation and could reach its full potential that way. Today I traded NFP with it successfully as well which I consider a powerful test for any strategy. Passing it prompted me to share it here.
It goes as follows:
1. Only tested it on GBPJPY due to the pair's volatility.
2. Only indicator needed for this is Wilder's DMI, which would be familiar to most of you I suspect. The timeframe which is used for this indicator is set to M30.
3. The two screenshots below will clarify the simple method. These are how I traded GBPJPY today and yesterday. To clarify: the histogram plots the wedge/distance between the DMI values. Given this, I declare Diff as the difference between 2 subsequent histogram values (the current bar and the previous one). Do note that it doesn't matter whether the histograms are red or green. The only thing that matters is that the wedge between the DMI sufficiently shrinks or expands, which we observe in the relative sizes of the histogram values. Green lines represent the moment of long-entry on the open and orange lines the moment of a short-entry on the open.
Thursday -- (I've accidentally put 2 vertical lines on the first bar where the Diff occur. The entries are on the close of those candles of course) Friday -- (I've put a red cross at a trade where the Diff was too small but only realised it after I entered the trade
)
Entry: 1. Diff > 0.8 at the minimum
2. Wait for the first candle to close in order to confirm the Diff
3. Go in immediately on the open of the 2nd candle. The green lines represent longs and orange lines represent shorts.
Exit:
1. TP (specified below)
2. We'll never stay in a trade for longer than 15 minutes. So if a trade hasn't reached the TP or SL, it shall be closed on the open of the 2nd candle following the entry.
3. SL rather distant (I've used 30 pips)
TP:
1. If 0.8 < Diff < 1.5 then TP = 10 pips
2. If 1.5 < Diff < 2.5 then TP = 20 pips
3. If 2.5 < Diff then TP = 30 pips
Naturally these values can be experimented with in order to find an optimal setup. Perhaps it may even be possible at a later stage to hold on to trades longer utilising jumping stops or so, given the fulfilment of certain conditions that is. A positive and most appealing point of this strategy is that any one trade won't be held for longer than 15 minutes (3 bars on M5), thus keeping market exposure at its minimum. Another is its simplicity. No bombardment and clutter of a dozen indicators to wait for perfect confluence in order to take a decision. A slight downside involves slightly raising the lot size to, say, 0.01 per 500 Dollop. Occasionally I suspect a SL will get hit, though I've yet to see one actually getting hit since we keep our market exposure tight. BE is more likely than a hit SL no doubt. The odds seem to be in our favour in any case. An unemotional and disciplined EA would do a much better job than a mere human here and could literally exploit all the opportunities, day and night. In combination with the power of compounding, this could prove to be of great value - that is, over time - to anybody wishing to use it. In attachment you can find the required indicator and the template I'm using. In any case, I'd be greatly honoured and grateful should anyone be willing to have a go at it (in the midst of fantastic advances here on SHF
). Have a great weekend!PS: since I've done this manually and by mere sight, I might have slightly broken the Diff rules by a few decimals.
Edit by Steve on 4th June 2016
The EA is in post 3.
Update 5/06/2016
Since the inherent nature of the use of the M30 Wildler's DMI may pose difficulties as Jal points out here, a more robust and better alternative is to be found in this post. Here you can set Wilder's DMI on M5 and use the values given in aforementioned post as reference points. By using M5 there as well we completely eliminate the possibilities of dynamic histograms since here the histogram values will become static once the candle closes. To clarify, I did have 3 days of positive results trading the M30 manually and merely on sight with tiny lots. Of course I did not take all the trades as I couldn't be behind my computer the entire time. In any case 3 days are absolutely insufficient to utter this method profitable over the long run. Therefore would like to thank Steve once more for giving it a shot anyway and spending time on it. The above and first few posts of this thread can be used as visual material and examples as to how this method works. I'll keep you posted
! Do feel free to share your settings, particular patterns you may notice, etc. Much appreciated!
